基于多元BKF模型的双树复小波图像去噪

本文提出了一种新的去噪方法。该方法将小波系数的内尺度和外尺度信息植入到多元BKF模型之中,从而充分考虑了内外尺度的相关性,并根据最大后验估计(MAP)准则,估计得到图像去噪的阈值化函数,最终,通过逆变换阈值化后小波系数得到去噪后的图像。同时该方法还利用双树复小波变换对图像进行分解和重构,克服了一般离散小波变换缺乏平移不变性和方向敏感性的弱点。去噪实验结果表明,该算法比传统的几种算法有更好的去噪效果。

Image denoising based on a multi-Bessel K form model using a dual-tree complex wavelet

We present an algorithm for image denoising based on a multi-Bessel K form(multi-BKF) model using a dual-tree complex wavelet.This method incorporates both interscale and intrascale wavelet coefficients into the multi-BKF model and takes sufficient account of the relevance of these interscale and intrascale coefficients.A new corresponding nonlinear threshold function is derived from the models using maximum a posteriori(MAP) estimation theory.The denoised image was finally obtained by computing the inverse...