Research Of Image Edge Detection Based On Lifting Wavelet And Morphology

With the rapid development and progress of the image communicationtechnology and computer processing technology, an important feature of the imageedges in an image analysis and research has become an urgent need in the field ofsignal and information processing. However, traditional edge detection method isachieved by calculating the gradient and set the threshold, has been unable to meetthe requirements of modern image communication and computer processingtechnology. In the current study, many of which are based on wavelet analysis to theimage edge detection, mathematical morphology, neural networks, fuzzy mathematics,fractal theory. But they are generally a way to achieve, can not achieve satisfactorydetection results.This article combines the lifting wavelet analysis and mathematical morphology,using lifting wavelet image decomposition and reconstruction, grayscale erosion anddilation of the image using mathematical morphology, the use of their advantages ofimage do processing to achieve the purpose of better detection of the target image.Firstly, through the study of image edge detection, in-depth analysis andsummary of these traditional methods, and evaluation of their advantages anddisadvantages. Then a lot of people using the wavelet analysis method to analyze andstudy the basic theory of wavelet transform, using the third-order B-spline waveletfunction to detect the image through the wavelet transform to be evaluated to detectthe image edge. Finally, further research and analysis of these edge detection methods,analysis of the advantages based on lifting wavelet transform method and twomethods based on mathematical morphology, and integration, through simulation andtesting the results of the analysis and comparison of this method with satisfactoryresults.Compared with the widely used wavelet analysis method, the advantages of theproposed algorithm is to take full advantage of the merits of both approaches, betterdetection performance than the separate use of a method, through a variety ofdifferent edge detection methods between fusion and thus produce a new detectionmethod, the field of image edge detection development trend.