FIGURE 3: Vegetation cover rate applicationFIGURE 5: Insect counting applicationSecond, we constructed an application that detects farm operations (Fig. 4). In thisexperiment, we used the background difference technique with cropping, subtracting andbinarizing Web application services, in that order. When the total number <strong>of</strong> white pixelscalculated by the binarizing function was greater than a threshold value, the system judgedthat a certain farm operation was performed in the image data. However, the collected imagedata were affected by various disturbances, and sometimes certain farm operations were notdetected in the judged image data when checked manually. To find image data with a farmoperation from judged image data effectively, we also developed a browsing Web applicationthat can display filtered image data as a motion picture. With this application, we can filterimage data based on not only the total number <strong>of</strong> white pixels but also any values collectedas sensor data, such as rainfall. By using the filtering function, we can reduce the time spentchecking extraneous image data. For example, there were many misjudged image datawithout a farm operation in rainy conditions. If a user knew that farming operations werenever performed in rain conditions, the browsing Web application can be altered toadditionally filter judged image data in rain conditions.Third, we constructed an application that automatically counts target insects in pheromonetraps (Fig. 5). In general, insect-counting applications use 'pattern recognition' that identifiesthe shape <strong>of</strong> the target insect, but the collected image data at a field site makes it difficult toperform such a traditional image analysis. In contrast, our image monitoring can calculate thetotal number <strong>of</strong> white pixels after processing via the background difference technique withthe cropping, subtracting and binarizing Web application services. With the combination <strong>of</strong>FIGURE 4: Farm operation detecting application
the pheromone trap to attract target insects specifically and cropping a low-noise part <strong>of</strong> theimage area to allow easy analysis, we can estimate the number <strong>of</strong> target insects in imagedata on the proposed system. The number <strong>of</strong> target insects can be estimated with highaccuracy, and this simple image analysis worked well to reduce the labor needed forcounting and to easily handle a large mass <strong>of</strong> image data (Fukatsu et al., 2012).4. DiscussionWe have proposed a practical image data monitoring system for agricultural use with FieldServers, and we tested the monitoring system with several agricultural applications toevaluate the effectiveness and potential <strong>of</strong> the system. As a practical image analysis methodfor image data collected at a field site, we proposed a simple image analysis method thatuses a combination <strong>of</strong> basic image processing techniques (cropping, subtracting andbinarizing functions). Although it is generally difficult for simple image analyses to producedetailed and accurate results, our proposed image analysis method can provide usefulinformation, and it has the potential to be effective for various applications.In the application for detecting farm operation, we used our image analysis method to filterextraneous image data with the browsing Web application. It has been important but difficultto manually check collected image data for detecting targets, because most <strong>of</strong> the imagedata were extraneous. The filtering method is effective and practical for manual checking,and it reduces user effort. In the application for counting target insects, we estimated thenumber <strong>of</strong> targets by simply applying the results <strong>of</strong> the background-difference technique tothe part <strong>of</strong> the image data that excluded the area most susceptible to disturbance. Thecounting method is also flexible and applicable because it doesn’t require special individualsettings and parameters.In conclusion, we constructed a system that can be easily used by non-expert users. Themonitoring system facilitates the use <strong>of</strong> varied applications for many users, and the simpleimage analysis method facilitates the practical use <strong>of</strong> image monitoring for various targets.The more that image monitoring comes into wide use, the more that researchers will be ableto easily apply advanced image analyses for practical use, based on our system.Reference listAkyildiz, I.F., Su, W., Sankarasubramaniam, Y., & Cayirci, E. (2002). Wireless sensornetworks: a survey. Computer Networks, 38, 393-422.Bauchi, C., & Rath, T. (2005). Prototype <strong>of</strong> a vision based system for measurements <strong>of</strong> whitefly infestation. Acta Horticulturae (ISHS), 691, 773-780.Boissard, P., Martin, V., & Moisan, S. (2008). A cognitive vision approach to early pestdetection in greenhouse crops. Computers and Electronics in Agriculture, 62, 81-93.Choi, K., Lee, G., Han, Y.J., & Bun, J.M. (1995). Tomato Maturity Evaluation Using ColorImage Analysis. Transactions <strong>of</strong> the ASAE, 38, 171-176.Fukatsu, T., & Hirafuji, M. (2005). Field Monitoring Using Sensor-Nodes with a Web Server.Journal <strong>of</strong> Robotics and Mechatronics, 17, 164-172.Fukatsu, T., Watanabe, T., Hu, H., Yoichi, H., & Hirafuji, M. (2012) Field monitoring supportsystem for the occurrence <strong>of</strong> Leptocorisa chinensis Dallas (Hemiptera: Alydidae) usingsynthetic attractants, Field Servers, and image analysis. Computers and Electronics inAgriculture, 80, 8-16.Fukatsu, T., Hirafuji, M., & Kiura, T. (2006). An Agent System for Operating Web-basedSensor Nodes via the Internet. Journal <strong>of</strong> Robotics and Mechatronics, 18, 186-194.Iwaya, K., & Yamamoto, H. (2005). Diagnosis <strong>of</strong> Optimal Harvesting Time <strong>of</strong> Rice UsingDigital Imaging, Journal <strong>of</strong> <strong>Agricultural</strong> Meteorology, 60(2005) 981-984.Tillett, R.D. (1991). Image Analysis for <strong>Agricultural</strong> Processes: a Review <strong>of</strong> PotentialOpportunities. Journal <strong>of</strong> <strong>Agricultural</strong> <strong>Engineering</strong> Research, 50, 247-258.