论文部分内容阅读
提出一种基于图像像素分类的小波阈值去噪方法.将图像进行小波变换后的结果看成一幅图像,对小波域中的低频信息利用自适应滤波器进行平滑,而对高频信息按图像像素分类的原则利用图像的方向信息测度来区分边缘和噪声,然后把噪声部分的小波系数置零,最后重构得到去噪图像.实验结果表明,算法可较好地改善图像的视觉效果.
This paper proposes a wavelet threshold denoising method based on image pixel classification.The result of image wavelet transform is regarded as an image and the low frequency information in wavelet domain is smoothed by adaptive filter while the high frequency information is processed by image pixel The classification principle uses the direction information measure of the image to distinguish the edge and noise, then sets the wavelet coefficients of the noise part to zero, and finally reconstructs the denoised image.The experimental results show that the algorithm can better improve the visual effect of the image.