基于暗通道先验去雾的SDSS图像去噪研究  

SDSS image denoising research based on dark channel prior defogging

作　　者：刘执靖 许婷婷[1] 邓雨禾 杨明存 李广平 高献军 曹婕 周卫红[1,2] LIU Zhi-jing;XU Ting-ting;DENG Yu-he;YANG Ming-cun;LI Guang-ping;GAO Xian-jun;CAO Jie;ZHOU Wei-hong(School of Mathematics and Computer Science,Yunnan Minzu University,Kunming 650500,China;Key Laboratory of the Structure and Evolution of Celestial Objects,Chinese Academy of Sciences,Kunming 650011,China)

机构地区：[1]云南民族大学数学与计算机科学学院,云南昆明650500 [2]中国科学院天体结构与演化重点实验室,云南昆明650011

出　　处：《云南民族大学学报（自然科学版）》2024年第3期351-358,共8页Journal of Yunnan Minzu University:Natural Sciences Edition

基　　金：国家自然科学基金(61561053);云南省教育厅科学研究基金(2023J0624).

摘　　要：美国的斯隆数字巡天(sloan digital sky survey,SDSS)于2000年开始正式的巡天观测,目前已经产生了海量的光谱和图像数据,但受仪器设备本身和观测条件的影响可能会出现观测数据细节模糊或损失等问题.在进行后续数据分析和挖掘之前,对观测图像进行去噪处理是一个非常必要的步骤.基于此,将暗通道先验(dark channel prior,DCP)去雾算法应用于SDSS测光图像的去噪处理中,并与B2U方法进行对比分析.结果表明:与B2U方法相比,DCP算法去噪效果在峰值信噪比(peak signal-to-noise ratio,PSNR)上提高了7.98 dB,结构相似性(structural similari⁃ty,SSIM)上提高了7%.The Sloan Digital Sky Survey(SDSS)in the United States began its formal survey in 2000,and it has produced a large number of spectral and image data.However,due to the influence of the equipment itself and ob⁃servation conditions,the details of the observation data may be blurred or lost.Before the subsequent data analysis and mining,it is a very necessary step to denoise the observed image.Based on this,the Dark Channel Prior(DCP)defogging algorithm is applied to the denoising of photometric images in SDSS,and compared with the B2U method.The results show that compared with the B2U method,the denoising effect of DCP algorithm is 7.98dB higher in Peak Signal⁃to⁃Noise Ratio(PSNR)and 7 percentage point higher in Structural Similarity(SSIM).

关 键 词：图像去噪 SDSS测光图像 暗通道先验 自监督算法 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术] P157[自动化与计算机技术—计算机科学与技术]