改进小波阈值和全变分图像去噪

Image Denoising Based on Improved Both Wavelet Threshold and Total Variation

针对图像去噪需求设计一种修复算法,引入小波变换并改进,提取高精度滤波阈值,清除噪声小波,实现图像初次去噪,然后采用改进的全变分去噪,再次进行图像去噪,提升图像去噪效果,最后引入Laplace算子,提高图像清晰度.从视觉效果、相似度和信噪比三个方面与当前主流算法进行分析和验证,结果表明该算法图像去噪效果更好.

A restoration algorithm was designed for image denoising,and wavelet transform was introduced and improved to extract high-precision filtering threshold and remove noise wavelet for initial denoising of image.The improved total variation was applied to improve the image denoising effect again.The Laplacian operator was applied to improve image sharpness in the end.From experiments and comparisons with current mainstream algorithms as well as the analysis from three aspects,namely visual effect,image SNR and similarity,the results demonstrate that this algorithm significantly improves images denoising effect.