A Self-adaptable Method to Detect Edge in Images
Yang Zhen · Jisuanji gongcheng · 2003
The capability of edge detection using LOG operators is analyzed and evaluated i n this paper. In order to avoid the defects of LOG operator. This thesis propose d an new edge detection method here. Firstly, by combining the first derivative s maximum and the second derivatives zerocrossing of image grayscale. It has made a good success in eliminating most of noise in image. Subsequently, by using variance of grayscale to selfadaptably definitize Gassian space coeff icient and edge detection threshold, the image edge can be automatically detecte d.