Acquired Population Features Of Cotton Image Based On Digital Image Processing

Digital image processing technology has been applied as a new method to many traditional subjects. In the field of crop science, research on population image based on Digital image processing technology has greatly developed. However, the work already done concentrated on population image of close planting crop, which was planted on homogenizing distribution, and individual crop. Population image of crop like cotton has not researched so far. This article introduced the technology of image processing and recognition into cotton image.Several population features have been acquired, such as vegetation coverage, regularity degree, height of plant, configuration of plant. Special arithmetic has been developed to quantify vegetation coverage and regularity degree of cotton images. Cotton image is segmented in order to acquire vegetation coverage. Vegetation coverage has been acquired in a much flexible way. We can select any polygonal area on cotton image by using mouse clicking vertexes of the polygonal area in turn. Segmentation arithmetic has been studied and effective arithmetic is chosen. Acquirement of regularity degree is based on managing vegetation coverage. One cotton image is divided uniformly into 16 areas. Work out vegetation coverage of these 16 areas one by one and then choose an arithmetic applying to those data. Sophisticated arithmetic has been developed to avoid negative effect caused by variety of vegetation coverage. Measuring distance is as convenient as acquiring vegetation coverage on polygonal area. Click proper point several times on the cotton image and you get the measurement figured out by software automatically. In comparison with manual measurements, the accuracy rates of measurements by software are as high as 99.1%, 94% and 95.32%. The results show that technology based on digital image processing technology can work successfully in the field of crop science. Population features have been acquired and analyzed in an efficient, rapid and nondestructive way, contrasted with traditional manual methods.In addition to population features, several image segmentation methods have been developed to acquire features of single cotton leaf and to analyze cotton leaf disease.