Development and assessment of a novel imaging software for optimizing the spray parameters on water-sensitive papers

Abstract

The overuse of pesticides is critically important for the environment because of contamination. The correct use of pesticides requires spraying optimization on the intended region. There are a lot of experimental and commercial softwares for determining the spraying parameters by using artificial targets such as water-sensitive papers. A novel image processing software was developed that could evaluate the spray coverage rates and droplet counts quickly and accurately as compared with DepositScan, ImageJ and SnapCard softwares. Water-sensitive papers were used as deposit collectors for providing a quick assessment of spray coverage. Each water-sensitive paper was assessed by comparing the three existing imaging softwares against the novel imaging software developed in Vision Acquisition Software (VAS) by National Instruments. A conveyor belt was set up for carrying the water-sensitive papers. Conveyor belt system was used for four nozzle types (AI11002, CN11002, FF11002, TTJ6011002) and three different belt speeds (4, 5 and 6 km h−1). Scanned and digitized water-sensitive paper images were evaluated with the existing software in order to determine the success of the new software.Results showed that the coverage rate results of developed imaging software had very close proximities, which were between 73.3–97.1%, 70.3–95.3% and 69.4–94.5% at belt speeds of 4, 5 and 6 km·h−1, respectively. Advanced morphological segmentation techniques used in image processing software caused more accurate performance than the other imaging programs. Droplet counts obtained by the novel image processing program were higher compared to the other image processing programs for all nozzle types and at all speeds. Detecting overlapping droplets was also an important novelty compared with the other image processing softwares as it provided both better accuracy and more data.