Local Energy Information Combined with Improved Signed Distance Regularization Term for Image Target Segmentation Algorithm

The uneven color image can not be segmented successfully with the traditional C-V model, and the C-V model is sensitive to the initial contour and the location. The existing signed distance regularization term has disadvantages, such as the periodic oscillation and the local extremum. This paper proposes the target segmentation algorithm, which combines the local energy information with improved signed distance regularization term. Firstly, the global image information can be expanded to the HSV space, and each pixels and its statistical properties are analyzed with the local energy information within the neighborhood, which can effectively realize the uneven distribution of color image segmentation in less iteration. Secondly, the improved signed distance regularization term avoids re-initialization of level set function, improving the computational efficiency, and maintains stability in the level set function evolution process. Finally, the termination criterion of threshold evaluation method for the level set function evolution is defined, in order to make the curve accurately evolution to the target contour. The experimental results show that the proposed algorithm has higher segmentation accuracy and robust than other similar models.