Research on image edge detection algorithm based on improved guided filtering
A Canny operator edge detection algorithm based on adaptive weighted guided filtering is pro-posed to solve the problem that the image smoothing in the Canny edge detection algorithm causes the image detail loss and the image edge cannot be dynamically discriminated and sharpened.This presented algorithm makes two improvements to the guided filtering,one is to increase the edge slope of the guided image by the shifting technology to achieve image sharpening,while the other is to introduce a segmented weight model to adaptively modify the regularization parameters of the guided filtering algorithm.The edges and non-edges can be distinguished according to the gradient information,thereby the degree of smoothness can be auto-matically changed.The PSNR value of the images processed by this proposed algorithm is increased by 1.50-6.92dB,and the PFOM value is increased by 0.02-0.31 compared with the traditional Canny algo-rithm.Experiment results show that the algorithm can effectively sharpen the images,extract complete and clear edges.
万方数据
维普
