Research On Image Edge Detection Method Based On Wavelet Transform

With the development of technology,digital image processing technology has played an important role in daily life.In the process of image acquisition and transmission,it is often interfered by noise.Noise can affect the expression of effective content in an image,and edges are a response to the main structure and important contours of the image.Better image contour extraction can better directly identify and extract effective information from the image,allowing further image segmentation as a basis.Therefore,removing or suppressing noise and improving the accuracy of image edge detection is an important content for better image segmentation processing.This paper enhances the wavelet threshold denoising and edge detection algorithms by utilizing wavelet theory.Moreover,it applies the refined algorithms to the segmentation of metal casting.The primary objectives of the paper are outlined below:The paper summarizes the theory of wavelet transform,elaborates on the relevant principles and multiscale theory of wavelet signal denoising,and uses MATLAB to conduct simulation experiments on the mallat algorithm.Simultaneously,it examines the principles and characteristics of one-dimensional and two-dimensional wavelet transforms.(2)Analyze and study the categories of image noise and traditional denoising algorithms,introduce evaluation indicators to conduct quantitative and qualitative analysis of image denoising,and summarize the advantages and disadvantages of traditional denoising algorithms.Due to the discontinuity of the hard threshold function at the threshold,the pseudo Gibbs effect can be caused when processing image noise;Soft threshold function can deal with the problem of high-frequency signal loss and easy loss when processing image noise.Improved threshold and improved wavelet threshold function are proposed,which can better determine the threshold.The improved threshold function has smoothness,continuity,and progressiveness.Through mathematical theoretical analysis and experiments,it is proved that the improved threshold function is effective.(3)Analyze the principles of traditional gradient edge detection operators and edge detection algorithms based on wavelet transform,and compare the advantages and disadvantages through experiments.Combining wavelet transform decomposition and reconstruction algorithm with detecting the continuity of image edges,an improved edge detection algorithm combining wavelet transform and morphology is proposed,which can better suppress noise and extract complete edge detection methods.Through experiments,the improved edge detection algorithm is compared with traditional edge detection algorithms,and the results show that the improved algorithm can better preserve the edge information of the image,and the detection effect on noisy images is better.(4)Taking metal casting segmentation as the research object,this paper adopts the denoising algorithm and edge detection algorithm to preprocess the metal casting segmentation process.The experimental results demonstrate the practicality of the enhanced algorithm.The improved wavelet coefficients are used to determine the peak and valley points to determine the segmentation threshold for segmentation.The experimental results show that the segmentation effect is better and meets the use requirements.