摘要
针对灰度图像受脉冲噪声污染后的恢复处理问题,提出了一种改进的自适应中值滤波算法。该方法根据脉冲噪声的分布特点,采用极大值、极小值和领域均值判定准则进行噪声点的检测,然后用检测窗口内最小非噪声点集合的中值作为噪声点的滤波输出。实验结果表明,与其他几种算法相比,文中算法不仅在峰值信噪比(Peak Signal to Noise Ratio)和结构相似度(Structural Similarity,SSIM)上有较大优势,而且还具有较低的时间复杂度和更好的自适应性。也进一步说明该方法不仅能有效地检测并滤除噪声点,还能较好地保护图像的边缘细节。
This paper proposes images polluted by impulse noise. an improved adaptive median filtering algorithm for the recovery of gray-scale According to the distributed noise, this method checks and identifies noise through maximum-minimum and field mean criterion, and then filters noise using the median of the smallest noise free pixels collection in the detected window. Experimental results indicate that the proposed method is of higher peak signal noise ratio （PSNR） and structural similarity （SSIM） and has lower time complexity and better adaptabili- ty than other methods. Furthermore, the proposed method can not only effectively detect and filter out the noise but also preserve the image details.
出处
《电子科技》
2013年第12期5-9,33,共6页
Electronic Science and Technology
