Abstract: The original image is depicted and sharpen by convolution filtering, and then put it in HSV model to obtain the image color extraction. In order to get the target area of the tea wheel spot,it is also used the image binaryzation to complete the refiltering.Finally,that is a picture rapid diagnosis of tea leaf spot disease. The results showed that the image recognition of tea leaf spot in leaves of tea leaves was finished. By using convolution filtering and HSV model image processing technology can be better identified the round spot of tea lesion area, especially the clear image,size of lesion area. Taking advantage of the embeddedness of Python language,it lays the foundation for the accurate spray system of tea tea wheel spot disease.

Key words: convolution filter, image processing technology, image filtering, tea disease detection, identification