A Pattern Recognition System About Maize Leaf Disease Based On Computer Image Processing

Crop disease,insect pest and weed affect the output and quality of crop badly.the papar mainly researched the theories and methods of recognition and diagnosis of main maize disease during maize growing by using computer vision technology,aimed at the problem of low precision and untimely diagnosis about crop disease.The study took common maize leaf diseases as study object and brought forward feasible method to improve diagnosis precision.Thereby it can provide theory support to the correlative research of the auto recognition and diagnosis of crop disease,insect pest and weed.According to the characteristics of maize leaf disease,Threshold method,the method of enhance contrast using the way of frequency and method of YCbCr color space have benn used to segmentate diseases image.The Redundant dots are removing by the Algorithm of open and close.Comparing the similarity during the images by using the algorithm of Parallel simulated annealing which can Optimizate the Neural Networks.it can improve the algorithm of comparing by distabce,and be used for classification recognition during Gray leaf spot disease, Brownleaf spot disease and big leaf spot disease.To extract texture feature based-on the lifting wavelet and the total space gray matrix. utilizing wavelets'perfect performance in spatial-frequency localization and the lifting wavelets'in-place operation. It computers the entropy values of the sub-frequency images disposed by the two-layers'lifting wavelet transform, and has merits of low feature dimensions,low computationl cost and good capability.Transforming RGB color space into HIS color for extracting color features.The study developed Software for maize disease recobnition by using of Visual C++.Three common kinds of maize leaf disease can be recognized and the recognition rate reach to ninty percent.It can provide soft and technology support for developing crop disease intelligent recognition system.The research applied computer vision technology in maize disease recognition,which extends the application area of machine vision and also provide a use for reference in the field of agriculture technology for machine vision.