南京理工大学学报（自然科学版）2025，Vol.49Issue(2)：167-173,7.DOI:10.14177/j.cnki.32-1397n.2025.49.02.004
基于矩阵补全的黄金比例原始对偶算法及其在煤矿图像恢复中的应用
Golden ratio primal-dual algorithm based on matrix completion and its application in coal mine image recovery
摘要
Abstract
To effectively address the issue of image degradation in coal mine underground video monitoring systems caused by environmental factors or equipment failures,this paper proposes a primal-dual algorithm for matrix completion based on the golden ratio.The proposed algorithm incorporates the golden ratio concept into the primal-dual framework,constructing a convex combination acceleration strategy leveraging the golden ratio,thereby significantly enhancing computational efficiency.The numerical analysis of coal mine underground image recovery tasks across varying scales demonstrates that the proposed algorithm achieves significant improvements over the classical primal-dual algorithm in both recovery quality and computational efficiency.Specifically,the signal-to-noise ratio(SNR)increases by an average of approximately 28%,while the runtime is reduced by as much as 77%.关键词
矩阵补全/图像恢复/凸组合加速/黄金比例Key words
matrix completion/image recovery/convex combination acceleration/golden ratio分类
数学引用本文复制引用
岑小丽,靳嘉琪,闫喜红,徐毅..基于矩阵补全的黄金比例原始对偶算法及其在煤矿图像恢复中的应用[J].南京理工大学学报（自然科学版）,2025,49(2):167-173,7.基金项目
山西省科技创新人才团队专项(202204051002018) （202204051002018）
山西省回国留学人员科研教研资助项目(2022-170) （2022-170）
国家自然科学基金(12371381) （12371381）
