Journal of South China University of Technology (Natural Science Edition) ›› 2009, Vol. 37 ›› Issue (10): 39-43.

• Electronics, Communication & Automation Technology • Previous Articles     Next Articles

Sub-Pixel Edge Detection Method Based on Sigmoid Function Fitting

Zhang Wu-jie1  Li Di 2  Ye Feng2   

  1. 1. School of Automation Science and Engineering, South China University of Technology, Guangzhou 510640, Guangdong, China; 2. School of Mechanical and Automotive Engineering, South China University of Technology, Guangzhou 510640, Guangdong, China
  • Received:2008-11-07 Revised:2009-02-27 Online:2009-10-25 Published:2009-10-25
  • Contact: 张舞杰（1970-）．男，博士后，主要从事图像处理、模式识别、过程监控、嵌入式装备控制等研究． E-mail:zwjllhu@seut．edu．cn
  • About author:张舞杰（1970-）．男，博士后，主要从事图像处理、模式识别、过程监控、嵌入式装备控制等研究．
  • Supported by:

    中国博士后科学基金资助项目（20070420784）;广东省博士启动基金项目（841064101000594）;广东省工业攻关项目（20081301040004）

PDF

1779

Abstract:

Sub-pixel edge detection is an effective way to improve the accuracy of edge detection using image processing algorithms. In this paper, the principles, advantages and shortcomings of such existing sub-pixel edge detection algorithms as the moment method, the fitting and the interpolation methods are analyzed, and a novel sub- pixel edge location algorithm is proposed based on the Sigmoid function fitting. This algorithm employs the Sigmoid function to obtain an edge model and uses the image edge gray data to perform a nonlinear least-square fitting for the edge model. Thus, the sub-pixel location of image edge is obtained. Theoretical analyses and experimental results demonstrate that the precision of the proposed algorithm based on Sigmoid function fitting achieves 0. 045 pixel, and the detection rate increases by one order of magnitude as compared with that of the gray moment method, and by two orders of magnitude as compared with those of the spatial moment, the Zemike moment and the interpolation methods. Thus, it well satisfies the requirements for strong stability, high precision and strong real-time performance in image measurement.

Key words: sub-pixel, Sigmoid function, edge detection, fitting, edge location