Adaptive Scale for Contour Completion

摘要：

　　Tensor voting provides a unified feature extraction framework, which incorporates perceptual organization laws into image processing and gains its popularity in many applications, what is more, scale is the only parameter for this algorithm.Here, we consider the selection of the scale for different images.This paper puts forward an autoregulation method for scale parameter and examines the method by experiments on four kinds of typical contours with different density of voting tokens.Experimental results show that the proposed scale computation method can be adaptive to change in voting tokens density.

关键词： adaptive scale voting curve completion

作者: Xiaofang Shao Xiaojun Chu Cuijuan Sun

作者单位: Department of Electronics, NAEI, Qingdao, China Gaming Center, NCC, Nanjing, China

会议类型: 国际会议

会议名称: International Conference on Computational Science and Engineering Applications(CSEA2015)2015计算机科学与工程应用国际会议

会议地点: 三亚

会议语种:英文

页码: 49-54

在线出版日期: 2015-12-26（万方平台首次上网日期，不代表论文的发表时间）