自适应中心加权的改进均值滤波算法

Modified Mean Filtering Algorithm based on Adaptive Center Weighting

多窗口滤波策略(MWA)算法对受到椒盐噪声污染严重的图像仍然有很好的滤波效果,但其运算量相对较大。针对这一问题,基于中值滤波法计算量偏小的优势,将这两种方法进行有机的融合和改进,提出了一种基于中值的极值算法。该算法的运算量较小,并能广泛应用于受椒盐噪声污染程度不同的图像。通过实验对比,提出的中值极值算法能在较短时间内实现对一幅噪声图像的滤波,具有可靠性和鲁棒性,应用前景广阔。

The MWA algorithm still has a good filtering effect on images that are heavily polluted by salt and pepper noise, but its computational complexity is relatively large. Aiming at this problem, based on the advantage of the median filtering method, the two methods are organically fused and modified, and a median-based extremum algorithm proposed. The algorithm has a small amount of computation and can be widely applied to images with different levels of salt and pepper noise pollution. Experiment and comparison indicate that the proposed median extremum algorithm can filter a noise image in a short time, and with reliability and robustness, has broad application prospects.