13.07.2015 Views

FACIAL SOFT BIOMETRICS - Library of Ph.D. Theses | EURASIP

FACIAL SOFT BIOMETRICS - Library of Ph.D. Theses | EURASIP

FACIAL SOFT BIOMETRICS - Library of Ph.D. Theses | EURASIP

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

94 8. USER ACCEPTANCE STUDY RELATING TO <strong>SOFT</strong> <strong>BIOMETRICS</strong>8.1.1 Testing MethodologyTest Setup : A set <strong>of</strong> 15 users (5 female, 10 male) <strong>of</strong> different nationality and ethnicity between26 and 37 years old was randomly selected from an <strong>of</strong>fice complex with the condition to notwork on biometrics. The participants were not paid. The test was consistent with the ITU-Trecommendation [usa00], that is to say we followed methods for interactive user tests <strong>of</strong> setting,equipment and environment, as well as subjects training and solicitation <strong>of</strong> opinions. We usedrating methods according to the absolute category rating (ACR). Specifically for the ratings <strong>of</strong> thefour access systems we presented each one at a time and let the user rate them independently. Wedenote this rating as MOS (mean opinion score), which spans on a five grade scale, five being"excellent" and one being "very poor". The user study took place in a computer laboratory, withsimilar conditions to an <strong>of</strong>fice. The duration per test was about half an hour. We performed aWizard <strong>of</strong> Oz study, see [Nie93], specifically the employed interfaces were functional, howeverthe acquired data was not processed. Hence processing time evaluation is not part <strong>of</strong> this study,neither the verification accuracy <strong>of</strong> the presented methods. The employed laptop was a DELLE4310. The documentation <strong>of</strong> the study contains the notes <strong>of</strong> questionnaires and related observernotes.Figure 8.1: Interfaces <strong>of</strong> the s<strong>of</strong>t biometrics, face, PIN and fingerprint based access methods.Procedure : The four access methods, see Figure 8.1, were presented and demonstrated by theobserver <strong>of</strong> the study. Additional information on the methods, as <strong>of</strong> the differences between themethods, was provided. Subsequently participants <strong>of</strong> the user study freely explored the availablesystems. In the next step users were asked to log in with each <strong>of</strong> the systems. Subsequentlyan interview about the user experience was conducted. Here users were asked to absolutely rate(from 1 to 5, 5 being excellent) and comment on different aspects (easiness, clarity, comfort andspeed <strong>of</strong> the methods). Then users were confronted with two scenarios, where in the first scenariothe user accesses his/ her personal computer; whereas in the second scenario the user acquiresthe right <strong>of</strong> entry for a lab in a crowded corridor. The suitability <strong>of</strong> the methods was enquiredfor both scenarios. Additionally to the absolute ratings the users were asked to rank the methodsin terms <strong>of</strong> speed, easiness, privacy preservation and overall satisfaction. Finally the users wereinvited to select freely one method to log in with a task to read a file. This preference was notedas spontaneous and practical preference. In the following we give details on the different accessmethods.8.1.2 Access methodsWe selected four substantially different accessing methods, in terms <strong>of</strong> both, interface andtechnology, namely s<strong>of</strong>t biometrics, face, PIN and fingerprint based access. In consent <strong>of</strong> recommendationsfor user friendly human computer interfaces (see [BH90], [CG05] and [Nie93]) wedeveloped four access systems, see Figure 8.1. All four interfaces were designed to be similar interms <strong>of</strong> structure and processing time in order to place emphasis on the four interaction elements:

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!