Wavelet Transform In Image Noise And Edge Detection

With internet and computer are used more and more today, the multimedia just like audio, video etc. enriches human' s life a lot. Meanwhile, the requirement of image quality is raising day by day. Therefore, it is necessary to do some research on image processing.First and foremost, this paper gives detailed introduction of wavelet analysis development in recent years and its application in image processing, then lists kinds of present conventional image denoising and edge detection algorithms. Moreover, some algorithms are carried out by programs and the experiment results are analyzed, the characteristics of these algorithms are given, too.This paper chooses wavelet analysis as a brand-new tool to propose a new image denoising algorithm, which is called WPT&AT algorithm based on wavelet packet transform and level dependent adaptive threshold. Furthermore, it proposes an algorithm for image edge detection based on the wavelet coefficients obtained from wavelet packet transform. Such edge detection algorithm can keep high frequency parts while the algorithms based on pure wavelet transform can not. It even gives a new method for calculating the threshold so that the denoised image has higher signal-to-noise ratio.At last, the algorithms for image denoising and edge detection proposed in this paper are tested by experiment and are proved to be efficient in image denoising and edge detection.