论文部分内容阅读
提出了一种基于多小波变换的综合阈值图像去噪方法。该方法通过对含噪图像进行多小波变换,克服了单小波变换中无法同时满足正交性和对称性的缺点。同时,在两种经典的硬、软阈值处理方法基础上,提出了一种综合阈值处理方法。将该阈值方法和多小波变换相结合,根据多小波分解后的能量分布特性,在不同尺度的子带选择不同的最佳阈值,有效地提高了重构图像质量。并且对阈值进行向零收缩处理,防止有用信息当作噪声滤除。实验结果表明,相对于传统的硬、软两种阈值处理方法,文中的去噪方法在输出信噪比和主观视觉效果上都有明显改善。
A multi-wavelet transform based threshold thresholding method is proposed. This method overcomes the shortcomings that the orthogonality and the symmetry can not be satisfied in the single wavelet transform by multi-wavelet transform of noisy images. At the same time, based on two classic hard and soft threshold processing methods, an integrated threshold processing method is proposed. Combining this threshold method and multiwavelet transform, according to the energy distribution characteristics after multi-wavelet decomposition, different optimum sub-bands are selected at different scales to effectively improve the reconstructed image quality. And the threshold value to zero shrink processing, prevent useful information as noise filter. The experimental results show that compared with the traditional hard and soft threshold processing methods, the denoising method in the paper has obvious improvement in output SNR and subjective visual effects.