Edge-detection in optical images using auto-organization
Fucheng Liu, Lifang Dong, Yafeng He, Weili Fan · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · 2007
Edge detection is one of the most demanding tasks in optical image processing for artificial vision and image matching works. In this paper, the self organization theory is used for edge detection. A new algorithm based on a two variable reaction-diffusion equations is proposed. A stable edge pattern can be obtained by choosing suitable control parameters. Compared with the conventional methods, such as Soble, Prewitt and Robert detectors, the new algorithm indicates a higher accuracy and continuity for the image. Moreover, it can also extract exactly the edge of the human face image.