基于数据驱动及多尺度非局部紧框架的乘性噪声去除  

Data-driven multiscale non-local tight frame for multiplicative noise removal

作　　者：王安林 吴玉莲 WANG Anlin;WU Yulian(Faculty of Medical Technology,Xi̓an Medical College,Xi̓an 710021,China;Faculty of Health Management,Xi̓an Medical College,Xi̓an 710021,China)

机构地区：[1]西安医学院医学技术学院,西安710021 [2]西安医学院卫生管理学院,西安710021

出　　处：《计算机应用文摘》2025年第6期81-83,共3页

基　　金：省级大学生创新创业训练计划资助项目(S202311840087)。

摘　　要：为了更有效地去除图像中的乘性噪声,提出了一种基于多尺度非局部紧框架的噪声去除算法。在图像的对数域中,该算法利用多尺度的非局部紧框架进行去噪处理,既考虑了图像灰度局部变化的稀疏性,又充分考虑了图像结构的全局重复性。随后,通过指数变换和误差修正将图像恢复到真实的图像域。实验结果表明,该算法在去除乘性噪声方面具有显著的效果。In order to more effectively remove multiplicative noise in images,a noise removal algorithm based on multiscale non-local compact framework is proposed.In the logarithmic domain of images,this algorithm utilizes a multi-scale non local compact framework for denoising,taking into account both the sparsity of local grayscale changes and the global repeatability of image structure.Subsequently,the image is restored to the true image domain through exponential transformation and error correction.The experimental results show that the algorithm has significant effectiveness in removing multiplicative noise.

关 键 词：乘性噪声 非局部 紧框架 去噪 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]