一种差分进化BM3D硬阈值参数的遥感影像去噪方法  

A differential evolution BM3D hard threshold parameter denoising method for remote sensing images

作　　者：胡鹏程 唐诗华 张炎 刘坤之 吕富强 李灏杨 HU Pengcheng;TANG Shihua;ZHANG Yan;LIU Kunzhi;LÜ Fuqiang;LI Haoyang(College of Geomatics and Geoinformation,Guilin University of Technology,Guilin 541004,China;Guangxi Key Laboratory of Spatial Information and Geomatics,Guilin 541004,China)

机构地区：[1]桂林理工大学测绘地理信息学院,广西桂林541004 [2]广西空间信息与测绘重点实验室,广西桂林541004

出　　处：《测绘通报》2023年第2期72-74,75-77,共6页Bulletin of Surveying and Mapping

基　　金：国家自然科学基金(41864002);广西自然科学基金(2018GXNSFAA281279)。

摘　　要：针对遥感影像高斯白噪声去噪的问题,本文首先利用差分进化(DE)优化三维块匹配(BM3D)中的硬阈值变换参数(参数包括块距离阈值和三维变换域硬阈值参数);然后使用优化的BM3D算法消除影像中高斯白噪声,以峰值信噪比(PSNR)、结构相似度(SSIM)和边缘保留指数(EPI)作为评价指标。试验结果表明,在噪声密度不同情况下,融合算法的PSNR、SSIM和EPI均有所提升,尤其EPI相较于BM3D算法提高约2%。整体上,融合算法的遥感影像高斯白噪声的去噪效果优于单一BM3D算法。Aiming at the problem of denoising of Gaussian white noise in remote sensing images, a differential evolution(DE) algorithm is used to optimize the hard threshold transformation parameters in the 3D block matching(BM3D) algorithm, which includes the block distance threshold and the 3D transformation domain hard threshold parameter, and then the optimized BM3D algorithm is used to eliminate Gaussian white noise in the image. Taking peak signal-to-noise ratio(PSNR), structural similarity(SSIM) and edge preservation index(EPI) as the evaluation indicators, the experimental results show that the fusion algorithm has better performance in PSNR, SSIM and EPI under different noise densities, especially the EPI has increased by about 2%. On the whole, the denoising effect of the remote sensing image Gaussian white noise of the fusion algorithm is better than that of the single BM3D denoising algorithm.

关 键 词：三维块匹配算法 差分进化 硬阈值参数 高斯白噪声去噪 融合 

分 类 号：P237[天文地球—摄影测量与遥感]