Pear tree leaves based on image analysis Research on disease identification methods

Pear trees are susceptible to black spot, brown spot, rust, black star, ring rot and other diseases in the planting process, resulting in large-scale pear fruit production and pear fruit quality decline. In this study, the identification and classification of pear brown spot and rust were realized based on image analysis combined with SVM algorithm, and the disease recognition model was established by MATLAB software, requiring the identification rate of pear diseases and pests to reach more than 90%. The main research contents and conclusions are as follows: (1) disease image preprocessing. The color image is transformed into gray image, image denoising, histogram equalization and other operations are carried out to enhance the image. In the aspect of image segmentation, HSV color space and RGB color space are analyzed and compared, and based on HSV color space, the threshold method is used to segment the color image spots. (2) Image feature value extraction. The color feature, texture feature and shape feature of the image are introduced and analyzed, and the texture feature is extracted by LBP operator. (3) Disease recognition model. Based on the principle of neural network, the disease recognition model is constructed with the GUI interface of MATLAB. The image to be tested is uploaded to the model and the result of disease identification is output. The disease recognition rate is above 95%.