基于块聚类匹配与PCA的图像去噪方法研究

The Research of Image Denoising Based on Block Clustering Matching and PCA

图像去噪是图像处理领域的重要研究方向,局部块匹配和主成分分析法是图像去噪处理的重要手段,传统的块匹配算法只在固定的窗口范围内进行一次相似度的块筛选,这种搜索方式保留了图像的局部特征但对纹理的保护较差,图像存在失真模糊的现象.为解决这一问题,将聚类匹配和局部筛选相结合,通过聚类类别对样本块进行进一步筛选,同时对匹配窗口的大小进行自适应调整,这种方法可以更好地平衡图像的纹理细节与整体噪声去除之间的矛盾.借助自适应块聚类匹配和主成分分析法对图像进行降噪处理,实验表明,改进后的算法比传统块匹配PCA算法具有更好的去噪效果.

Image denoising is an important research direction in the field of image processing.Local-block-matching algorithm and principal component analysis are important means of image denoising processing.The traditional block-matching algorithm only do the similarity screening once,this searching has strong locality and poor texture protection,which blurs the image.In order to solve this problem,this paper combines cluster matching and partial screening,further screens sample blocks by clustering category,and adaptively adjusting the size of the matching window.It can keep balance between the texture details and overall noise removal better.Base on self-adaption block clustering matching and principal component analysis method,the image is denoised.Experiments show that the improved algorithm has better denoising effect than the traditional block matching PCA algorithm.