改进的小波域耦合偏微分方程图像去噪模型

摘要

针对全变分及四阶偏微分方程图像去噪模型的不足,利用小波变换能够聚焦到图像细微变化的优势,提出一种基于小波域的偏微分方程图像去噪算法.并通过对小波的阈值和阈值函数做适当的改进以及利用加权函数将全变分和四阶偏微分方程去噪模型相结合的方法,得到一种改进的小波域耦合偏微分方程图像去噪模型.MATLAB仿真结果表明,该模型和小波软阈值去噪、全变分模型以及四阶偏微分方程图像去噪模型相比,峰值信噪比有明显的提高,而且能够在很好地保留图像的边缘和细节信息的同时提高处理噪声的效率.%Aiming at the shortcomings of the total variation and fourth order partial differential equation(PDE)image denoising model,by using the advantage of wavelet transform can focus on the subtle changes of the image.An image de-noising algorithm based on partial differential equation in wavelet domain was proposed.By making appropriate improve-ments to the wavelet threshold and threshold function using the weighted function combining total variation and fourth order partial differential equations.An improved image denoising model based on partial differential equation in wavelet domain was obtained.Compared with the wavelet soft threshold denoising and fourth order partial differential equations,MATLAB simulation results show that the peak signal to noise ratio of the new model have obvious enhancement.Meanwhile,new model can keep image edges and details very well.Finally,it can also improve the efficiency of handing noise.