智能系统学报2024，Vol.19Issue(6)：1593-1603,11.DOI:10.11992/tis.202306002
基于光流和多尺度特征融合的视频去噪算法
Video denoising based on optical flow and multi-scale features
摘要
Abstract
To effectively eliminate noise from videos while preserving texture details,a cascade video denoising al-gorithm that integrates optical flow and multi-scale features is proposed.The process begins by accurately aligning se-quence frames using a grouping strategy.These frames are then processed through a multi-scale architecture that com-bines residual refinement and selective skip connection.This approach not only preserves detailed features but also en-hances alignment and fusion.Furthermore,a non-local attention mechanism is employed to deeply mine spatiotemporal features,enabling the reconstruction of high-quality videos.To preserve detailed textures,a target function supervision training method that combines perceptual loss is proposed.Experimental results show that the proposed algorithm re-tains more texture features and aligns well with human visual perception.It is also highly robust,has low computational complexity under strong noise,and meets real-time denoising requirements.关键词
多帧去噪/视频去噪/光流对齐/感知损失/非局部注意力/图像处理/计算机视觉/深度学习Key words
multi-frame noise reduction/video denoising/optical flow alignment/perceptual loss/non-local attention/image processing/computer vision/deep learning分类
信息技术与安全科学引用本文复制引用
孙立辉,陈恒,商月平..基于光流和多尺度特征融合的视频去噪算法[J].智能系统学报,2024,19(6):1593-1603,11.基金项目
河北省重点研发计划项目(20350801D). （20350801D）
