GEOmedia 3 2022

Rivista bimestrale - anno XXVI - Numero 3/2022 - Sped. in abb. postale 70% - Filiale di Roma
LAND CARTOGRAPHY
GIS
CADASTRE
GEOGRAPHIC INFORMATION
PHOTOGRAMMETRY
3D
SURVEY TOPOGRAPHY
CAD
BIM
EARTH OBSERVATION SPACE
WEBGIS
UAV
URBAN PLANNING
CONSTRUCTION
LBS
SMART CITY
GNSS
ENVIRONMENT
NETWORKS
LiDAR
CULTURAL HERITAGE
May/June 2022 year XXVI N°3
Mobile Robotics and
Autonomous Mapping
DATA COLLECTION
AND PUBLICATION
WITH QGIS
PLAYING WITH COLORS
ON PANCHROMATIC
AERIAL PHOTOGRAPHS
MODELLING WATERSHED
PHENOMENA WITH QGIS

Inspiration for a smarter World
This is the motto of the Intergeo Conference 2022, where the English language edition of
GEOmedia, usually issued as the third of the year in the summer, will be distributed on next
October.
The Conference will highlight current developments in surveying with following main topics:
Digital Twins and their value creation
4D-geodata and geospatial IoT
Potentials of remote sensing
Industrial surveying, measurement systems and robotics
Smart Cities and mobility in the context of climate change and sustainability
Mobile mapping, Web services and geoIT in disaster management
Spatial reference and positioning
Earth observation and Galileo
Trend topics such as Building Information Modeling (BIM) and the diverse application
possibilities of the Digital Twins, but also the current requirements for the Smart City and
rural areas have their fixed place in the Conference. The Digital Twins will be a matter of
particular importance in this edition. The focus will be on their use in Building Information
Modeling, smart planning and construction.
But let’s go to see the content of this issue where we start with a Focus “From the field to
the clouds: data collection and publication with QGIS” by Paolo Cavallini, Matteo Ghetta
and Ulisse Cavallini, about the main solutions available for data collection and seamless
publications over the web: MerginMaps, Qfield, Lizmap, with an example form a water
resources project in Gambia. A following Focus is on “Open-source GIS software and
components for modelling watershed phenomena”, by Flavio Lupia and Giuseppe Pulighe,
over the recent version of the Soil and Water Assessment Tool (SWAT) that was implemented
by a dedicated QGIS plugin (QSWAT), widening the userbase and the potential modelling
application worldwide. Then we’ll go to the Reports “Mobile Robotics and Autonomous
Mapping: Technology for a more Sustainable Agriculture” by Eleonora Maset, Lorenzo Scalera
and Diego Tiozzo Fasiolo, concerning the automation in geomatics for agriculture using
robotics platforms, that must be equipped with appropriate technology. And “Geographical
Information: the Italian Scientific Associations and... the Big Tech” from Valerio Zunino,
observing that while the World is changing, the Italian Scientific Associations of Geographical
Information are not.
Marco Lisi, in “Time and Longitude: an unexpected affinity”, talks about the Time, the fourth
dimension, becoming increasingly important in all aspects of technology and science.
Finally, don’t miss “Potatoes, Artificial Intelligence and other amenities: playing with Colors
on Panchromatic Aerial Photographs”, by Gianluca Cantoro from Italian National AirPhoto
Archive (Aerofototeca Nazionale, AFN), discussing about the use of historical photographs,
whether taken from the air or from the ground, are usually synonyms of grayscale or sepia
prints.
Enjoy your reading,
Renzo Carlucci

In this
issue...
FOCUS
REPORT
COLUMNS
From the field
to the clouds:
data collection and
publication with QGIS
By Paolo Cavallini,
Matteo Ghetta,
Ulisse Cavallini
6
24 ESA Image
32 NEWS
40 AEROFOTECA
46 AGENDA
12
Open-source GIS
software and components
for modelling watershed
phenomena
Understanding the soil
and water components
under different
management options with
QGIS and the SWAT
By Flavio Lupia
and Giuseppe Pulighe
On the cover the
sensorized mobile
platform developed
at University of
Udine, Italy.
geomediaonline.it
GEOmedia, published bi-monthly, is the Italian magazine for
geomatics. Since more than 20 years publishing to open a
worldwide window to the Italian market and vice versa.
Themes are on latest news, developments and applications in
the complex field of earth surface sciences.
GEOmedia faces with all activities relating to the acquisition,
processing, querying, analysis, presentation, dissemination,
management and use of geo-data and geo-information. The
magazine covers subjects such as surveying, environment,
mapping, GNSS systems, GIS, Earth Observation, Geospatial
Data, BIM, UAV and 3D technologies.

Mobile Robotics and
Autonomous Mapping:
Technology for a more
Sustainable Agriculture
by Eleonora Maset,
Lorenzo Scalera,
Diego Tiozzo Fasiolo
16
ADV
Ampere 45
Epsilon 33
Esri Italia 23
Geomax 31
Gter 15
INTERGEO 35
Nais 39
Planetek 48
Stonex 47
Strumenti Topografici 2
Geographical
Information: Our
Associations and ...
the Big Tech
by Valerio Zunino
20
Teorema 46
In the background image:
Bonn, Germany. This Esa
Image of the week, also featured
on the Earth from
Space video programme, was
captured by the Copernicus
Sentinel-2 mission, that with
its high-resolution optical camera,
can image up to 10 m
ground resolution.
(Credits: ESA)
26
Time and
Longitude: an
unexpected
affinity
by Marco Lisi
Chief Editor
RENZO CARLUCCI, direttore@rivistageomedia.it
Editorial Board
Vyron Antoniou, Fabrizio Bernardini, Caterina Balletti,
Roberto Capua, Mattia Crespi, Fabio Crosilla,
Donatella Dominici, Michele Fasolo, Marco Lisi,
Flavio Lupia, Luigi Mundula, Beniamino Murgante,
Aldo Riggio, Monica Sebillo, Attilio Selvini, Donato Tufillaro
Managing Director
FULVIO BERNARDINI, fbernardini@rivistageomedia.it
Editorial Staff
VALERIO CARLUCCI, GIANLUCA PITITTO,
redazione@rivistageomedia.it
Marketing Assistant
TATIANA IASILLO, diffusione@rivistageomedia.it
Design
DANIELE CARLUCCI, dcarlucci@rivistageomedia.it
MediaGEO soc. coop.
Via Palestro, 95 00185 Roma
Tel. 06.64871209 - Fax. 06.62209510
info@rivistageomedia.it
ISSN 1128-8132
Reg. Trib. di Roma N° 243/2003 del 14.05.03
Stampa: System Graphics Srl
Via di Torre Santa Anastasia 61 00134 Roma
Paid subscriptions
Science & Technology Communication
GEOmedia is available bi-monthly on a subscription basis.
The annual subscription rate is € 45. It is possible to subscribe at any time via
https://geo4all.it/abbonamento. The cost of one issue is € 9 €, for the previous
issue the cost is € 12 €. Prices and conditions may be subject to change.
Issue closed on: 28/07/2022
una pubblicazione
Science & Technology Communication

FOCUS
From the field to the clouds:
data collection and
publication with QGIS
By Paolo Cavallini, Matteo Ghetta, Ulisse Cavallini
Open source GIS, and in particular QGIS, is a leading free and open
source solution for desktop mapping since many years already.
Its versatility, ease of use, and analytical power have made it the
software of choice for many professionals around the world (see
https://analytics.qgis.org). Field data collection and checking, and
web publication are attracting more attention in the recent years.
A whole suite of integrated tools is now available to implement a
complete workflow, all centered on QGIS.
Central to all tools is the QGIS project, designed and created using
QGIS Desktop. Its power in creating beautiful and rich styling is
probably unsurpassed, with expression-based styling, fusion modes,
and a huge set of other functions. The same project can be used on a
mobile device, and exposed through a web service (WMS, WFS, WCS,
WPS) and a complete web interface.
Mobile
Over the 20+ of life of QGIS,
a number of mobile interfaces
have been designed, from
the first attempt to run the
whole of QGIS on Android,
or a simplified interface on
Windows Mobile, to proper
mobile apps. Currently the
most important and used
solutions to use QGIS on a
mobile are Qfield (https://
qfield.org/) and MerginMaps
(https://merginmaps.com/).
Both are generic tools, that
can be effectively employed
in a wide variety of contexts.
Their flexibility stems from
the ease with which they
can be configured, simply
through QGIS projects, that
include sophisticated styling
and simple to complex forms,
with all expected functionalities
such as custom menus
(drop-down, checkbox,
calendar etc.), relations,
constraints, default values,
user guiding tips, etc. While
still relatively new entries in
the market, they have been
successfully employed in
extensive surveys, from single
users up to thousands of field
surveyors simultaneously collecting
data in the field.
MerginMaps
MerginMaps is a web service,
written in Python with flask,
that manages the synchronization
process of a QGIS
6 GEOmedia n°3-2022

FOCUS
project, and all the related
files. For media files, it is
not unlike other cloud file
management service. Unlike
other cloud file managers,
though, MerginMaps is able
to manage geospatial information,
primarily in the form
of geopackage files. When a
new version of a geopackage
is uploaded, for example because
a surveyor added some
features and is uploading
them back on the centralized
server, the geodiff library is
used to check for changes,
merge them and solve any
conflicts. This enables some
flexibility in the downloading
and uploading of new data,
since multiple surveyors can
add features for their area
of interest, upload different
versions of the modified geopackage,
and MerginMaps
will take care of adding every
new feature to the centralized
repository.
Lutra Consulting, the firm
developing MerginMaps,
offers an official hosted instance,
a reliable way to use
MerginMaps without the
need for configuration and
installation on a local server.
The official hosted instance
offers a generous free trial
for non-commercial usage.
Pricing is clear and reasonable,
with no per-user pricing,
and the support is quick and
responsive.
The surveying process is effectively
split in two. The
first phase involves the generation
of the QGIS project,
the related layers, and the
form structure, and the
subsequent upload to the
MerginMaps web service. Far
from being complicated, this
phase still requires a good understanding
of GIS software
and data formats.
Once the project is uploaded
and ready, the surveying phase
can begin. Due to the easeof-use
of the mobile application,
the surveying requires
minimal technical skill, and
operators can be trained in a
matter of few few days. From
their point of view, the intricacies
of the project are invisible:
they just need to add or
modify the features according
to the form, preconfigured
through QGIS, and click on
the synchronization button
once they are online. A very
recent addition, the option
to automatically upload new
changes whenever an internet
connection is detected, further
simplifies this.
The project folder itself is
what is visible from the web
interface. In order to manipulate
the project, and the
geospatial data, MerginMaps
can be accessed from QGIS,
through the official plugin,
and through the MerginMaps
mobile application, available
for Android and iOS.
The MerginMaps application
has a special focus on simple
UI and UX design, in order
to be accessible by everyone,
regardless of their GIS experience
and in demanding field
conditions.
Qfield
Qfield has been the first natively
Android mobile application
connected to QGIS.
Downloaded around halfmillion
times it is available
for Android and now iOS.
The idea of the usage is very
simple: the user sets up a
project in QGIS and thanks
to the plugin QfieldSync it
will be packaged in a folder.
The folder created has to be
copied to the device and with
the App data can be collected
on the field. Back to the office
the data collected with
the mobile device have to be
copied back on the machine
and re-synchronized to the
original data source with the
QfieldSync plugin.
GEOmedia n°3-2022 7

FOCUS
The App has a very simple
design and it comes with a lot
of features: snapping facilities,
advanced form layout, pictures
and interaction with the legend
to name a few.
The manual synchronization
can be nowadays avoided
thanks to QfieldCloud, a
Django framework that is able
to store and automatically
synchronize the data from the
computer to the mobile device
and vice versa. Open Source,
QfieldCloud is still in Beta
version and let the user choose
between installing the software
on the server or register to the
web with a free plan (limited
space) or buy additional space.
The main advantage of
QfieldCloud is that the user
can log in both on the machine
and on the device with the
same name and immediately
synchronize the data between
all the devices. The QfieldSync
plugin in QGIS has all the
options needed to log in, synchronize
and also see the data
changes difference.
8 GEOmedia n°3-2022

FOCUS
As for Mergin, also Qfield
has the possibility of using a
server managed by OpenGIS.
ch, the firm developing it
without the need for configuration
and installation on a
local server.
Web
A number of different web
interfaces have been designed
around QGIS. For all of
them the basic mechanism is
the same: all user requests are
sent to the backend (QGIS
Server) that creates the map
and other objects (legend,
print layouts etc.) and send
them back to the web app.
The main advantages over
other free and open source
webGIS solutions are the ease
to create both visually sophisticated
maps, and complex
print and reporting through
QGIS Desktop layouts, without
the need for specific
web skills.
The most widely used is
Lizmap, created and maintained
by 3Liz, a South French
company, who also substantially
contributes from years
to core QGIS development.
As for the other solutions
described, also Lizmap can
be used without the need for
configuration and installation
on a local server through a
service managed and maintained
by 3Liz, the firm developing
it.
Case study
MerginMaps was recently
used in a project geared towards
the improvement of
water resources infrastructure
in The Gambia, financed by
the African Development
Bank; the project is headed
by Hydronova, and its GIS
section is technically managed
by Faunalia.
Special acknowledgment
for this project goes for the
continuous support to the
Climate Smart Rural WASH
Development Project Office
Team and to the Department
of Water Resources Staff, under
the Ministry of Fisheries
and Water Resources of the
Gambia.
Data collection is the first
task upon which the whole
project is built, since in order
to improve resource management,
key stakeholders need
to know the current situation
and distribution of the resources
at their disposal.
An open source solution
is ideal in most contexts,
even more so in a context
where free access to data is
paramount, and budget constraints
are tight.
The MerginMaps mobile
application, backed by the
Mergin web service, was chosen
due to its ease of use and
synchronization. Due to the
possibility and ease of setting
up a Mergin instance, all the
data was kept in-situ at the
relevant ministry, retaining
control on this crucial information.
A QGIS project with four
layers, each with a custom
form, was created. In order
to have all the data fully offline,
vector tiles were used.
These were generated, for the
whole country, by extracting
OpenStreetMaps data, packaging
it in an mbtiles file, and
GEOmedia n°3-2022 9

FOCUS
by styling them with one of
the OpenMapTiles styles.
The resulting project was
only 16 MB, a small fraction
of the 160+ MB that would
have been needed for rasterized
tiles.
Using the QGIS drag-n-drop
form, extensive logic was
introduced in the data entry
user interface. With this
setup, users were guided in
choosing the three different
administrative levels from a
drop down, with automatic
filtering of the available options,
and constraints with
appropriate description were
implemented. For water
sources, which are of upmost
importance, a photo was also
required.
After the project was tested,
teams of surveyors covered
the whole country in the span
of a few months, while the
survey manager constantly
analyzed data quality with
spot crosschecks.
Periodically, the tablets were
brought back online, and
the data was synchronized.
In this process, the selective
sync option, introduced in
MerginMaps (at the time called
Input) 1.0.1, was crucial.
This feature instructs all the
tablets to upload the pho-
tos that were taken locally,
without downloading all
of the other media in the
project, that was added by
other surveyors. Without
this, more than 15 GB of
photos would have been
downloaded into each tablet,
severely impairing the
synchronization process and
requiring a stable and fast
internet connection.
At the completion of the
survey, the data was checked
and cleaned, then it
was synchronized with a
PostgreSQL/PostGIS database,
using the mergin-dbsync
tool, as described in
the “Extensions and integrations”
section. This procedure
initialized the new
database, and ensures that
any change in the data will
be reflected in the database
tables.
Using the newly initialized
database, a second QGIS
project based on the same
data was created and published
on a WebGIS based on
QGIS server and Lizmap,
thus reusing QGIS styling
without the need for restyling
and conversion. In
this phase, the advanced
forms could be reused, thus
showing on the website all
the information as entered
by the surveyor, including
the water source photo.
Other layers, such as the
administrative subdivisions,
were added, as well as the
localization tool, that enables
any user to quickly find
the current location, a village,
or an area of interest.
By combining the efficiency
of PostgreSQL materialized
view, and the flexibility
of the QGIS print layout,
multiple layouts were created
and personalized for
each administrative level,
10 GEOmedia n°3-2022

FOCUS
from country aggregates to
the village level. Graphs created
with DataPlotly, a recent
addition to the QGIS print
layout, were also used.
These layouts were then exposed
in Lizmap with the
AtlasPrint QGIS Server plugin.
Extensions and integrations
MerginMaps, being written
in Python, with good documentation,
has quite a few
extensions, that enable it to
adapt to the specific needs of
most survey projects.
Mergin-db-sync, first released
in June 2020, is a crucial
part of the MerginMaps
offering. Also written in
Python, it interfaces with the
main Mergin web service and
with a PostgreSQL/PostGIS
database, keeping the two
in constant sync. Whenever
a change is detected in the
specified geopackage, the
changes are propagated to the
PostgreSQL database, and
vice versa. Strict versioning
is still maintained, since the
tool creates a new version of
the MerginMaps project, just
as a user uploading new data
would. The tool uses two
PostgreSQL schemas, one in
which changes can be made
directly, and a backup copy
used to check for changes.
It utilizes the geodiff library
to check and merge changes,
even if they happen in the
two backends at the same
time. Mergin-db-sync can
also be used to expose the
data on a WebGIS such as
Lizmap.
Mergin-media-sync, first
released in December 2021,
allows for the offloading of
the media files, often representing
a good chunk of the
project size, to a local drive,
or to the MinIO object storage.
When new media files
are added, the tool downloads
them, uploads them to
the configured service, and
updates the relevant rows in
the geopackage, pointing the
media path to the new url. In
a wide-area survey, covering
many features and containing
photos, this tool can effectively
be used to avoid cluttering
the MerginMaps project
with hundreds of gigabytes of
images.
Both MerginMaps and
QField can be used with an
external GPS/GNSS device,
that can be obtained at a low
cost, enabling high precision
location, up to a centimeter
of accuracy. These devices,
once highly priced, are now
accessible and reliable.
METAKEYS
field survey; water resources; qgis; qfield
ABSTRACT
QGIS is the leading free and open source
desktop GIS. It is also a complete ecosystem,
that allows to build complete workflows, from
field data collection to publication on the web.
Central to it are QGIS projects, that define
data sources, projections, styling and integration,
and are reused from mobile to the web
without a need to reconfigure them. We describe
the main solutions available for data collection
and seamless publication over the web:
MerginMaps, Qfield, Lizmap, with an example
form a water resources project in Gambia.
AUTHOR
Paolo Cavallini
cavallini@faunalia.it
Matteo Ghetta
matteo.ghetta@faunalia.eu
Ulisse Cavallini
ulisse.cavallini@faunalia.it
Faunalia
www.faunalia.eu
Piazza Garibaldi 4, 56025 Pontedera
(PI), Italy
GEOmedia n°3-2022 11

FOCUS
Open-source GIS software and components
for modelling watershed phenomena
Understanding the soil and water components under
different management options with QGIS and the SWAT+
By Flavio Lupia and Giuseppe Pulighe
Fig. 1 – Workflow with input and output components for running SWAT+ through the QSWAT plugin for QGIS.
Modelling the
watershed balance
Current and future climate
change are expected to increase
our challenges in preserving
natural resources and ecosystem
services. At the watershed scale,
the processes taking places
relate to interactions between
soil and water and are influenced
by land use management.
Precipitation, infiltration, runoff,
evapotranspiration, soil
erosion, soil and water pollutions
are the main components
to be considered whenever for
the simulation of the watershed
system within the current and
future conditions under changing
drivers (i.e., human interventions
and climate change).
One of the most popular watershed
modelling tools is the
Soil and Water Assessment Tool
(SWAT), a public domain model
jointly developed by USDA
Agricultural Research Service
and Texas A&M University
(Arnold et al., 1998). SWAT
has been used worldwide for
different applications (water
quality, land use, soil erosion,
crop yield, etc.) during the last
four decades. As of May 2022,
a total of 5154 articles report
SWAT applications within
different journals according to
the SWAT Literature Database
(https://www.card.iastate.edu/
swat_articles/).
SWAT enables the simulation of
watershed and river basin quantity
and quality of surface and
ground water under the influence
of land use, management,
and climate change. It can be
used to monitor and control soil
erosion, non-point source pollution
and basin management. An
entirely reconstructed version of
SWAT, nicknamed SWAT+, was
only launched in recent years
to improve the capabilities of
SWAT code maintenance and
future development. Reservoir
operation functions have been
added to SWAT+ in addition
to the new model structure to
increase model simulation performance.
Now, the physical
objects (hydrologic response
units (HRUs), aquifers, canals,
ponds, reservoirs, point sources,
and inlets) are built as separate
modules.
QSWAT: the QGIS plugin for
the new SWAT+ model
SWAT model was implemented
withing the GIS environment
with dedicated plugins both for
commercial and open-source
GIS platforms. The GIS imple-
12 GEOmedia n°3-2022

FOCUS
mentations has allowed users to
manage more efficiently the watershed
modelling process in its
natural environment, the GIS,
where the spatial component
of the various datasets can be
handled straightforwardly.
QSWAT is the most recent
implementation as QGIS plugin,
written in Python, of the
new version SWAT+. As of 6th
April 2022, QSWAT3 v1.5 for
QGIS3 was released for 32 and
64bit machines. SWAT+ is written
in FORTRAN and is also
available as command-line executable
file that runs text file inputs
without interface (SWAT+
installer 2.1.0 was released on
31 March 2022 for Windows,
Linux, and MacOS). QSWAT
is increasingly gaining momentum
thanks to the spreading
and robustness of the opensource
GIS platform. QGIS has
a huge number of users and a
solid reputation. It is utilized in
academic and professional settings,
and it has been translated
into more than 48 languages.
Moreover, the release of SWAT
code as open source has benefitted
the diffusion and improvement
of the model by making
it more robust and suitable for
different applications thanks
to the collaboration of several
users with various expertise.
Different channels are available
for users’ collaboration such as
QSWAT user group, SWAT+
Editor user group and SWAT+
model user group (https://swat.
tamu.edu). Other plugins are
also available, such as the one
developed for the commercial
ESRI ARCGIS (ArcSWAT).
Beyond the functions provided
by QSWAT for setting
the watershed to be analysed,
SWAT+ is complemented by
additional software: SWAT
Editor (a user interface for
modifying SWAT+ inputs and
running the model installed
Fig. 2 – Spatial distribution of annual means of the actual evapotranspiration from the soil at subbasin scale for a
watershed through QSWAT.
along with QSWAT), SWAT+
Toolbox (a user-friendly tool
for SWAT+ for sensitivity
analysis, manual and automatic
calibration), SWATplus-CUP
(the Calibration Uncertainty
Program for SWAT+ requiring
a license purchase) and
SWATplusR (a set of tools
taking advantage of the R environment
for parameter sensitivity
analysis, model calibration
and the analysis model results).
Moreover, the SWAT website
provides datasets for running
the model even if specific datasets,
with adequate spatial and
temporal resolution, are always
recommended for the study
areas to be analysed. The datasets
available have often global
coverage and are relative to climate,
soil, land use and Digital
Elevation Models (DEMs).
Look up tables are also supplied
with QSWAT to properly
match to standard legends the
land use and soil codes.
The four-steps process and the
minimum set of data for running
SWAT+
The following spatial and tabular
data are required for running
SWAT+: Land use/cover, DEM,
Soil data (hydrological group,
clay, silt, sand), Climate data
(temperature, precipitation,
humidity, solar radiation, wind
speed) and Hydrology (river
discharge).
QSWAT runs SWAT+ by following
a four-step procedure: 1)
Delineate watersheds, 2) Create
Hydrologic Response Units
(HRUs), 3) Edit inputs and run
SWAT, and 4) Visualize.
The first step deals with the
definition of the watershed and
its structure by extracting the
channels and the watershed
boundary by processing the
DEM of the study area with
the classical Terrain Analysis
Using Digital Elevation Models
(TauDEM) functions that
divide the watershed in subbasins
(areas with a principal
stream channel). The second
step involves the creation of
the HRUs, lumped areas with
the same combination of soil,
topography, and land use, not
spatially related to each other
(Rathjens et al., 2016). The
third step concerns the weather
data selection and the set-up
of model parameters. For instance,
the latter are relative
to the potential evapotranspiration
method (e.g., Priestley-
GEOmedia n°3-2022 13

FOCUS
Taylor, Penman-Monteith or
Hargreaves), curve number
method for soil moisture, land
use management and conservation
practices. The fourth step
is dedicated to the visualization
of the results at basin and subbasin
scale and to the outputs
exploration for a given channel
and gauge.
The model can simulate daily,
monthly, yearly, and average
outputs for different model
components (e.g., channel,
aquifer, reservoir, etc.), for nutrient
balance, water balance,
plant weather and losses from
the basin, HRUs and landscape
units. Results can be printed
and exported in different formats
such as tabular or text
structure.
Calibration and validation
After the SWAT+ run and the
production of the outputs, model
calibration and validation
are strongly recommended.
The model can be calibrated
and validated for hydrologic,
sediment, nitrogen, and phosphorus
components. These
last steps guide the user on
the fine-tuning process of the
model parameters to produce
results coherent with the real
watershed processes. It requires
the collection of data on river
discharge that are often missing
for several watersheds or
available in analogic format,
moreover water quality data
(e.g., sediments load, dissolved
oxygen, nitrate and phosphorous
concentrations, etc.) can
be used. Alternative approaches
for hydrology calibration may
involve the use of evapotranspiration
data from satellite data
to overcome the lack of river
discharge data from the gauge
stations at the outlets. SWAT+
allows several options for calibrating
and validating the
simulation under the different
parametrizations defined by
the users. A sensitivity analysis
is quite common approach
undertaken to pinpoint the
main sensitive parameters and
to reduce their redundancy during
cal/val process. Literature
review is always useful to start
listing a set of common parameters
affecting streamflow and
sediment yield process. SWAT-
Calibration and Uncertainty
Programs (SWAT-CUP) is by
Fig. 3 – Schematic representation of the hydrology outputs at watershed level through QSWAT.
far the most known tool for
assessing the sensitivity of parameters
by providing several
model evaluation techniques
based on the relevant statistics
(e.g., Pearson’s correlation
coefficient, root mean square
error (RMSE), etc.). Following
the identification of the most
sensitive parameters the calibration
and validation phase
are carried out by focussing on
specific components (e.g., daily
discharge). The time series of
the available data (e.g., t0-tn) is
divided to provide a reference
period for the warm-up (e.g.,
t0-t5), calibration (e.g., t6-t15)
and validation (e.g., t16-tn).
Finally, model performances
are assessed by using classical
statistic measures (e.g., RMSE).
Calibration and validation can
be long and tedious. Therefore,
it is always recommended to
follow a precise work protocol
(Abbaspour et al. 2018).
Extending model simulation capabilities:
land use management
and climate change
The impact of alternative land
uses, and climate change are
pressing concerns in different
regions of the world. SWAT+
allows modelling diverse land
use scenarios to meet sustainability
goals (Pulighe et. al.,
2020) through a module where
alternative land management
practices can be defined (e.g.,
ploughing, seeding, tillage, irrigation,
fertilization rates and
crop nutrients uptake, etc.),
by defining dates or specific
land use classes and regions.
Similarly, climate projections
from climate models under different
representative concentration
pathways (RCPs) scenarios
of greenhouses emissions can be
loaded as weather data to create
climate scenarios for the future
decades that can be compared
to the baseline period covering
the historical meteorological
data. SWAT+ can ingest these
14 GEOmedia n°3-2022

FOCUS
data for simulating seasonal
changes in precipitation and
temperatures, hydrological extremes,
flow regime alterations
and river discharge, future water
quality (i.e., nitrogen and
phosphorus) and soil erosion
conditions, and future biomass
production. Estimating
the mentioned effects on the
hydrological regime might have
strong impacts also on agricultural
activities posing challenges
to land use management and
irrigation (Pulighe et al., 2021).
Conclusions
Open-source GIS (QGIS)
and free to use models such
as SWAT+ can be considered
effective and strategic tools for
monitoring and assessing water
and soil interactions at the
watershed level. In addition,
the growing availability of public
domain geospatial datasets
can increase the applicability
of the simulation of the watershed
processes worldwide and
for a wide variety of use cases.
QSWAT will be a valuable
tool for the SWAT scientific
community thanks to the full
integration with the geospatial
functions, the new functionality
offered by SWAT+ and
the contribution of a wide and
growing open-source community.
QSWAT could be a powerful
tool to assess the effects
of climate change and land use
management and the impacts
on water quality and land degradation.
We believe that in
the near future, the evaluation
of the effectiveness of policy interventions
and the deployment
of sustainable soil/water management
practices will become
an interesting arena for experimenting
and acknowledging the
potentiality of SWAT+.
REFERENCES
Arnold, J. G., Srinivasan, R., Muttiah, R. S., & Williams,
J. R. (1998). Large area hydrologic modeling
and assessment part I: model development 1. JAWRA
Journal of the American Water Resources Association,
34(1), 73-89.
Abbaspour KC, Vaghefi SA, Srinivasan R. A Guideline
for Successful Calibration and Uncertainty
Analysis for Soil and Water Assessment: A Review of
Papers from the 2016 International SWAT Conference.
Water. 2018; 10(1):6. https://doi.org/10.3390/
w10010006
Pulighe, G., Lupia, F., Chen, H., & Yin, H. (2021).
Modeling Climate Change Impacts on Water Balance
of a Mediterranean Watershed Using SWAT+. Hydrology,
8(4), 157.
Pulighe G, Bonati G, Colangeli M, Traverso L, Lupia
F, Altobelli F, Dalla Marta A, Napoli M. Predicting
Streamflow and Nutrient Loadings in a Semi-Arid
Mediterranean Watershed with Ephemeral Streams
Using the SWAT Model. Agronomy. 2020; 10(1):2.
https://doi.org/10.3390/agronomy10010002
Rathjens, H., Bieger, K., Srinivasan, R., Chaubey, I.,
& Arnold, J. G. (2016). CMhyd user manual. Doc.
Prep. Simulated Clim. Change Data Hydrol. Impact
Study.
https://www.card.iastate.edu/swat_articles/
https://swat.tamu.edu
METAKEYS
QGIS; QSWAT; watershed; river basin; SWAT+;
climate change
ABSTRACT
The Soil and Water Assessment Tool (SWAT) enables
the simulation of watershed and river basin quantity
and quality of surface and ground water under the influence
of land use, management, and climate change.
It can be used to monitor and control soil erosion,
non-point source pollution and basin management.
The recent version (SWAT+) was implemented by
a dedicated QGIS plugin (QSWAT) widening the
userbase and the potential modelling application
worldwide. QSWAT, along with additional software
for preparing the input dataset and for performing
the calibration/validation phase, further extends the
watershed modelling capabilities. Such tools and the
growing diffusion of public open geospatial datasets
are expected to increase the range of applications especially
with the availability climate projections datasets.
The latter will enable users to simulate all the watersoil
phenomena at watershed level under future conditions
to better understand and plan suitable action for
preserving the natural resources.
AUTHOR
Flavio Lupia
flavio.lupia@crea.gov.it
Giuseppe Pulighe
giuseppe.pulighe@crea.gov.it
CREA - Council for Agricultural Research
and Economics - Roma (Italy)
GEOmedia n°3-2022 15

REPORT
Mobile Robotics and Autonomous
Mapping: Technology for a more
Sustainable Agriculture
by Eleonora Maset, Lorenzo Scalera, Diego Tiozzo Fasiolo
Fig. 1 - The mobile robot traversing under the canopy in a
maize field (Manish et al., 2022).
Today, food systems account
for nearly onethird
of global greenhouse gas
emissions, consume
large amounts of natural
resources and are among the
causes of biodiversity loss.
As part of the actions of
the European Green
Deal, the Farm to Fork
Strategy (European Commission,
2020) plays therefore a
crucial role to reach the ambitious
goal of making Europe a
climate-neutral continent by
2050. In fact, it aims to accelerate
the transition towards a
sustainable food system, reducing
dependency on pesticides,
decreasing excess fertilization
and protecting land, soil, water,
air, plant and animal health. All
actors of the food chain need to
contribute to the implementation
of this strategy, starting
from the transformation of
production methods that can
benefit from novel technological
and digital solutions to deliver
better environmental and climate
results.
In this context, we are witnessing
an increasing demand for
automated solutions to monitor
and inspect crops and canopies,
that are driving the adoption of
autonomous and robotic systems
with computational and
logical capabilities. The introduction
of robotics and automation,
coupled with Geomatics
techniques, could provide notable
benefits not only in terms
of crop production and land use
optimization, but also to reduce
the use of chemical pesticides,
improving sustainability and
climate performance through
a more results-oriented model,
based on the use of updated
data and analyses. For these reasons,
the implementation of autonomous
and robotic solutions
together with advanced monitoring
techniques is becoming
of paramount importance in
view of a resilient and sustainable
agriculture.
Applications of mobile robotics
in agriculture span from a large
variety of tasks, as for instance,
harvesting, monitoring, phenotyping,
sowing, and weeding. A
particular task in which mobile
robots are currently employed
at a faster pace than in previous
years is 3D mapping, as testified
by a flourishing literature
on the topic (Tiozzo Fasiolo et
al., 2022). Indeed, 3D maps of
agricultural crops can provide
valuable information about the
health, stress, presence of diseases,
as well as morphological and
biochemical characteristics. Furthermore,
3D surveys of plants
and crops are fundamental in
the computation of geometrical
information, such as volume
and height, to be used to reduce
pesticide and fertilizer waste
and water usage, and, therefore,
improve sustainability and environmental
impact.
Obviously, to provide useful
information for crop management
and to perform the survey
in the most automatic way possible,
robotics platforms must
be equipped with appropriate
technology. In the following, we
will therefore try to summarize
trends and future developments
16 GEOmedia n°3-2022

REPORT
in this domain.
The first requirement of mobile
robots in agriculture applications
is the availability of onboard
sensors and computational capabilities.
Common sensors are
2D and 3D LiDAR (Light Detection
and Ranging), cameras
(monocular, stereo, RGB-D, and
time-of-flight ones), as well as
RTK-GNSS (Real-Time Kinematics
Global Navigation Satellite
System) receivers, and IMUs
(Inertial Measurement Units),
the latter two used mainly for
localization tasks.
Among the robotic systems recently
proposed in the literature
for 3D mapping in agriculture,
it is worth mentioning the platform
developed in (Manish et
al., 2021) and shown in Figure
1. That system is capable of collaborating
with a drone to build
a dense point cloud of the field.
Another interesting mobile robot
is BoniRob (Figure 2), developed
by Bosch Deepfield® Robotics
(Chebrolu et al., 2017). It is an
omnidirectional robot and carries
a multispectral camera, able
to register four spectral bands,
and an RGB-D sensor to capture
high-resolution radiometric data
about the inspected plantation.
Multiple LiDAR sensors and
GNSS receivers as well as wheel
encoders provide at the same
time observations employed for
localization, navigation, and
mapping. An example of robot
with an onboard manipulator
is given by BrambleBee (Ohi et
al., 2018). That robotic system
features a custom end effector
designed to pollinating flowers
in a greenhouse.
Images from standard RGB
cameras only supply information
about the plants in the visible
spectrum. To investigate vegetation
indexes related to the crop
vigor, as for instance the NDVI
(Normalized Difference Vegetation
Index), multispectral and
Fig. 2 - Agricultural field robot BoniRob with onboard sensors (Chebrolu et al., 2017).
hyperspectral sensors are needed,
which can measure the near
infrared radiation reflected by
the vegetation leaves. However,
only a paucity of robotic platforms
described in the literature
manage to perform this task.
The mobile lab developed at
the Free University of Bolzano
and shown in Figure 3 is among
them (Bietresato et al., 2016).
A prototype of mobile robot for
3D mapping in agriculture is
currently being developed at the
University of Udine, based on
an Agile-X Robotics Scout 2.0
platform (Figure 4). The robot
can navigate in harsh terrain and
narrow passages thanks to its
four-wheel drive and differential
kinematics. The platform is
equipped with a low-cost GNSS
receiver and a 9-degree-offreedom
(DOF) IMU as direct
georeferencing systems. Moreover,
it features a great computational
capability thanks to the
NVIDIA Jetson AGX Xavier
board, developed to exploit artificial
intelligence (AI) algorithms
even in embedded systems. The
perception of the environment
is guaranteed by a rotating 360°
LiDAR and an RGB-D camera.
Finally, for phenotyping purposes
it exploits a multispectral
camera pointing forward to
acquire information on the near
infrared and the red edge portion
of the light spectrum.
As far as the sensorial and
computational capabilities are
Fig. 3 - Agricultural robot developed at the Free University of Bolzano, Italy: robot in a orchard,
and onboard sensors (Bietresato et al., 2016).
GEOmedia n°3-2022 17

REPORT
Fig. 4 - The sensorized mobile platform developed at University of Udine, Italy.
considered, from the literature
it can be noticed that most of
the robotic platforms operating
in the agricultural environment
usually employ physical devices
to store the acquired data, whose
can require a frequent manual
intervention of an operator. The
implementation on Internet
of Things (IoT) approaches,
together with the storage of
data on clouds could be a great
improvement, making data
remotely available. Future improvements
in this context will
also include the integration of
renewable energy sources, such
as solar panels, to increase the
autonomy of the systems, especially
in large scale operations
(e.g., autonomous 3D mapping
of a whole vineyard). Moreover,
to avoid occlusion problems
that can occur in image-based
phenotyping, sensors can be
mounted on a robotic arm that
can optimize the camera pose,
guaranteeing the best point of
view for data acquisition. However,
it should also be underlined
that eye-in-hand configurations
for LiDAR sensors and multispectral
cameras are not exploited
yet. A further important
aspect is the durability of these
systems and sensors, that should
be designed to operate in severe
outdoor scenarios.
Fig. 5 - Person following with YOLO object detection (Masuzawa et al., 2017).
To navigate autonomously in
the surrounding environment, a
mobile robot needs a robust localization
method that can georeference
the data acquired by
means of the onboard sensors.
Direct georeferencing methods
are usually based on the RTK-
GNSS, that provides position at
low update rate, generally coupled
with a 9-DOF IMU, which
however is sensible to noise in
rough terrains. Higher accuracy
for the localization of the robot
and the generated 3D map can
be achieved using in addition
Simultaneous Localization and
Mapping (SLAM) approaches.
As well-known also in the Geomatics
community, SLAM problem
consists in the estimation
of the pose of the robot/sensor,
while simultaneously building a
map of the environment. Stateof-the-art
methods are divided
into two main groups: visual
SLAM and LiDAR SLAM. The
former approach relies on images
and sequentially estimates
the camera poses by tracking
keypoints in the image sequence.
The popular approach for Li-
DAR SLAM is instead based
on scan matching: the pose is
retrieved by matching the newly
acquired point cloud with the
previously built map, which is
constantly updated as soon as
new observations are available.
Although not yet fully implemented
in mobile robots for precision
agriculture applications,
an optimal solution could be
data fusion, taking the advantages
of both visual and LiDAR
SLAM methods. In addition,
since external conditions can significantly
vary among different
application and the environment
dictates the most advantageous
sensor, the robot itself should be
able to choose and use the most
suited data source according to
the environmental conditions.
Many open-source SLAM algo-
18 GEOmedia n°3-2022

REPORT
rithms are currently available,
that can run in real time or in
post-processing mode. To this
regard, a comparison among
the state-of-the-art SLAM algorithms
could be interesting,
together with a quantitative
evaluation of the obtained 3D
maps performed with respect
to ground truth datasets. Conversely,
there is a lack of methods
to efficiently fuse spectral data
acquired by multispectral and
hyperspectral cameras with Li-
DAR point clouds, fundamental
for agriculture applications.
Another important aspect that
must be considered for the
profitably application of mobile
platforms is the autonomous
navigation ability, guaranteed by
path planning algorithms. Path
planning is a mature field in
mobile robotics, and, in the crop
monitoring context, it generally
consists of providing a global
path to map an entire area. This
approach is called coverage path
planning and is usually coupled
with a row-following algorithm
to provide local velocity command
to the robot.
A recent trend in the coverage
path planning is the development
of algorithms that avoid
repetitive paths to minimize soil
compaction. This approach generally
relies on prior information
of the working area, that could
be acquired thanks to the collaboration
with drones, useful to
capture an up-to-date 2D or 3D
model of the environment. Furthermore,
a promising solution
could be extending the range
of action with swarm robotics,
that is the collaboration of several
unmanned ground vehicles
(UGVs).
Nowadays, another fundamental
aid for agricultural applications
based on mobile robotics is given
by artificial intelligence (AI). In
fact, classification and segmentation
algorithms of images and
point clouds based on AI are increasingly
used to enrich the 3D
map with semantic information,
also in real time. This is possible
mostly thanks to the advances in
the computational performance
of modern embedded computers
that can be installed onboard a
mobile platform.
For instance, a convolutional
neural network (CNN) applied
to the acquired images can provide
bounding boxes of objects
of interest, that can constitute
the basis to build a topological
map with key location estimation
and semantic information.
This is exploitable also to give
the robot person following capability,
as done in the work by
(Masuzawa et al., 2017) (Figure
5). Another example of machine
learning application is given
by the work in (Reina et al.,
2017), which employed a support
vector machine to classify
the terrain on which the robot
is navigating, by means of wheel
slip, rolling resistance, vibration
response experienced by the
mobile platform and visual data.
Recent trends in this field also
comprise the use of generative
adversarial networks to generate
photorealistic agricultural images
for model training, as well as
the recognition of diseases with
CNN.
In the coming years we will
witness great progress in all the
domains highlighted by this
work, from sensors to mobile
platforms, from localization algorithms
to artificial intelligence
methods, with the hope that
these innovations will effectively
contribute to the transition to
a more sustainable, healthy and
environmentally-friendly food
system.
REFERENCES
European Commission (2020). Farm to Fork
strategy for a fair, healthy and environmentallyfriendly
food system. https://ec.europa.eu/food/
horizontal-topics/farm-fork-strategy_en
Tiozzo Fasiolo, D., Scalera, L., Maset, E.,
Gasparetto, A. (2022). Recent Trends in Mobile
Robotics for 3D Mapping in Agriculture. In International
Conference on Robotics in Alpe-Adria
Danube Region (pp. 428-435). Springer, Cham.
Manish, R., Lin, Y. C., Ravi, R., Hasheminasab,
S. M., Zhou, T., Habib, A. (2021). Development
of a miniaturized mobile mapping system
for in-row, under-canopy phenotyping. Remote
Sensing, 13(2), 276.
Chebrolu, N., Lottes, P., Schaefer, A., Winterhalter,
W., Burgard, W., Stachniss, C. (2017).
Agricultural robot dataset for plant classification,
localization and mapping on sugar beet
fields. The International Journal of Robotics
Research, 36(10), 1045-1052.
Ohi, N., Lassak, K., Watson, R., Strader, J., Du,
Y., Yang, C., et al. (2018). Design of an autonomous
precision pollination robot. In 2018 IEEE/
RSJ international conference on intelligent robots
and systems (IROS) (pp. 7711-7718). IEEE.
Bietresato, M., Carabin, G., D'Auria, D., Gallo,
R., Ristorto, G., Mazzetto, F., Vidoni, R., Gasparetto,
A., Scalera, L. (2016). A tracked mobile
robotic lab for monitoring the plants volume and
health. In 2016 12th IEEE/ASME International
Conference on Mechatronic and Embedded
Systems and Applications (MESA). IEEE.
Masuzawa, H., Miura, J., Oishi, S. (2017). Development
of a mobile robot for harvest support
in greenhouse horticulture—Person following
and mapping. In 2017 IEEE/SICE International
Symposium on System Integration (SII) (pp.
541-546). IEEE.
Reina, G., Milella, A., Galati, R. (2017). Terrain
assessment for precision agriculture using vehicle
dynamic modelling. Biosystems engineering, 162,
124-139.
KEYWORDS
Mobile robotics; autonomous mapping; sustainable
agriculture
ABSTRACT
The introduction of robotics and automation,
coupled with Geomatics techniques, could
provide notable benefits not only in terms of crop
production and land use optimization, but also to
reduce the use of chemical pesticides, improving
sustainability and climate performance through
a more results-oriented model, based on the use
of updated data and analyses. For these reasons,
the implementation of autonomous and robotic
solutions together with advanced monitoring
techniques is becoming of paramount importance
in view of a resilient and sustainable agriculture.
AUTHOR
Eleonora Maset
eleonora.maset@uniud.it
Lorenzo Scalera
lorenzo.scalera@uniud.it
Diego Tiozzo Fasiolo
diego.tiozzo@uniud.it
Polytechnic Department of Engineering and
Architecture (DPIA), University of Udine,
Italy
GEOmedia n°3-2022 19

REPORT
Geographical Information: the Italian
Scientific Associations and...
the Big Tech
by Valerio Zunino
In Italy, the Scientific
Associations that deal with
geographical data are simply
not up to speed, and just a few
of them have some sort of idea
on how to get into it. The huge
and mind-bending projects
that Big Tech is dealing with in
our same world of professional
expertise, must be firstly
understood -deeply-, weighted
and brought to the table of
our heritage of knowledge,
culture, education, and lastly
of consulting services that our
Associations are required to
offer to the national market.
These days, we can no
longer afford to pretend
we can just barely
glimpse the revolutionary contribution
that has been made
to the world of professionals
around the planet by the big
techs (in our sector), among
others, through their generalist
geographic portals.
Nor can we continue to brand
as scientific approximation their
method in strategically tackling
that world that we, Scientific
Associations, together with the
most established companies
in the sector, believed we were
presiding over with the exclusivity
of knowledge and the
most advanced technology: the
time it takes to access an immense
amount of geographical
data residing on the internet
has soon become one tenth of
what we used to expect, and our
locking up within our Tender
Special Specifications will simply
make us less credible in the
eyes of that growing audience of
subjects that we call users, and
which evidently represents the
market of the Associates that we
have a duty to approach, starting
with small and mediumsized
enterprises and ending
with the individual professional
who is struggling at work and
in life. If we do not do this, we
will disappear.
The technological framework
and business model outlined
by those companies that today
- whether one accepts it or
not - mark the times, methods
and rules of consultation and
processing of the vast majority
of published geographical data,
and that have so far often been
seen as a kind of obstruction
to the scientific conversation,
it should be clearly stated as of
now that they will have to be
carefully studied and if possible
brought to the table of the
Associations, so as to first and
foremost qualify them. It is necessary
to at least begin to show
a general humility on the contents,
seek an encounter with
the Industry and attempt to
generate a global, adjustable and
- even more urgently- mutual
learning, that is for us the only
effective and possible means of
sharing.
On 11 June 2001, the global
market witnessed the release of
Google Earth. It took none of
us 'insiders' more than a couple
of minutes, the time to get to
grips with the surprise effect,
to measure an extraordinary
performance, that was not com-
20 GEOmedia n°3-2022

REPORT
parable (for how far it was superior)
to that of products such
as Autodesk Mapguide or the
direct competitors of the time,
positioned by ESRI, Intergraph
or Bentley, at the time leading
tools dedicated to the web consultation
of raster and vector geographic
data. The very concept
of raster was debased in just a
few moments, almost as if it
had been overtaken by still unknown
words, which had been
able to refer to and perhaps
even describe the practicality
and simplicity of the dynamic
and intelligent management of
remote sensed images, which
populated this formidable application
capable of occupying
the globe in a representation
without geographical hesitation
and continuity.
Earth then remained essentially
the same, integrating some
interesting functionalities over
time, which, however, could not
affect the first violent impact of
product innovation. It is a fact
that the other Big Techs were
not able to respond rapidly, and
did not want to do so in the
years immediately following.
Consequently, it is precisely
from 2001 onwards that Google
began to dig a trench that for
a while increased in width (in
terms of the quantity of the
information entered) and in
depth (in terms of its quality
and geographical accuracy);
then, the depth was filled by a
number of competitors, as a result
of which the big geographical
data market was hit by an
unprecedented rivalry based on
quality: an arms race that saw
first Apple enter the game, with
the initial and fundamental
support of a very strong and acclaimed
segment brand as Tom-
Tom still is... then Amazon, and
closely followed by Facebook
and Microsoft, all of which,
although leading the growth of
a robust proprietary mapping
sector, are to a greater or lesser
extent still heavily anchored,
and we are talking about truly
significant investments, to a
global geocartographic system,
probably born (Joe Morrison)
out of a conversation between
recent graduates in an English
pub in 2004, and whose commercial
value is now out of control:
OpenStreetMap.
Of the interesting and singular
reasons that have at the moment
prevented the big corporations
from replicating OSM,
suggesting them instead to
invest in it by bringing in their
own teams of editors, we will
perhaps speak on another occasion.
What seems more on topic
now, however, is to report on
the evolutionary framework of
OSM content at the hands of
the big brands of the IT world.
Within the geographic areas
(States, regions, crucial Human
Settlements, etc.) where the
white collar teams of the big
names are active, the average
incidence of editors operating
on a voluntary basis is today less
than 25% of the entire OSM
geographic road/building data
operation, whereas in 2017 this
figure was around 70% (Jennings
Anderson). Now, given
that the big corporations are
in this way impoverishing the
ideological path from which the
GEOmedia n°3-2022 21

REPORT
spirit of community that inspired
the birth of the great free
geographic portal had sprung,
it remains to be seen what else
these corporations are doing at
the moment, each on their own
account and more or less with
the headlights off.
Otherwise known as
F.A.A.N.G. (Facebook,
Amazon, Apple, Netflix and
Google), sometimes with the
inclusion of Microsoft, these
are the Big Techs that for some
time have been influencing our
behavior, our choices, our purchases
and probably even our
attitudes.
In the vicinity of Seattle (by the
way, not without the support
of the large Indian headquarters
in Hyderabad, announced
in September 2019 and now
operating with about 15,000
engineers, structured in an
area of almost 170,000 square
meters) they are working on at
least two fronts (we are talking
about mapping, of course):
on the one hand, the Amazon
Location Service project, born
together with ESRI and HERE
Technology B.V. from a rib of
Amazon Web Service, the latter,
today a cloud computing platform
with a major competitive
advantage over the analogues
provided by Microsoft (Azure)
and Google (GCP). The abovementioned
partnership serves
to fill the gap that Amazon, like
the others, also suffers in terms
of proprietary cartography (or
acquired in perpetual license
without disbursement of any
fee for the benefit of the relevant
suppliers): but while ESRI
makes available to its partner
some high-definition satellite
databases, HERE contributes
through the provision of its
own geographic vectorial data
referring in particular to road
circulation, real-time traffic and
address location.
Of course, partners receive payment
on an on-demand (click)
basis whenever the Amazon
Location Service user performs
an operation on geography, processes
a route request, performs
a different search, etc... And it
is also for this reason that Amazon
is also moving on its own
account, in order to free itself
from such costs. As is typical for
big brands, once the allurement
of a market has been verified, it
is considered a serious strategic
mistake to wait too long before
being present, consequently
Amazon Location Services, like
other platforms, simply had to
be born, necessarily together
with selected and established
partners in the segment. So,
the goal was to get into it right
away, more or less, to steady a
service and then innovate and
improve it, just as it was battling
on market share points
against its longtime competitors.
Before long, Amazon will
reduce the quantitative contribution
of its consume-based
geographic data providers, and
resubmit a quasi-proprietary
version of Amazon Location
Services on the most important
element, the mapping system:
you can bet on it.
Facebook and Microsoft are also
investing in the same endeavor
to reduce the – as the present
day - gross imbalance between
third-party geographic data and
data owned or acquired outright
without conditions on publication;
the platforms are called,
for the uninitiated, Facebook
Maps and Microsoft Azure
Maps, respectively. But while
Zuckenberg's creature (today
“Meta”) does not seem to show
any particular interest in a race
for proprietary geographic information
endowments - if not
for certain categories or thematic
classes - thereabout Redlands
we are today witnessing an
interesting acceleration, which
also concerns 2D buildings,
published and also made available
in Opendata just in the last
few weeks by the Bing subsidiary
and with reference to a long
series of countries, including
Italy. The fullness of the data is
good (we tested it for parts of
our country) and the quality
certainly more than decent.
And while what happens in
Mountain View is always
22 GEOmedia n°3-2022

REPORT
somewhat surrounded by that,
more or less, invisible aura of
mystery that almost always intrigues,
Cupertino, eventually,
shows detail and quality with
the famous proprietary map of
California, representing and
symbolizing down to individual
plants, in some cities. Apple
New Map is indeed the qualitative-quantitative
manifesto,
reporting the formidable global
geographic wishes of the brand
founded by Steve Jobs. From a
strictly GIS-oriented point of
view, it is the most ambitious
project.
In conclusion, will we, as Associations
for the protection,
dissemination, and comparison
of geographic data in Italy, be
able to keep up with the pace of
a category credibility that envisages,
without compromises, a
greater openness of our awareness
and a more convincing
manifestation of our somewhat
repressed humility?
These are reactions that are neither
easy nor quick, but necessary.
To young people, who are
entering the world of the Associations
federated in ASITA we
say and recommend that they
open up, open their vision of
the market that will one day be
theirs alone, in the direction of
others, Public sector, Big Tech,
Professional world, international
segment majors, national
Industry, etc... Geoinformation,
sooner or later, will have to
become one: we better realize it
sooner than later.
KEYWORDS
ASITA; Geographic information; big
tech; GIS; AMFM; location services
ABSTRACT
The world is changing. The Italian
Scientific Associations of Geographical
Information are not. The opportunities
are endless, but what is missing
is humility and ideas, and often these
two shortcomings feed off each other.
AUTHOR
Valerio Zunino
valerio.zunino@studiosit.it
Vice-President Association AMFM
GIS Italia
GEOmedia n°3-2022 23

Rhine River, Germany
The Rhine River, the longest river in Germany, is
featured in this colourful image captured by the Copernicus
Sentinel-2 mission. The Rhine River, visible here in black, flows from
the Swiss Alps to the North Sea through Switzerland, Liechtenstein, Austria,
France, Germany, and the Netherlands. In the image, the Rhine flows from
bottom-right to top-left. The river is an important waterway with an abundance of
shipping traffic, with import and export goods from all over the world. The picturesque
Rhine Valley has many forested hills topped with castles and includes vineyards, quaint
towns and villages along the route of the river. One particular stretch that extends from Bingen
in the south to Koblenz, known as the Rhine Gorge, has been declared a UNESCO World
Heritage Site (not visible). Cologne is visible at the top of the image. This composite image was
created by combining three separate Normalised Difference Vegetation Index (NDVI) layers from
the Copernicus Sentinel-2 mission. The Normalised Difference Vegetation Index is widely used in
remote sensing as it gives scientists an accurate measure of health and status of plant growth.Each
colour in this week’s image represents the average NDVI value of an entire season between 2018
and 2021. Shades of red depict peak vegetation growth in April and May, green shows changes in
June and July, while blue shows August and September. Colourful squares, particularly visible
in the left of the image, show different crop types. The nearby white areas are forested areas
and appear white as they retain high NDVI values through most of the growing season,
unlike crops which are planted and harvested at set time frames. Light pink areas are
grasslands, while the dark areas (which have a low NDVI) are built-up areas and
water bodies.
[Credits: contains modified Copernicus Sentinel data (2018-21),
processed by ESA - Translation: Gianluca Pititto]

REPORT
Time and Longitude:
an unexpected affinity
by Marco Lisi
Time, the fourth dimension, is becoming increasingly important in all
aspects of technology and science.
The generation and distribution of an accurate reference time is a
strategic asset on which the most disparate applications depend: from
financial transactions to broadband communications, from satellite
navigation systems to large laboratories for basic physical research (the
so-called "Big Physics ").
But time is also the dimension through which technology evolves (as,
for example, in the case of Moore's law which describes the increase in
complexity of integrated electronic circuits) and obsolescence spreads.
Obsolescence will be the great challenge, often ignored or
underestimated, that the economically and technologically advanced
societies of the world will have to face in the years to come. The more
technology increases its evolutionary pace, the more things that
surround us quickly become “old”, as they are no longer able to interface
with each other and be maintained.
Maintenance and updating of obsolete parts (the so-called “logistics”)
are essential aspects in the operational life of a system and both have to
do with time.
The importance of a precise
time reference in our society
and economy
The determination and the accurate
measurement of time are
the basis of our technological
civilization. The major advances
in this field have taken place
in the last century, with the
invention of the quartz crystal
oscillator in 1920 and the first
atomic clocks in the 40s. Nowadays
time measurement is by far
the most accurate among the
measures of other fundamental
physical quantities. Even the
measurement unit for lengths,
once based on the mythical reference
meter, a sample of Platinum-Iridium
preserved in Paris,
was internationally redefined
in 1983 as "the length of the
path covered by light in vacuum
during a time interval equal to
1/299792458 of a second ".
The second (symbol “s”) is the
unit of measurement of the official
time in the International
System of Units (SI). Its name
comes simply from the second
division of the hour, while the
minute is the first. The second
was originally defined as the
86400-th part of the mean solar
day, i.e., the average, taken over
a year, of the solar day, defined
as the time interval elapsing between
two successive passages of
the Sun on the same meridian.
In 1884 the Greenwich Mean
Time (GMT) was officially
established as the international
standard of time, defined as the
mean solar time at the meridian
passing through the Royal
Observatory in Greenwich
(England).
26 GEOmedia n°3-2022

REPORT
GMT calculates the time in each
of the 24 zones (time zones) into
which the earth's surface has
been divided. The time decreases
by one hour for each area west of
Greenwich, and increases by one
hour going east. GMT is also
defined as "Z" time, or, in the
phonetic alphabet, "Zulu" time.
The time standard underlying
the definition of GMT was
maintained until astronomers
discovered that the mean solar
day was not constant, due to the
slow (but continuous) slowdown
of the Earth's rotation around
its axis. This phenomenon is essentially
linked to the braking
action of the tides. It was then
decided to refer the average solar
day to a specific date, that of
January 1, 1900. This solution
was very impractical since it is
not possible to go back in time
and measure the duration of that
particular day.
In 1967 a new definition of the
second was proposed, based on
the motion of precession of the
isotope 133 of Cesium. The second
is now defined as the time
interval equal to 9192631770
cycles of the vibration of Cesium
133. This definition allows scientists
anywhere in the world
to reconstruct the duration of
the second with equal precision
and the concept of International
Atomic Time or TAI is based on
it.
The first atomic clock was developed
in 1949 and was based
on an absorption line of the
ammonia molecule. The cesium
clock, developed at the legendary
NIST (National Institute
of Standards and Technology)
in Boulder, Colorado, can keep
time with an accuracy better
than one second in six million
years. It was precisely the
extreme accuracy of atomic
clocks that led to the adoption
of atomic time as an official reference
worldwide. However, a
Fig. 1 - UTC and critical infrastructures.
Fig. 2 - Stonehenge, a prehistoric astronomical observatory.
new problem was been indirectly
generated: the discrepancy between
the international reference
of time, based as mentioned on
atomic clocks, and the average
solar time. An average solar year
increases by about 0.8 seconds
per century (i.e., about an hour
every 450,000 years). Consequently,
universal time accumulates
a delay of approximately 1
second every 500 days compared
to international atomic time.
This means that our distant
great-grandchildren, in the
distant future just 50,000 years
from now, would read “noon” on
their atomic clocks, even though
they are actually in the middle of
the night. To overcome this and
many other more serious drawbacks,
the concept of Universal
Coordinated Time (UTC) was
introduced in 1972, which definitively
replaced GMT.
In the short term, UTC essentially
coincides with atomic time
(called International Atomic
Time, or TAI); when the difference
between UTC and TAI approaches
one second (this occurs
approximately every 500 days),
a fictitious second, called "leap
second", is introduced.
In this way, the two time-scales,
TAI and UTC, are kept within
a maximum discrepancy of 0.9
seconds.
UTC ("Universal Coordinated
Time"), defined by the historic
“Bureau International des Poids
et Mesures” (BIPM) in Sevres
GEOmedia n°3-2022 27

REPORT
Fig. 3 - Ancient Egyptian stone obelisks.
(Paris), is since 1972 the legal
basis for the measurement of
time in the world, permanently
replacing the old GMT. It is
derived from TAI, from which it
differs only by an integer number
of seconds. TAI is in turn
calculated by BIPM from data
of more than 200 atomic clocks
located in metrology institutes
in more than 30 countries over
the world.
But why is it so important to
have an accurate and unambiguous
definition of time?
It is a matter not only for scientists
and experts. A universally
recognized and very accurate
reference time is in fact at the
base of most infrastructures of
our society (figure 1).
All cellular and wireless networks,
for example, are based on
careful synchronization of their
nodes and base stations (obtained
receiving GNSS signals,
as we will see). The same is true
for electric power distribution
networks. Surprisingly, even
financial transactions, banking,
and stock markets all depend on
an accurate time reference, given
the extreme volatility in equity
and currency markets, whose
quotations might vary within a
few microseconds.
Time and its measurement
The history of the measurement
of time is as old as the history of
human civilization.
In prehistoric England, the megalithic
monument of Stonehenge
seems to have been a sophisticated
astronomical observatory
to determine the length
of the seasons and the date of
the equinoxes (figure 2).
Already in 3500 BC the ancient
Egyptians invented the
sundial and erected stone obelisks
throughout their country
which had the primary purpose
of marking the movement of
the sun with their shadow and,
therefore, the passage of time
(figure 3).
In ancient Roman times and up
until late in the Medieval Age,
sundials, marked candles, water
and sand hourglasses were used
to measure time (figure 4).
A milestone in the history of the
measurement of time was, in
more recent times, Galileo's discovery,
in 1583, of the constancy
of the pendulum swing period,
on which all mechanical clocks
are based (figure 5).
In 1656 Christiaan Huygens,
Dutch mathematician, astronomer,
and physicist (famous
among other things for having
defined the principle of diffraction
that bears his name)
designed the first weight-wound
pendulum clock, which deviated
by ten minutes a day (figure 6).
But the major impetus for the
development of ever more accurate
techniques for measuring
time came from the need to
determine one's position (particularly
longitude) aboard a
ship in the open sea. From then
on, time and positioning became
irreversibly connected.
Fig. 4 - Roman and Medieval time measurement methods.
“Longitude problem” and
measurement of time
The latitude and longitude coordinate
system is commonly used
to determine and describe one’s
position on Earth’s surface and it
was also known by astronomers
28 GEOmedia n°3-2022

REPORT
and navigators since the Greek
and Roman times.
Determining latitude north or
south with respect to the equator
posed no major problems:
it could be calculated through
angular measurements of the sun
and stars made with relatively
simple instruments.
Measuring longitude, that is,
identifying the east-west position
on Earth between meridians,
lines running from pole to pole,
was a completely different story.
Longitude was far more difficult
than latitude to measure by astronomical
observation.
Because of the Earth’s rotation,
the difference in longitude between
two locations is equivalent
to the difference in their local
times: one degree of longitude
equals a four-minute time difference,
and 15 degrees is equal to
one hour (making 360 degrees,
or 24 hours, in total).
While a sextant with which to
determine the height of the sun
at noon was sufficient to determine
one's latitude, the determination
of longitude, due to the
earth's rotation, required the use
of both the sextant and a very
precise clock.
Several methods had been
proposed over the centuries by
scientists and astronomers (including
Galileo and Newton), all
based on the observation of specific
astronomical events, such as
lunar eclipses.
All these methods turned out
to be rather cumbersome and
inaccurate by several hundred
kilometers.
Even Christopher Columbus
made two attempts to use lunar
eclipses to discover his longitude,
during his voyages to the
New World, but his results were
affected by large errors.
The lack of an accurate longitude
determination method created
innumerable problems (at
times, real disasters) for sailors of
Fig. 5 - Galileo Galilei discovered in 1581 the isochronism of the pendulum.
the 15th and 16th centuries.
At the beginning of the eighteenth
century, with the rapid
growth of maritime traffic, a
sense of urgency had arisen. The
search for longitude cast a shadow
over the life of every man at
sea, and the safety of every vessel
and merchant ship.
The exact measurement of longitude
seemed at that time an
impossible dream, a sort of perpetual
motion machine.
There was a need for an instrument
that recorded the time (at
the place of departure) with the
utmost precision during long sea
voyages, despite the movement
of the ship and the adverse climatic
conditions of alternating
Fig. 6 - Christiaan Huygens and the first pendulum clock.
hot and cold, humid and dry.
On the other hand, seventeenthcentury
and early eighteenthcentury
clocks were crude devices
that usually lost or gained up
to a quarter of an hour a day.
The “longitude problem” however
became so serious that in
1714 the British Parliament
formed a group of well-known
scientists to study the solution,
the “Board of Longitude”. The
Board offered twenty thousand
pounds, equivalent to more than
three million pounds today, to
anyone who could find a way
to determine the longitude of a
ship on the open sea with accuracy
within one-half of a degree
(thirty nautical miles, about
GEOmedia n°3-2022 29

REPORT
Figure 7: Right: John Harrison – Left (clockwise): H1 thru H4 Harrison’s chronometers
Fig. 8 - James Cook’s Pacific Voyages.
Fig. 9 - Galileo Passive Hydrogen Maser (PHM) clock.
fifty-five kilometers, at the equator).
The approach was successful,
despite the many (and often
completely crazy) proposals. In
fact, in 1761, a self-educated
Yorkshire carpenter and amateur
clock-maker named John Harrison
built a special mechanical
clock to be loaded on board
ships, called the “marine chronometer”,
capable of losing or
gaining no more than one second
per day (an incredible accuracy
for that time) (figure 7).
Harrison did not receive the
prize from the Board until after
fighting for his reward, finally
receiving payment in 1773, after
the intervention of the British
parliament.
And it was thanks to a copy of
Harrison's H4 chronometer
that Captain James Cook made
his second and third legendary
explorations of Polynesia and
the Pacific islands on board the
HMS Resolution (figure 8).
A copy of the H4 chronometer
was also used in 1787 by
Lieutenant William Bligh, commander
of the famous HMS
Bounty, but it was retained by
Fletcher Christian following his
mutiny. It was later recovered
in Pitcairn Island to eventually
reach the National Maritime
Museum in London.
GNSS and Timing
An extremely accurate UTC reference
is today provided worldwide
by satellite navigation
systems (GNSS) such as GPS
(Global Positioning System),
GLONASS, Beidou, and the
European Galileo system. They
are systems of satellites orbiting
around the Earth, each containing
onboard extremely precise
atomic clocks which are all synchronized
to a system reference
clock.
GNSS technologies are intrinsically
linked to accurate timing.
30 GEOmedia n°3-2022

REPORT
This is because of the specific
principle (trilateration) on which
position determination is based,
i.e., the method of measuring
the distance of a user from each
satellite, involving the measurement
of the time delay experienced
by the signal-in-space.
The most accurate and numerous
atomic clocks around the
world are those belonging to
GNSS, thus contributing substantially
to the derivation of
TAI and UTC (figure 9).
UTC can be derived from
the Galileo and GPS signals,
through a series of corrections
based on data provided by the
signals themselves. The accuracy
obtainable, even with very cheap
commercial receivers (or in
those integrated into our smartphones)
is easily better than one
microsecond.
KEYWORDS
GNSS; GPS; Galileo; GLONASS; Beidou; time; longitude; GMT; TAI; UTC;
ABSTRACT
To have an accurate and unambiguous definition of time is a matter not only for scientists
and experts. A universally recognized and very accurate reference time is in fact at the
base of most infrastructures of our society. All cellular and wireless networks, for example,
are based on careful synchronization of their nodes and base stations (obtained receiving
GNSS signals, as we will see). The same is true for electric power distribution networks.
Surprisingly, even financial transactions and banking and stock markets all depend on
an accurate time reference, given the extreme volatility in equity and currency markets,
whose quotations might vary within a few microseconds. The history of the measurement
of time is as old as the history of human civilization. But the major impetus for the
development of ever more accurate techniques for measuring time came from the need to
determine one's position (particularly longitude) aboard a ship in the open sea. In 1761,
a self-educated Yorkshire carpenter and amateur clock-maker named John Harrison built
a special mechanical clock to be loaded on board ships, called the “marine chronometer”,
capable of losing or gaining no more than one second per day (an incredible accuracy for
that time). From then on, time and positioning became irreversibly connected.
AUTHOR
Marco Lisi
ingmarcolisi@gmail.com
Time to THE FUSION !
BLK2GO & X-PAD OFFICE FUSION
3D scanning and data processing
are now simpler than ever.
GeoMax provides the solution to increase your efficiency and accuracy.
BLK2GO, a handheld 3D imaging scanner, captures models and point clouds and
X-PAD OFFICE FUSION, GeoMax geodata office software, processes them in a few clicks.
©2022 Hexagon AB and/or its
subsidiaries and affiliates.
All rights reserved.
Part of Hexagon
GEOmedia n°3-2022 31
Works when you do

NEWS
in January 2021. In
December 2020, the
Space Development
Agency selected L3Harris
to build and launch four
space vehicles to demonstrate
the capability to
detect and track ballistic
and hypersonic missiles.
L3HARRIS INFRARED SPACE
TECHNOLOGY TO ENHANCE
BATTLEFIELD IMAGERY AND
MISSILE DEFENCE DETECTION
L3Harris is providing the instrument as part of a
wide-field-of-view satellite that also will help inform
future space-based missile defense missions
and architectures. The satellite will be positioned
22,000 miles from Earth, enabling the infrared system
to see a wide swath and patrol a large area for
potential missile launches.
“The L3Harris instrument can stare continuously
at a theater of interest to provide ongoing information
about the battlespace, which is an improvement
over legacy systems,” said Ed Zoiss,
President, Space & Airborne Systems, L3Harris.
“It also provides better resolution, sensitivity and
target discrimination at a lower cost.”
The instrument was built for Space Systems
Command and is integrated into a Millennium
Space Systems satellite, scheduled to launch from
Cape Canaveral, Florida. The payload, which is
more than six feet tall and weighs more than 365
pounds, was developed in Wilmington, Mass.
L3Harris is prioritizing investments in space-based
missile defense programs and has accelerated
the development of resilient, end-to-end satellite
solutions in spacecraft, payloads and ground software,
and advanced algorithms.
In a related effort, the Missile Defense Agency
awarded L3Harris a missile-tracking study contract
in 2019 and the prototype demonstration
About L3Harris
Technologies
L3Harris Technologies is
an agile global aerospace
and defense technology
innovator, delivering
end-to-end solutions that
meet customers’ missioncritical
needs. The company
provides advanced defense and commercial
technologies across space, air, land, sea and cyber
domains. L3Harris has more than $17 billion in
annual revenue and 47,000 employees, with customers
in more than 100 countries. L3Harris.
com.
Forward-Looking Statements
This press release contains forward-looking statements
that reflect management's current expectations,
assumptions and estimates of future performance
and economic conditions. Such statements
are made in reliance upon the safe harbor provisions
of Section 27A of the Securities Act of 1933
and Section 21E of the Securities Exchange Act of
1934. The company cautions investors that any
forward-looking statements are subject to risks
and uncertainties that may cause actual results
and future trends to differ materially from those
matters expressed in or implied by such forwardlooking
statements. Statements about the value or
expected value of orders, contracts or programs
are forward-looking and involve risks and uncertainties.
L3Harris disclaims any intention or obligation
to update or revise any forward-looking
statements, whether as a result of new information,
future events, or otherwise.
32 GEOmedia n°3-2022

NEWS
EMLID RELEASED THE PPK APP—EMLID
STUDIO FOR MAC AND WINDOWS
Emlid announced new PPK software—Emlid Studio. It’s a
cross-platform desktop application designed specifically for
post-processing GNSS data. The app is free and available
for Windows and Mac users.
Emlid Studio features a simple interface that makes postprocessing
easier than ever. The app allows users to convert
raw GNSS logs into RINEX, post-process static and kinematic
data, geotag images from drones, including DJI, and
extract points from the survey projects completed with the
ReachView 3 app.
With Emlid Studio, you can post-process data recorded
with Emlid Reach receivers and other GNSS receivers
or NTRIP services. For post-processing, you will need
RINEX observation and navigation files. You can also use
raw data in the UBX and RTCM3 format—Emlid Studio
will automatically convert them into RINEX.
The post-processing workflow is very straightforward.
You can receive precise positioning of a single point or
track depending on your positioning mode. Just add several
RINEX files and enter the antenna height. Click the
Process button, and Emlid Studio will do the rest. Once
the resulting position file is ready, you will see the result
on the plot.
One more tool is available for the users of Reach receivers
and the ReachView 3 app. The Stop & Go feature allows
you to improve the coordinates of points collected in Single
or Float modes.
Another helpful feature is geotagging for drone mapping.
To add geotags to the images’ EXIF data, you’ll need aerial
photos and the POS file with the events. Emlid Studio also
provides a chance to update your data from the RTK drone
in case you had a float or single solution during your
survey. You will need a set of RINEX logs from a base and
drone, MRK file, and images from the drone. Just drag and
drop data in the file slots and you’ll see the result in a few
seconds.
To start using Emlid Studio, simply download the app for
your computer—either Windows or macOS. To learn
more about Emlid Studio features, visit the Emlid website.
There's life in our world
We transform and publish data, metadata
and services in conformance to INSPIRE
We support Data Interoperability,
Open Data, Hight Value Datasets,
APIs, Location Intelligence, Data Spaces
INSPIRE Helpdesk
We support all INSPIRE implementers
Epsilon Italia S.r.l.
Viale della Concordia, 79
87040 Mendicino (CS)
Tel. (+39) 0984 631949
info@epsilon-italia.it
www.epsilon-italia.it
www.inspire-helpdesk.eu
GEOmedia n°3-2022 33

NEWS
TOPCON REPRESENTS CONSTRUCTION
INDUSTRY IN "CAMPUSOS" 5G RESEARCH
PROJECT
Topcon Positioning Germany is one of 22 partners involved
in CampusOS; a research project with the goal of developing
a modular ecosystem for open 5G campus networks
based on open radio technologies and interoperable
network components. As part of the German technology
program titled "Campus networks based on 5G communication
technologies," innovative solutions for open 5G
networks are being developed and tested in conjunction
with the German Federal Ministry for Economic Affairs
and Climate Protection. The program was launched at the
beginning of 2022 and will run through 2025.
The use of artificial intelligence in the operation of autonomous
plants and construction machinery requires the
highest level of digital sovereignty. If Construction 4.0, including
far-reaching automation, is to become a reality in
Germany and the rest of the world, the processes of such
data-driven solutions must run reliably, quickly and autonomously.
Sponsored by the Federal Ministry of Economics
The German Federal Ministry for Economic Affairs and
Climate Protection is providing around 18.1 million euros
in funding for the technology program over the next three
years, which will cost around 33 million euros in total. The
Fraunhofer Institutes FOKUS and HHI are coordinating
the project. 22 partners from industry and research are involved.
They include Deutsche Telekom, Siemens, Robert
Bosch and more.
"To enable companies to operate their own campus networks,
certain requirements must be met; from standardized
technology building blocks to network structures. As the
sole representative of the construction industry, Topcon
will test the technologies on reference test sites and therefore,
will help shape the solutions for the future," explains
Ulrich Hermanski, Chief Marketing Officer of the Topcon
Positioning Group. "We look forward to working with our
research partners to take the digital construction site to the
next level."
The future of the construction industry is digital
With this research project, construction companies will one
day be able to operate plants and machinery autonomously
in open campus networks. This will allow the fluid and
uninterrupted monitoring of construction sites in real time,
as well as the networking of all sensors and construction
machines in use on construction sites.
Completely autonomous from public networks, 5G technology
guarantees seamless machine-to-machine communication
and transmits data ten times faster than 4G.
The campus networks required for this, based on 5G frequencies,
are practically digital ecosystems. They operate
with open radio technologies and dialog-enabled components.
The campus networks are geographically limited and
can operate on a factory floor or on a construction site.
Hermanski explains: "We will put a lot of time and energy
into this project, because 5G campus networks are an important
key technology for the construction site of the future."
Lead project CampusOS: The consortium and its partners
In addition to Topcon Deutschland Positioning GmbH,
the collaborative partners of the CampusOS lead project include:
atesio GmbH, brown-iposs GmbH, BISDN GmbH,
Robert Bosch GmbH, Deutsche Telekom AG, EANTC
AG, Fraunhofer Institutes FOKUS and HHI (project coordinators),
GPS Gesellschaft für Produktionssysteme
GmbH, highstreet technologies GmbH, Kubermatic
GmbH, MUGLER SE, Node-H GmbH, Rohde & Schwarz
GmbH, rt-solutions. de GmbH, Siemens AG, Smart
Mobile Labs AG, STILL GmbH, SysEleven GmbH, the
Technical University of Berlin and the Technical University
of Kaiserslautern.
About Topcon Positioning Group
Topcon Positioning Group is an industry leading designer,
manufacturer and distributor of precision measurement
and workflow solutions for the global construction, geospatial
and agriculture markets. Topcon Positioning Group
is headquartered in Livermore, California, U.S. (topconpositioning.com,
LinkedIn, Twitter, Facebook). Its European
head office is in Capelle a/d IJssel, the Netherlands. Topcon
Corporation (topcon.com), founded in 1932, is traded on
the Tokyo Stock Exchange (7732).
34 GEOmedia n°3-2022

NEWS
INSPIRATION
FOR A SMARTER
WORLD
GET YOUR FREE TICKET NOW!
VOUCHER CODE: IG22-GEOM
TICKETSHOP GOES LIVE ON APRIL 28th
Veranstalter / Host: DVW e.V.
Ausrichter Conference / Conference organiser: DVW GmbH
Ausrichter Expo / Expo organiser: HINTE GmbH
WWW.INTERGEO.DE
GEOmedia n°3-2022 35

NEWS
GALILEO GNSS FOR THE ASSET MAPPING
PLATFORM FOR EMERGING COUNTRIES
ELECTRIFICATION
The purpose of the AMPERE (Asset Mapping Platform for
Emerging countRies Electrification) project is to provide
a dedicated solution for electrical power network information
gathering. AMPERE can actually support decision making
actors (e. g. institutions and public/ private companies
in charge to manage electrical network) to collect all needed
info to plan electrical network maintenance and upgrade.
In particular, the need for such a solution comes in emerging
countries where, despite global electrification rates are
significantly progressing, the access to electricity is still far
from being achieved in a reliable way. Indeed, the challenge
facing such communities goes beyond the lack of infrastructure
assets: what is needed is a mapping of already deployed
infrastructure (sometime not well known!) in order
to perform holistic assessment of the energy demand and
its expected growth over time. In such a context, Galileo
is a key enabler -especially, considering its free-of-charge
High Accuracy Service (HAS) and its highly precise E5
AltBOC code measurements- as a core component to map
electric utilities, optimise decision making process about the
network development and therefore increase time and cost
efficiency, offering more convenient way to manage energy
distribution. These aspects confer to the AMPERE project
a worldwide dimension, having European industry the clear
role to bring innovation and know-how to allow network
intervention planning with a limited afforded financial risk
above all for emerging non-European countries.
AMPERE proposes a solution based on a GIS Cloud mapping
technology, collecting on field data acquired with
optical/thermal cameras and LIDAR installed on board a
Remote Piloted Aircraft (RPA). In particular, an RPA will
be able to fly over selected areas performing semi-automated
operations to collect optical and thermal images as well
as 3D LiDAR-based reconstruction products. Such products
are post processed at the central cloud GIS platform
allowing operators in planning and monitoring activities
by means of visualization and analytics tools can resolve
data accessibility issues and improve the decision-making
process. On this context, EGNSS represents an essential
technology ensuring automated operations in a reliable
manner and guaranteeing high performance.
AMPERE use Galileo advanced features -namely, High
Accuracy Service (HAS) and E5 AltBOC- as a core element
of the added-value asset mapping proposition. The nature
of HAS is fitting very well the requirements of this application,
especially due to the re-shaping of the once fee-based
accuracy capability to an open, free-of-charge service delivering
around 20 centimeter accuracy, versus the belowten-centimeters
PPP services, in lower convergence time.
The key of Galileo HAS stands upon the high bandwidth of
its E6-B channel, well suited to transmit PPP information,
especially relevant for satellite clock corrections, which are
not as stable in the medium and long term as the orbits.
Additionally, the use of E5 AltBOC pseudo-ranges (which
are cm-level precise with maximum multipath effects in the
order of 1 m) supports fast ambiguity resolution for carrier
phase observations. The Alternative BOC, (AltBOC)
modulation on E5, is one of the most advanced signals the
Galileo satellites transmit. Galileo receivers capable of tracking
this signal will benefit from unequalled performance
in terms of measurement accuracy and multipath suppression.
The market is responding actively and positively to multi-frequency
enhanced capabilities provided by Galileo.
Around 40% of receiver models on the market are now
multi-frequency. Also, in the mass market, with the launch
of the world’s first dual-frequency GNSS smartphone by
Xiaomi, and u-blox, STM, Intel and Qualcomm launching
their first dual-frequency products earlier this year, multifrequency
is becoming a reality for user needing increased
accuracy.
More information on: https://h2020-ampere.eu
36 GEOmedia n°3-2022

NEWS
RHETICUS® NETWORK ALERT TO ASSIST
UNITED UTILITIES OF ENGLAND
CHC Navigation (CHCNAV) today announced the availability
of the i73+ Pocket GNSS receiver. The i73+ is a compact,
powerful and versatile GNSS receiver with an integrated UHF
modem which can be used indifferently as a base station or
rover. Powered by 624 full GNSS channels and the latest iStar
technology, it delivers survey-grade accuracy in all job site configurations.
"Building on the legacy of the i73 GNSS, the new i73+ receiver
is designed to maintain its proven compact and lightweight
concept, but additionally adds the ability to be operated
as either a GNSS RTK base station or a rover.” said Rachel
Wang, Product Manager of CHC Navigation's Surveying and
Engineering Division. “To enable this extra feature, we have
built in the latest UHF modem technology allowing the reception
and transmission of RTK corrections without sacrificing
receiver size and power consumption.”
Integrated Tx/Rx UHF modem extends the i73+ capacity
The i73+ has a built-in transceiver radio module compatible
with major radio protocols, making it a perfect portable builtin
UHF base and rover kit with fewer accessories. The i73+ is
a highly productive NTRIP rover when used with a handheld
controller or tablet and connected to a GNSS RTK network
via CHCNAV LandStar field software.
Best-in-class technology with 624-channels advanced tracking
The integrated advanced 624-channel GNSS technology takes
advantage of GPS, Glonass, Galileo
and BeiDou, in particular the latest
BeiDou III signal, and provides robust
data quality at all times. The
i73+ extends GNSS surveying capabilities
while maintaining centimeterlevel
survey-grade accuracy.
Built-in IMU technology highly
enhances surveyors’ work efficiency
With its IMU compensation ready
in 3 seconds, the i73+ delivers 3 cm
accuracy at up to 30 degrees pole tilt, increasing point measurement
efficiency by 20% and stakeout by 30%. Surveyors are
able to extend their working boundary near trees, walls, and
buildings without the use of a total station or offset measurement
tools.
Compact design, only 0.73kg including battery
The i73+ is the lightest and smallest receiver in its class, weighing
only 0.73 kg including battery. It is almost 40% lighter
than traditional GNSS receivers and easy to carry, use and
operate without fatigue. The i73+ is packed with advanced
technology, fits in hands and offers maximum productivity for
GNSS surveys.
Learn more about i73+: https://chcnav.com/product-detail/
i73+-imu+-+rtk-gnss
BRINGING REAL LIFE LO-
CATION BASED DATA TO
THE METAVERSE WITH
METAGEO
METAGEO is an easy to use map (GIS)
platform that brings imagery, maps,
Digital Twins, and sensor data into one
3D universe, and then streams to any
internet enabled device, or metaverse
platform. The new GIS platform aim to
enable organizations of all sizes to host,
analyze, find and share 3D map datasets
between any internet-capable device.
The platform processes any locationbased
map or sensor data from the real
world, combines it into a single 3D virtual
environment and streams it to any
device or Metaverse platform.Today’s
traditional GIS platforms are expensive,
primarily offer 2D mapping features, are
highly complicated and often require an
advanced degree to master. 3D map and
scan datasets are large, expensive and
often hidden. Furthermore, these large
files are often unsuitable for viewing
on mobile devices or rendering in AR/
VR environments. METAGEO addresses
these issues with an affordable and
easy-to-use platform that can load data
from multiple sources. These sources include
satellites, drones, mobile devices,
public and crowdsourced repositories,
IoT sensor data, 3D models and topographic
maps. The data is then processed
by the METAGEO platform into a
3D world and streamed to any internetconnected
device, enabling live collaboration
between the office and field via
mobile or AR device. Key innovations
in the METAGEO 3D map platform
include:Fast and intuitive multi-user interface
for easy data sharing and collaborationAggregation
of map and locationbased
data from a multitude of sources
on a global scaleSeamlessly import and
sync data from multiple different systems
into a single platformEasily host
and stream large datasets between internet-connected
devicesProvide ability to
find open source and private dataPlugin
SDK will allow for 3rd party tools to
scale and fit any user needsMETAGEO
has been designed for a wide range of applications
in academia, architecture, engineering,
construction, energy, natural
resource management, environmental
monitoring, utilities, and public safety,
among others. The platform uses include
planning and managing construction
sites, organizing the layouts of events,
maps for disaster management, public
safety, visualizing inspection imagery
from drones and mobile devices, and
much more.“After working with 3D
map data for several years, it became apparent
that there was no easy way to share
big datasets with those who need the
information most, those with the boots
on the ground,”said Paul Spaur, Founder
of METAGEO. “Now with the rapid
advancement of mobile hardware, and
using advanced processing techniques,
we can now leverage this data in real life,
and in the metaverse.” METAGEO will
be offered in several affordable subscription
tiers, including Free Single User,
Free Educational, Standard, Commercial
and Enterprise. Each tier provides added
features and
benefits, enabling organizations to scale.
METAGEO is available to a limited
number of beta subscribers.Interested
parties can get started today at www.metageo.iowww.geoforall.it/kcax
GEOmedia n°3-2022 37

NEWS
LEICA GEOSYSTEMS ANNOUNCES
MAJOR PERFORMANCE INCREASE IN
AIRBORNE BATHYMETRIC SURVEY
Leica Geosystems, part of Hexagon, announced today
the introduction of Leica Chiroptera-5, the new highperformance
airborne bathymetric LiDAR sensor for
coastal and inland water surveys. This latest mapping
technology increases the depth penetration, point density
and topographic sensitivity of the sensor compared
to previous generations. The new system delivers highresolution
LiDAR data supporting numerous applications
such as nautical charting, coastal infrastructure
planning, environmental monitoring as well as landslide
and erosion risk assessments.
Higher sensor performance enables
more cost-effective surveys
Chiroptera-5 combines airborne bathymetric and topographic
LiDAR sensors with a 4-band camera to collect
seamless data from the seabed to land. Thanks to higher
pulse repetition frequency (PRF), the new technology
increases point density by 40% compared to the previous
generation system, collecting more data during
every survey flight. Improved electronics and optics
increase water depth penetration by 20% and double
the topographic sensitivity to capture larger areas of
submerged terrain and objects with greater detail. The
high-performance sensor is designed to fit a stabilising
mount, enabling more efficient area coverage which decreases
operational costs and carbon footprint of mapping
projects.
Leica Geosystems’ signature bathymetric workflow supports
the sensor’s performance. Introducing near realtime
data processing enables coverage analysis immediately
after landing, allowing operators to quality control
the data quickly before demobilising the system.
The Leica LiDAR Survey Studio (LSS) processing suite
provides full waveform analysis and offers automatic calibration,
refraction correction and data classification,
as well as advanced turbid water enhancement.
Expanding bathymetric application portfolio to support
environmental research
Combining superior resolution, depth penetration and
topographic sensitivity, Chiroptera-5 provides substantial
benefits for various environmental applications like
shoreline erosion monitoring, flood simulation and
prevention and benthic habitat classification.
Bundled with the FAAS/EASA certified helicopter pod,
the system enables advanced terrain-following flying
paths for efficient river mapping and complex coastlines
surveys. Owners of previous generation systems are
offered an easy upgrade path to Chiroptera-5 to add
capabilities to their existing sensor and leverage their
initial investment.
“The first generation Chiroptera airborne sensor was
flown in 2012. During its ten years of operation, the
system has seen constant evolution that continuously
improved the productivity and efficiency of the entire
bathymetric surveying industry,” says Anders Ekelund,
Vice President of Airborne Bathymetry at Hexagon.
“By collecting detailed data of coastal areas and inland
waters, Chiroptera-5 provides an invaluable source of
information that supports better decision making, especially
for environmental monitoring and management,
in line with Hexagon’s commitment to a more sustainable
future.”
For more information please visit: http://leica-geosystems.com/chiroptera-5
38 GEOmedia n°3-2022

NEWS
GEOmedia n°3-2022 39

AEROFOTOTECA
POTATOES, ARTIFICIAL INTELLIGENCE AND
L’Aerofototeca
Nazionale
racconta…
OTHER AMENITIES: PLAYING WITH COLORS ON
PANCHROMATIC AERIAL PHOTOGRAPHS
by Gianluca Cantoro
Towards color photography
During the first half of
19th century, photography
stimulated people’s imagination
and wonder. Despite the
impressive quality of the first
trials, photographs lacked the
realism provided by natural
colors, at times added in postproduction
–as we would
say today– by specialized
painters (Coleman, 1897, p.
56) who felt threatened by the
emergence of photography.
Following Rintoul, “when the
photographer has succeeded
in obtaining a good likeness,
it passes into the artist’s hands,
who, with skill and color, give
to it a life-like and natural
appearance” (Rintoul, 1872, p.
XIII–XIV).
A French physicist, Louis
Ducos du Hauron, announced
a method for creating color
photographs by combining
colored pigments instead of
light, as suggested by Maxwell’s
demonstration of 1861. His
process required long exposure
times, and this problem
built on top of the absence
of photographic materials
sensitive to the whole range
of the color spectrum. Other
inventors and scientists tried
to solve the challenge of color
photographs, but all trials were
quite expensive and needed
specific equipment and complex
procedures.
Fig. 1 - Example of pan-sharpening between a satellite image and an historical panchromatic
photograph (top and bottom left). In the column to the right, three different algorithms, respectively
(from top to bottom) Brovey, IHS and PCA.
The first patent of color
photograph, combining both
screen and emulsion on the
same glass support under
the name Autochrome, was
registered by Auguste and
Louis Lumière in 1895, the
same year of their invention
of the Cinématographe. The
manufacturing of autochrome
plates was a complex process,
starting with the sieving
of potato starch (to isolate
individual grains between 10–
15 microns in diameter), whose
grains were then dyed red, green
and blue-violet, mixed and
spread over a glass plate (around
four million transparent starch
grains on every square inch of
it), and coated with a sticky
varnish. Next, charcoal powder
was spread over the plate to fill
any gaps between the colored
starch grains.
Autochrome plates were simple
to use, they required no special
apparatus and photographers
were able to use their existing
cameras. Exposure times,
however, were long –about 30
times those of conventional
plates. Nevertheless, by 1913,
the Lumière factory in Lyon was
40 GEOmedia n°3-2022

AEROFOTOTECA
producing 6,000 autochrome
plates every day. This testifies
of the appeal of color
photographs already in those
early times.
Is it possible today to convert
native black and white images
(raster digital pictures, not
prints or negative anymore)?
And why should one take
the trouble to convert
panchromatic into color
images after all? This paper
presents some experiments
to colorize historical
photographs, in the effort
to boost our capabilities to
undisclose information from
frozen moments captured
by cameras and to –ideally–
promote further the use of
aerial images in various fields.
Colors in Remote Sensing
Some procedures in remote
sensing are known and
frequently applied to
satellite images, to improve
the resolution of a color
image with the details of its
panchromatic twin. This
fusion procedure, known with
the term pan-sharpening,
can be applied to satellite
imagery through numerous
algorithms, and it produces
a sensible increase in the
accuracy of photo-analysis
and derived feature extraction,
modeling and classification
(Yang et al., 2012). The most
commonly used algorithms
include IHS (Intensity, Hue
and Saturation) (Schetselaar,
1998), PCA (Principal
Component Analysis) (Chavez
et al., 1990), the Gram-
Schmidt Spectral Sharpening
(Laben Craig and Brower
Bernard, 2000) and the
Weighted Brovey transform
(Chavez et al., 1990).
The various pan-sharpening
techniques have two main
factors in common: 1) they are
normally applied to satellite
images, namely multispectral
and panchromatic bands; 2)
the two datasets to be fused
Fig. 2 - Example of visual trick for image colorization inspired by the Color-Assimilation-Grid-Illusion. Top-
Left: historical vertical image of Ostia of 1985 precisely georeferenced over the bottom satellite image. Bottomleft:
Landsat/Copernicus satellite image of the same area of 2019. Top-Right: historical panchromatic with
over-saturated color grid extracted from the available satellite image. Bottom-Right: Detail of the image above
to show a close-up look at the colored grid and the black and white background.
Fig. 3 - Application of automatic colorization algorithms (Deoldify, Algorithmia and Automatic Colorizer) to
three oblique photographs (Original Image). Photographs by Otto Braasch (Musson et al., 2005, fig. 10.8, 10.9,
10.7) edited for the proposed approach.
should have been captured
(almost) simultaneously. For
these reasons it is apparently
not possible to fuse an historical
aerial image with a satellite
image, which is what we are
going to try here. Proposed
methods are not conventional
and may therefore attract
comprehensible skepticism, but
they should be intended as a
proof-of-concept or experiments
GEOmedia n°3-2022 41

AEROFOTOTECA
Fig. 4 - Interactive Deep Colorization User Interface. After clicking on a specific point on the
black and white image (see colored spots on left image in the interface), the user can assign a
color from the “ab Color Gamut”, the “Suggested colors” or the “Recently used colors”. Results
are presented in real time to the right of the UI and can be saved at any time.
Fig. 5 - Comparison of processing of the same pictures as for Fig. 3 with Interactive Deep Colorization
(or iColor).
to test the capabilities of
modern computer approach to
obtain a realistic representation
of past environments in natural
colors for the benefits of photoreaders.
For example, since our objective
is mainly to get a colorized
historical image, we can adjust
reciprocal resolution of our
vertical and satellite images of
exactly the same area. A similar
approach has been explored
recently (Siok and Ewiak,
2020) with aerial and satellite
images of about the same
period and without dramatic
changes in cultivations or plot
sizes. Indeed, the processing
of areas that changed across
time (i.e. between the date of
the historical photograph and
the date of the chosen satellite
image in terms of time of the
day, season or cultivations/
urbanization processes) may
produce some unpleasant
artifact (see in Fig. 1 the details
of the trees and bushes colors
which are larger than in the
historical image), but in this
case a targeted editing with
computer graphic software
may minimize the problem, if
needed.
Once such high-resolution
color image is being generated,
the applicability of multiple
operations can be explored,
such as image classification and
feature extraction.
Another approach, completely
different in terms of processing
and output, is the use of
an operation called Color-
Assimilation-Grid-Illusion
(Kolås, 2019). As the name
suggests, this approach is a mere
visual trick and it is presented
here mainly for dissemination
purposes (not for an improved
photo-interpretation) and as
a sort of mild invasive way to
add colors to black and white
images. It consists in overlaying
grids (or lines or dots) of oversaturated
colors over black
and white images; our brain
essentially fills in the missing
colors that it would anticipate,
or expect, to be there in a full
color image.
The image processing is
intended to be used on one
single color image, that is
converted to grayscale and
overlaid with the original colors
only through a grid. Instead, in
our case, starting from the same
inspiring principle, we take
the historical aerial photograph
that we want to colorize, and a
satellite image (ideally of about
the same season) of the same
42 GEOmedia n°3-2022

AEROFOTOTECA
Fig. 6 - Possible variants generated with iColor of the same image with deliberate selection of false colors to make specific features more visible or for
other applications.
area; we generate the color grid
from the satellite image, oversaturate
it and overlay it over the
panchromatic picture (Fig. 2).
Calibration of the above
experiment may be needed
according to the printing or
screen zoom size, nevertheless,
it is an easy effect to colorize
historical photographs.
The Machine Learning
automatic and
semi-automatic approach
Machine learning has been
explored in several fields for its
ability to “learn” the intended
process from A to B with
the help of prepared dataset.
Algorithms are available to
convert black and white images
into color one, based exactly on
a learning dataset. In this sense
it could be applied to historical
aerial images as well, but again,
here the intent is to generate
an image with colors that are
plausible, not to produce an
accurate representation of the
actual snapshot in time.
Below (Fig. 3) are some
examples of oblique
photographs originally taken
with digital camera in color,
then converted to black and
white for the sake of the
experiment, and colorized back
with three automatic algorithms
for image processing: Deoldify
(Antic, 2021) (or DeepAI),
Algorithmia (Zhang et al.,
2016) and Automatic Colorizer
(Larsson et al., 2017).
The chosen algorithms, selected
for their simplicity of use, for
their advertised capabilities and
for their availability as opensource
code or online demo,
are examples of a computer
problem called image-to-image
translation, whose success
depends by the provision of
sufficient (and compatible)
training data (Tripathy et al.,
2018). Since the training data
is mostly made of ground
photographs of natural subjects,
portraits or architectures, the
obtained result in our case is
mostly unsatisfactory, especially
when looking at the original
(our “ground truth”) but also
if we consider the generated
images on their own for
photointerpretation.
Different result is instead
achievable with another
algorithm of the same family,
which has an interactive model
that allows user to manually
input colors on black and white
image based on chrominance
gamut: it is the case of
Interactive Deep Colorization
(or iColor) (Zhang et al., 2017)
(Fig.4).
By default, the first proposed
colorized image in this
algorithm is very much similar
to the ones generated by similar
algorithms (see Fig. 3), but once
specific colors are selected, the
result improves considerably
reaching a good proximity to
the original images of our test
cases (Fig. 5).
If from one side the Interactive
Deep Colorization algorithm
allows one to create images with
plausible colors, possibly similar
to the originally depicted
subject, it also provides the
option to deliberately choose
“wrong” colors, and somehow
create a completely unreal
scenario (Fig. 6), which may
make sense if they are employed
in our case to highlight specific
features or shadows.
Conclusions
In modern photointerpretation,
crop-marks
– as well as weed-marks,
germination-marks and grassmarks
– by definition are
made of vegetational stress or
differential growth in green
fields. Even with soil-marks,
shades of brown help us
recognizing patterns in arable
lands.
Seeing black-and-white images
in color has the potential to
brings certain details to life that
would otherwise be missed or
hardly be visible. This sense
of immediacy is why color
images feel more relatable.
Historical vertical photographs
traditionally served (and still
serve) immensely for the study
of landscape changes and the
identification of archaeological
traces (among others) for
the reconstruction of topic
palimpsests. They often provide
details that have no equals in
color images so far and, training
on photo-interpretation black
and white images cannot be
ignored or replaced in any way.
In the paper an effort is
presented to push the
boundaries of consolidated
GEOmedia n°3-2022 43

AEROFOTOTECA
practice in remote sensing
and artificial intelligence,
together with the attempt of
presenting a visual trick for
dissemination purposes. The
proposed methods change
the current paradigm with
respect to employed algorithms
and dataset to which the
algorithms are applied, aiming
at a new way of looking at
historical aerial photographs
and ideally unveiling a new
dimension in past-landscape
studies and dissemination.
The various procedures are all
oriented towards the artificial
colorization of historical aerial
photographs, natively black
and white. These “bizarre”
trials are intended as ways to
promote new approaches to
legacy data, with the ultimate
goal to simplify or enhance
aerial photo-interpretation
and involve non-experts in
the narration of the past
made through photographic
documents.
Lastly, artists dealing with
historical image colorization
admit the intense and timeconsuming
effort required to
achieved a realistic result and
a philological reconstruction,
involving historical research,
comparative materials and
interviews with witnesses
or experts. Therefore, black
and white colorization may
be a creative process that can
increase focus and attention on
what we see (or we don’t see) in
historical aerial images.
REFERENCES
Antic, J., 2021. DeOldify [WWW Document]. Deoldify, a deep learning based project
for coloring and restoring old images. URL https://github.com/jantic/DeOldify (accessed
1.15.21).
Chavez, P.S., Jr, Sides, S.C., Anderson, J.A., 1990. Comparison of three different methods
to merge multiresolution and multispectral data: LANDSAT TM and SPOT panchromatic.
AAPG Bulletin (American Association of Petroleum Geologists); (USA) 74:6.
Coleman, F.M., 1897. Typical Pictures of Indian Natives, Being Reproductions from
Specially Prepared Hand-coloured Photographs with Descriptive Letterpress. “Times of
India” office, and Thacker & Company, Limited.
Kolås, Ø., 2019. Color Assimilation Grid Illusion [WWW Document]. Color Assimilation
Grid Illusion. URL https://www.patreon.com/posts/color-grid-28734535 (accessed
2.2.21).
Laben Craig, Brower Bernard 2000: A. Laben Craig, V. Brower Bernard, Process For Enhancing
The Spatial Resolution Of Multispectral Imagery Using Pan-sharpening - (US
Patent: US 6011875 A), https://lens.org/135-660-046-023-136.
Larsson, G., Maire, M., Shakhnarovich, G., 2017. Learning Representations for Automatic
Colorization. arXiv:1603.06668 [cs].
Musson, C., Palmer, R., Campana, S., 2005. In volo nel passato: aerofotografia e cartografia
archeologica, Biblioteca del Dipartimento di archeologia e storia delle arti, Sezione
archeologica, Università di Siena. All’insegna del giglio, Florence?
Rintoul, A.N., 1872. A guide to painting photographic portraits, draperies, backgrounds,
&c. in water colours : with concise instructions for tinting paper, glass, & daguerreotype
pictures and for painting photographs in oil colours and photo-chromography.
Schetselaar, E.M., 1998. Fusion by the IHS transform: Should we use cylindrical or
spherical coordinates? International Journal of Remote Sensing 19, 759–765. https://doi.
org/10.1080/014311698215982
Siok, K., Ewiak, I., 2020. The simulation approach to the interpretation of archival aerial
photographs. Open Geosciences 12, 1–10. https://doi.org/10.1515/geo-2020-0001
Tripathy, S., Kannala, J., Rahtu, E., 2018. Learning image-to-image translation using
paired and unpaired training samples. arXiv:1805.03189 [cs].
Yang, S., Wang, M., Jiao, L., 2012. Fusion of multispectral and panchromatic images
based on support value transform and adaptive principal component analysis. Information
Fusion 13, 177–184. https://doi.org/10.1016/j.inffus.2010.09.003
Zhang, R., Isola, P., Efros, A.A., 2016. Colorful Image Colorization. arXiv:1603.08511
[cs].
Zhang, R., Zhu, J.-Y., Isola, P., Geng, X., Lin, A.S., Yu, T., Efros, A.A., 2017. Rea
KEYWORDS
Color photography; artificial intelligence; image-to-image; remote sensing;
air-photo interpretation;
ABSTRACT
Historical photographs, whether taken from the air or from the ground, are usually synonyms
of grayscale or sepia prints. From the very beginning of photography, during the first
half of 19th century, people were amazed by this new media that could record all aspects
of a scene with great detail. Soon though, everybody started wondering why would such
an impressive innovation fail to record colors? A process of trials and errors then started
(including the most successful and pioneer one, involving the use of potato starch, by Lumière
brothers) aiming to add colors to photographs, till the consolidation of new systems
(camera and film) capable to collect photographs directly in color. In the past, before and
during this innovative approach, native black and white photographs were painted in the
effort to give them life. Today, only few methods are available to convert a panchromatic
image into a color one, and they need a number of steps and further development to work
properly. The paper tries to present different methods to colorize native black and white
photographs, based on available automatic or interactive Artificial Intelligence (Machine
Learning or Deep Learning) algorithms, on revised remote sensing procedures and on
visual tricks, aiming at exploring the possible improvement in readability and interpretation
of photographed contests in the usual analytic process of photo-interpretation. At the
same time, colorized historical photographs hold different appeal in the general public
and have the potential to attract and involve non-experts in the archaeological/historical
reconstruction phases.
AUTHOR
Gianluca Cantoro
gianluca.cantoro@cnr.it
Institute of Heritage Science (ISPC) – Italian National Research Council (CNR)
Area della Ricerca di Roma 1, Via Salaria km 29,300 - 00010 Montelibretti (RM)
Italian National AirPhoto Archive (Aerofototeca Nazionale, AFN) – Istituto Centrale
per il Catalogo e la Documentazione (ICCD)
Via di San Michele 18, 00153 Rome (RM)
44 GEOmedia n°3-2022

AMPERE
a a GNSS-based integrated platform
for for energy decision makers
AMPERE Working Group in Santo Domingo
AMPERE PARTNERS
Barrio Los Tres Brazos
Santo Domingo Este
Asset Mapping Platform Platform for
Emerging CountRies Electrification
Emerging CountRies Electrification
Despite global electrification rates are significantly progressing, the
Despite access global to electricity electrification in emerging rates countries are significantly is still far from progressing, being achieved. the access
Indeed, the challenge facing such communities goes beyond the lack of
to electricity infrastructure emerging assets; what countries is needed is is still a holistic far from assessment being achieved. of the
energy demand and its expected growth over time, based on an accurate
Indeed, assessment the challenge of deployed facing resources such communities and their maintenance goes beyond status. the lack of
infrastructure assets; what is needed is a holistic assessment of the energy
demand and its expected growth over time, based on an accurate
assessment of deployed resources and their maintenance status.
AMPERE Consortium
www.h2020-ampere.eu
AMPERE project has received funding from the European GNSS Agency (grant
agreement No 870227) under the European Union’s Horizon 2020 research and
innovation programme.

EVENTS
22-28 Agosto 2022
FOSS4G 2022 Academic
Track
Firenze (Italy)
http://www.geoforall.it/
kcp3h
11-15 September 2022
FIG Congress 2022
Warsaw (Poland)
http://www.geoforall.it/
kc3kd
17-20 October 2022
IAG International
Symposium on Reference
Frames for Applications in
Geosciences (REFAG 2022)
Thessaloniki (Greece)
http://www.geoforall.it/
kc3kr
18-20 October 2022
INTERGEO 2022
Essen (Germany)
http://www.geoforall.it/
kc3kh
20-21 October 2022
SAR Analytics Symposium
Berlin (Germany)
http://www.geoforall.it/
kc3f8
Cartographic Conference
Firenze www.geoforall.it/
kyk8k
NEW LEICA BLK360
The all-new Leica BLK360 breaks open
the possibilities of reality capture.
With unprecedented, best-in-class scanning speed,
the BLK360 makes you faster.
◗ A supercharged next-gen imaging laser scanner, it captures
a full scan with spherical images in only twenty seconds over
five times faster than the BLK360 G1.
◗ With fast and agile in-field workflows, along with live feedback
on your mobile device, you can be absolutely sure you’ve captured
everything you need. VIS technology automatically combines
your scans to speed up your workflow and help you make sure
your datasets are complete.
◗ BLK360 data is highly valuable for so many uses - from AEC to VR.
Easily transfer and work with data in your software ecosystem
to create immersive and highly accurate deliverables.
◗ You can also upload BLK360 data to HxDR, the Hexagon cloud-based
data storage, visualization and collaboration platform.
to know
more
Contact us to know more!
Via A. Romilli, 20/8 - 20139 Milano • Tel. 02 5398739
E-mail: teorema@geomatica.it
www.geomatica.it • www.disto.it • www.termocamere.com

EASY. RELIABLE. STONEX.
SURVEYING
3D SCANNING
SOFTWARE
www.stonex.it