Automatic segmentation of virtual human organ based on improved color similarity coefficient
【摘要】：正Along with the in-depth study of virtual human, human datasets with different territorial race have been released in many countries. As the first stage of virtual human research, studying on visualized virtual human is the focus in many countries. To obtain the actual 3-D models of human's organ, fast and exact segmentation approaches for color images are the necessary conditions. An improved segmentation approach is utilized in this paper. It is often made use of in the color object searching for vision system of soccer robot. The swift and automatic segmentation of serial color images for virtual human's organ is realized. It provided potent support to the 3-D reconstruction research of visualized virtual human.