摘要
一种新的基于小波变换的图像消噪方法是在运用小波变换对含噪图像进行消噪前,先对图像进行小波级数分解,对其中的低频系数和高频系数进行适当的放大;然后对图像采取局部阈值消噪法进行消噪;最后运用小波变换对所得到的图像小波系数进行适当的缩小并将其重构.仿真实验证明这种方法比一般的诸如中值滤波和维纳滤波等图像消噪方法有很大的改进,特别是图像均方差(MSE)有很大的降低,而图像的信噪比也有较为明显的提高.
A new image denoising method based on wavelet transform is proposed. The coefficients including low frequency and high frequency of wavelet transform are magnified properly before using image denoising method based on wavelet transform. Then the noise in image is eliminated by using local threshold wavelet method. In the last, the coefficients of wavelet transform are reduced aptly and restructured. According to the result of experiment, the given algorithm is much better than the conditional median value filtering method and Wiener filtering method. Especially, the MSE of image has been decreased a lot while the PSNR being improved markedly.
出处
《云南民族大学学报(自然科学版)》
CAS
2005年第1期71-74,共4页
Journal of Yunnan Minzu University:Natural Sciences Edition