Gradient Color Leaf Image Segmentation Algorithm Based on Meanshift and Kmeans

摘要

In order to segment the gradient color leaf image accurately in complex background, a method of image segmentation combining meanshift algorithm and kmeans algorithm is proposed. In this paper, meanshift algorithm is used in LUV space to smooth the color image, then kmeans algorithm is used to divide and correct similar areas, and finally OTSU algorithm is used for image binary segmentation. The method uses the smoothing ability of meanshift algorithm to process the image gradient area, and uses the clustering ability of kmeans algorithm to solve the problem of under-segmentation of the gradient color region. The method can effectively separate the gradient color leaf image from different backgrounds. The results of the proposed method are better than the color index method.