一种改进的基于偏微分方程图像的降噪算法

An Improved Image Denoising Algorithm Based on Partial Differential Equation

针对图像去噪过程中存在边缘保持与噪声抑制之间的矛盾,提出了一种改进的基于偏微分模型的图像去噪算法.引入了正则化、绝对差值排序检测法,结合Chao和Tsai模型,构造了一种新的扩散系数函数,兼具了正则化解决方程病态问题和绝对差值排序检测法有效区分噪声与边缘的优点.实验结果表明:与其他基于常用偏微分模型的去噪算法相比,所提算法能更加有效地去除噪声,保留了更多的细节信息,提高了图像去噪的信噪比.

Concerning the contradiction between edge - preserving and noise - suppressing in the process of image denoising, an improved image denoising method was proposed based on partial differential equation （PDE） model. Regularization and rank-ordered absolute differences were introduced into the proposed algo- rithm, which constructed a new diffusion coefficient combined with the model of Chao and Tsai. The method takes advantages of the regulation and rank-ordered absolute differences, so it can solve ill-posed problem of the equation and distinguish between noise and edges more effectively. Experimental results show that the proposed method can retain more information of details and get rid of noise better than the other common PDE models, and improve the SNR of image denoising.