摘要： 基于轮廓线模型的目标分割、物体跟踪方法是近十几年来图像和视频领域研究的热点。因为轮廓线模型可以将待处理问题的先验知识与各种图像处理算法有效地融合在一起,所以比以往的计算视觉理论具有更强的实用性。虽然很多模型算法已经和最初的Snakes大相径庭,但它们的指导思想却是一脉相承的。该文以Snakes模型为基础,系统地介绍了轮廓线模型的发展脉络和各种最新成果。

关键词: 轮廓线模型, Snakes, 图像分割, 跟踪

Abstract: Active contour models are the hotspot in the field of computer vision. They are more practically and powerful than other computer theories because the contours could merge prior knowledge and image processing algorithms. Although the forms of these models have many differences, the essence is the one. The paper tries to introduce the whole active contour models family.

中图分类号: 

• TP391.41