Edge Detection and Analysis for Gastric Tumor Pathologic Cell Images Based on Multi-Wavelet
基于多小波的胃癌病理细胞图像边缘检测与分析

A research of edge detection based on multi scale wavelet transform for gastric tumor cell images was made, it could offer doctors a new idea and approach for diagnosis and estimation of relative disease by means of modern information theory. A new edge detection method was put forward by the multi scale wavelet, and some strategies were summed up to improve the effect of the edge detection and the experiment results have been obtained. From comparison with the results, it is discovered that traditional methods of the edge detection may produce the false edges and directional errors for complex texture of pathologic cell images, which influence the reliability of the edge detection. However, the actual edges of images uncontaminated by noise can be gotten by the wavelet method, which has spatial-temporal characteristic and excellent detection capability of fantastic changes.