【摘要】 小波收缩是一种非线性小波变换,这种算法的关键问题在于收缩算法中的收缩阈值和收缩函数(规则)。自适应理论是现代信号处理中强有力的工具。该文将自适应理论和传统的小波收缩算法相结合,提出了基于图像奇异特性自适应小波收缩"去噪"算法。该算法根据高频子带小波系数的均方根来确定最佳小波收缩阈值。阐明了最佳阈值与图像本身特性之间的关系。实验表明,该算法比一般软、硬阈值的小波收缩算法有更好的"去噪"效果,既克服硬阈值函数所产生的人为的噪声点和数学上不易处理等缺点,又避免了软阈值算法所带来的边缘模糊。从而进一步提高了图像的峰值信噪比,改善图像质量。更多还原

【Abstract】 <Abstrcat> Wavelet shrinkage is a nonlinear wavelet transform, whose key is shrinkage thresholds and shrinkage functions (shrinkage rules). The adaptive theory is a strong tool in modern signal processing.Combining the adaptive theory with classical wavelet shrinkage, a wavelet shrinkage denoising method with adaptive thresholds based on image singularity characteristic is proposed. The optimal wavelet shrinkage threshold in this method is determined by the root mean square value of high frequency wavelet coefficient. The idea illustrates the relation between the near-optimal threshold and image’s characteristic.Experiment shows that the method proposed in the paper has better denosing effect than that of other methods based on hard-thresholds or soft-thresholds. The new method overcomes the shortcomings in hard-thresholding scheme: more tractable mathematically in the analysis of algorithms and the reconstructed images appear as annoying, spurious ’blips’.It also avoids blurring the brim of the reconstructed images in soft-thresholding scheme. Thus, it further increases the peak signal-to-noise ratio and improves the image’s quality.更多还原