RGB color channel variation based segmentation of crop leaf lesion

Accurate segmentation of crop leaf lesion is the precondition of precise spraying. A method of lesion segmentation is proposed for image recognition of the precise spraying system in this paper. The method is operated in RGB color space, to separate the normal area and the lesion by intensity change of three color components. The experimental results show that the proposed method achieves a high level of accuracy and real-time performance in terms of the segmentation of brownish red lesion and yellow lesion. Therefore, this method can obtain fairly accurate leaf lesion image.