【摘要】 Donoho所提出的去噪算法由于没有考虑到图像的局部特征,而滤除了过多的小波系数,影响了图像的去噪效果。本文介绍了一种自适应阈值的小波系数收缩算法,该算法利用邻域小波系数与噪声方差关系,同时根据小波分解级数的不同而动态地改变小波系数收缩的幅度。由于阈值的自适应性。从而可以利用更多小波分解级数而不会滤除过多的小波系数,因此在去噪方面达到了更好的效果。本文还讨论了如何选取合适的小波分解级数、窗口大小、以及算法的复杂性。通过大量的实验结果可以看出,本算法在去噪的视觉效果和峰值信噪比(PSNR)指标上,优于其他去噪算法,同时,算法的执行和原理非常简单。更多还原

【Abstract】 Donoho’s denoising scheme by thresholding the wavelet coefficients kills too many wavelet coefficients without consid- ering the image’s local characteristic.We introduce an adaptive wavelet coefficients shrinkage method,which can modify the magnitude of shrinkage by considering neighboring wavelet coefficients,variance of noise and decomposition levels.Therefore,we can use more wavelet decomposition levels than other wavelet shrinkage methods to recover the noisy images.We also discussed how to preset the num- ber of decomposition level,window size and the complexity of the algorithm.The proposed method outperforms the other methods in the literatures,while its implement and concept are both simplicity.The results have good visual effect and low mean square error.更多还原