The Research On Image Edge Detection Algorithm Based On Wavelet And Improved Morphology

Edge detection is the basis of research in the field of image processing,is the precondition of image segmentation and pattern recognition,and it is a key step in image processing.The existing algorithm for edge detection often has the problem that the edge is not continuous,the positioning is not accurate,the detected edge has the false edge,and the anti noise ability is poor.The maximum value of wavelet transform dual threshold method based on the theory of selecting edge points by using the dual threshold method in wavelet transform modulus,has obvious advantages in edge positioning accuracy,anti noise ability,but this method in the detection process,only for the high frequency part of wavelet decomposition,ignoring the low-frequency information;therefore,for the low frequency part.,using mathematical morphology algorithm to improve the processing of these three factors,considering the scale,orientation and shape,choosing the proper structure element is applied to the anti noise morphological operator,good continuity so that the detected edge,can extract more effective edge,and anti noise ability is very strong;and then the fusion of high low frequency edge image subtraction method,we get the final image.The experimental results show that the improved algorithm has better noise suppression ability,the edge detection is more continuous and clear,and the positioning accuracy is higher.