【摘要】 双边滤波器是一种以高斯核为基础的具有图像边缘保留和噪声滤除双重滤波作用的滤波器,能同时兼顾各像素值几何上的临近信息和亮度上的相似信息.为了进一步增加双边滤波器的保边作用,在传统双边滤波器的基础上,结合图像的梯度信息,将传统滤波模板改进为局部滤波模板,使图像自适应地在给出的滤波模板中选择较优模板进行滤波.实验结果证明,改进的双边滤波算法能够更好地保留图像边缘,与经典的双边滤波算法相比,在峰值信噪比和结构相似性上有了提升.更多还原

【Abstract】 Bilateral filter based on Gauss kernel has the double image edge reservation and noise filtering. It can consider the geometric proximity information and brightness similarity information of each pixel value at the same time. In order to further increase the edge-preserving function of bilateral filter,the traditional filter template is improved to local filter template by combining with the image gradient information. So the image can adaptively select the better template in the given filter template for filtering. The experimental results show that the improved bilateral filtering algorithm can better retain the image edge. Compared with the classic bilateral filtering algorithm, the peak signal-to-noise ratio and structural similarity of it are improved compared to the classic bilateral filtering algorithm.更多还原