关键词: 图像分割, 活动轮廓模型, 边缘停止函数, 小波变换

Abstract: The edge stopping function is very important in edge-based active contour models. It typically depends on the gradient magnitude of Gaussian smoothed image, which leads the corresponding models evolving slowly and having difficulty to segment images with strong noise and complex background. In this paper, a novel edge stopping function related to wavelet transform is proposed. Experimental results show that, based on the new edge stopping function, edge-based models can not only effectively address the problems above, but also be applied to selective detection of multiple objects.

Key words: image segmentation, active contour model, edge stopping function, wavelet transform