Key points are not available for this paper at this time.
Fast and accurate segmentation for color images are important and difficult in computer vision and image analysis. To improve the quality of image recognition, take into account pixel color information and adjacency relationship between pixels based on spatial information, integrate color histogram-based method and region growing and merging-based segmentation method, proposed a new algorithm for color image segmentation. First, select an appropriate color space; followed by saturation threshold segmentation to extract nucleus from white blood cell; after that, according to certain rules, using region growing of color information to extract the cytoplasm; Finally, do region merging. Compared to similar methods, the segmentation results show that this method is faster and more accurate.
Duan et al. (Sat,) studied this question.