一种自适应加权的灰度形态学图像滤波算法

An Adaptive Weighted Image Filtering Algorithm Based on Grayscale Morphology

数学形态学滤波属于非线性滤波,针对传统的线性滤波对图像边缘等细节特征容易模糊的缺点,本文提出一种基于灰度形态学的自适应加权复合滤波优化算法,采用凸性多尺度结构元素对灰度图像进行多级形态开闭运算,根据形态滤波等幂特性,实现结构元素序列的自适应加权复合滤波,并对滤波效果进行PSNR评价,实验仿真结果表明该复合滤波算法在保留图像细节特征的同时能有效去除噪声,提高了图像信噪比.

Mathematical morphology filter is a nonlinear filter. In viewing of the disadvantages for the traditional linear filtering is easy to blur the image edges,this paper presents an adaptive weighted composite filter optimization algorithm based on grayscale morphology. Base on the characteristic of nonlinear filtering,under the effect of convex multi-scale structural elements,this algorithm performs multistage morphological open-closed operation on grayscale image,and implements adaptive weighted composite filtering,and evaluates the effect of composite filtering by PSNR. The simulation results show that this adaptive weighted composite filtering algorithm can not only effectively remove noises but also preserve grayscale image details,and the PSNR values get improved.