一种基于图像边缘检测的小波阈值去噪方法

Wavelet image threshold denoising based on edge detection

边缘特征是图像最有用的高频信息,因此,在图像去噪的同时,应尽量保留图像的边缘特征。为实现这一想法,提出了一种基于图像边缘检测的小波阈值去噪新方法。该方法在去噪前,先用定位精度高的小尺度LOG算子检测图像的边缘,对检测出的边缘进行均值平滑滤波,以减少边缘图像中的孤立点噪声;进而再对图像边缘和含噪图像分别进行小波分解,根据分解后的小波系数以确定图像的边缘特征和非边缘特征;最后,再对图像边缘对应的小波分解系数进行小阈值处理,而对非边缘的则进行大阈值处理,从而实现了在去噪的同时保留了图像边缘特征的目的。实验结果表明,与普通的小波阈值去噪方法相比,该方法可有效地保持图像的边缘信息,去噪效果则优于前者。

Edge information is the most important high frequency information of an image. Therefore it is vital to maintain more edge information in the process of denoising. The image denoising method, wavelet image threshold denoising based on edge detection, was presented. Before denoising, the edge of a noised image was detected with smaU-seale LOG operator which had higher orientation precision, and the image edge was smoothed with average filter to reduce a great lot of isolated point. Next, the image edge and non-edge character in wavelet coefficient were ascertained by decomposing the noised image and edge image, and the edge coefficient of wavelet decomposed with lower thresholds and non-edge coefficient were dealt with higher thresholds. In this way, the edge could be protected while denoising. Experiment results show that, compared with the eommouly-used wavelet threshold denoising methods, this method can keep image＇s edges from damaging and excels eommouly-used wavelet threshold denoising methods.