关键词: 超声图像, 图像分割, Li模型, C-V模型, 变分水平集

Abstract: Aiming at the problem that ultrasonic image segmentation has low precision because of restriction of image principle，this paper proposes an ultrasonic image segmentation method based on variation level set，combining both boundary information and regional information.This method improves the energy function of Li model，introduces external energy function of C-V model，combines the edge gradient information and region information of image，and applies it to the ultrasonic image segmentation.The experimental results show that the new model not only inherits the advantages of the two models，but also performs better than the above-mentioned model in the speed of segmentation，accuracy，noise immunity，robustness，etc.It can extract the object ideally from ultrasonic image.

Key words: ultrasound images, image segmentation, Li model, C-V model, variational level set