The Study Of Corn And Cucumber Leaf Disease Based On Image Processing Technology

Along with increasing ability of computer process as well as rapid development of image processing and recognition technology, this technology not only is applied to agriculture more and more widely, but also is an important technology to achieve agricultural information and automatization. The [0] amount of work for gathering agricultural information is tremendous; besides the realtime and accuracy of information are paid attention to in agricultural production and scientific research. How to diagnose diseases of crops timely and accurately is always an important part of research facing to the agricultural field through computer technology. Therefor, by means of computer image processing technology as an important means, this paper researches the recognition and diagnosis of the corn and cucumber leaf diseases applying the knowledge on image processing and plant physiology synthetically.Firstly, according to sampling requirements of diseased leaves, the images of diseased samples are collected by illumination system and computer image processing equipments. No matter which kinds of equipments are used, the collected images often are not satisfying. In order to improve the image quality, the noise in the image needs to be removed by the method of median filter. Secondly, various algorithms of image segmentation and features of disease images are researched carefully and deeply. The cluster analysis is introduced to the image segmentation, and the segmentation algorithm properties of hard C-means clustering and fuzzy C-means clustering are analyzed. In this paper, disease images are segmented by fast fuzzy C-means clustering algorithm. The experiments show that this clustering algorithm can depress the cost of computation to reduce the time of segmentation obviously on the premise that clustering optimizing performance does not change. Thirdly, in this paper, chain code is used to contour tracking. Features of disease images which include perimeters, areas and shape parameters are obtained according to the principle of feature parameter extraction. The values of these features are made standardization and disease images are classified to make the disease recognition accurate. Finally, the design and implementation of this system are given. The function modules, concrete flow and main interfaces of this system are introduced.