基于最小方差的自适应中值滤波算法.pdfVIP

中国科技论文在线 基于最小方差的自适应中值滤波算法 钟子豪，刘森，陈岱** （中国矿业大学计算机科学与技术学院，江苏 徐州 221116） 5 摘要：在图像处理应用中，图像噪声消除的好坏决定了图像的质量及后续处理的效果。针对 图像处理中常常遇到的椒盐噪声，提出了一种基于最小方差的自适应中值滤波算法。该算法 采用 n*n 的模板作用于待处理图像的每一个像素点，根据其邻域的阀值极值判断其是否为准 噪声点；然后利用其邻域内的最小方差的大小决定该像素点是否为真噪声点，若为真噪声点 则执行该算法。实验结果表明该算法简单高效，比传统的中值滤波算法具有更高的信噪比和 10 更好的执行效率。 关键词：计算机应用技术；中值滤波；椒盐噪声；最小方差；自适应 中图分类号：TP391 A New Adaptive Median filter algorithm Based On 15 Minimum Variance ZHONG Zihao, LIU Shen, CHEN Dai (School of Computer Science and Technology, Cumt, JiangSu XuZhou 221116) Abstract: In image processing applications, image noise cancellation is good or bad determines the image quality and the effect of subsequent processing. In view of the salt and pepper noise is 20 often met in image processing, put forward a kind of adaptive median filter algorithm based on minimum variance. The algorithm uses the n * n template role in each pixel of the image to be processed, according to its neighborhood threshold extremum judgment whether the accurate noise; And then take advantage of its neighborhood within the size of the minimum variance to decide whether the pixel is true noise points, if is true noise the execution of the algorithm. 25 Experimental results show that the algorithm is simple and efficient than the traditional median filtering algorithm has a higher signal to noise ratio and better performance. Keywords: Computer application technology; median filter; impulse noise; minimum variance; self-adapti