作　　者： ; ;

机构地区： 吉林师范大学信息技术学院

出　　处： 《长春工业大学学报》 2008年第2期162-164,共3页

摘　　要： 提出基于小波变换与中值滤波相结合的方法,实现了图像去噪。该方法在去噪之前,先通过小波边缘检测确定图像边缘特征的小波系数,保留这些位置的小波系数,其不受阈值去噪影响,对其它位置的小波系数进行自适应阈值去噪,去除高斯噪声。然后对图像进行中值滤波,去除椒盐噪声。该算法的实验结果表明,不仅能滤出图像中高斯噪声和椒盐噪声的混合噪声,而且能较好地保留图像的边缘细节,其滤波效果优于传统的图像去噪方法。 An efficient algorithm based on wavelet transform to remove the noise in the image is proposed, in which, the wavelet coefficients of the image edge features are determined before denoising. The coefficients are kept while the other are denoised with adaptive threshold to remove the Gaussian noise. Then the image is dealt with median filter to remove the noises. The experiment results show that the algorithm not only can efficiently remove the mixed and Gaussian noises but can keep image edge details as well. The filtering performance is better than that of traditional methods.

关 键 词： 小波变换 高斯噪声 椒盐噪声 图像去噪

领　　域： [电子电信] [电子电信]