一种基于Canny算子的多重边缘检测算法

Multiple Edge Detection Algorithm Based on a Canny Operator

选择Canny算子为基础算子,对图像进行第一次边缘检测,在对原图进行顶帽变换后进行二次边缘检测,然后在顶帽变换的基础上对图像进行对数变换进行第三次边缘检测,将三次边缘检测的结果叠加,最后进行骨骼化处理,得到清晰简约的边缘检测图.实验结果表明:该算法能较好地实现对边缘尤其是弱边缘的提取.

Canny operator was used as a basic operator to have the first time edge detection of the image. After the top-hat transformation of the original image, the second time edge detection was applied. On the base of the top-hat transformation image, logarithm transformation was carried out to start the third edge detection. After stacking up the three results of edge detections and skeletonizing it, a clear and simple image of edge detection was finally got. The experiment showed that the algorithm can well realize the extraction of the edges, especially the weak edges.