Development of Image-Processing Software for Simple and High Precision Measuring of Cover-area Ratio on Water-Sensitive Paper

Yosuke Kubota, Yoshihiko Usui, Kazunobu Hayashi, Tomomichi Mizukami, Sumihiko Miyahara, Dai Osato3
and Kazuhiro Nakano

Summary

The use of pesticides is absolutely necessary for the cultivation of agricultural products in Japan. However, there is a risk that sprayed pesticides will drift towards non-target crops; a phenomenon known as drift hazard. In the present study, a simplified method for both measuring and evaluating spray drift was developed using an image-processing system. The experimental pesticide spraying was conducted using a boom sprayer, a large fan, water-sensitive paper, a diluted solution of Sumithion (MEP) emulsion, stimulant crops and glass plates. The diluted MEP emulsion was sprayed following the placement of the water-sensitive paper, stimulant crops and glass plates at arbitrary measurement points. Each material was then collected from the measurement points. Images of the collected water-sensitive paper were scanned and saved as red-green-blue (RGB) image data. Subsequently, the cover-area ratio was calculated from the RGB image data using the following methods: decomposition of RGB components, extraction processing, binary conversion by brightness threshold and arithmetic processing of pesticide attachment stain. When only R-image data were used, the cover-area ratio was calculated with a high degree of accuracy. Moreover, the measurement time of the image-processing algorithm was greatly shorted. This algorithm has obtained the objectivity and the repeatability using the image-processing and the statistical procedure. These results suggest that it is possible to both measure and evaluate spray drift using water-sensitive paper in conjunction with the image-processing algorithm developed in the present study.



Return