结合核函数与非线性偏微分方程的图像去噪方法

Image Denoising Method Combining Kernel Function and Nonlinear Partial Differential Equation

传统的非线性扩散滤波方法在图像去噪时使用传统的梯度算子,易导致细节缺失。针对该缺点,以非线性偏微分方程和图像结构信息为基础,构造一类带有自适应权值的非线性扩散滤波去噪方式。这种方式选用多方向的拉普拉斯算子模板,结合核函数,自适应地调节权重系数;利用非局部信息选取合适的搜索窗宽,以减小图像噪声的影响。实验证明,该方法既能较好地保存图像纹理细节,又可以达到很好的去噪效果。

The traditional nonlinear diffusion filtering methods use the traditional gradient operator in image denoising,which may easily lead to missing details.In view of this shortcoming,a denoising method based on nonlinear diffusion filter was constructed according to the nonlinear partial differential equation and the image structure information.In this method,the kernel function is used to adaptively adjust the weight coefficient in the multi direction Laplasse operator template,and the influence of image noise is weaken by selecting the appropriate search window width by using the nonlocal information.The experimental results show that this method can not only save the image texture details,but also achieve good denoising results.