计算机与数字工程2019，Vol.47Issue(10)：2589-2596,8.DOI:10.3969/j.issn.1672-9722.2019.10.041
基于语义分割和像素非线性核的视频去模糊∗
Video Deblurring Based on Semantic Segmentation and Pixel Nonlinear Kernel
摘要
Abstract
Due to factors such as camera shake,object motion and depth change will inevitably cause video blur,this paper uses the semantic segmentation in each fuzzy frame to understand the scene content,and uses different image region motion models to achieve optical flow estimation. The relationship between motion blur trajectory and optical flow is analyzed,and a pixel fuzzy nonlinear kernel(PWNLK)model is proposed to explain motion blur. The proposed fuzzy model is more effective in describing com?plex based on nonlinear optical flow. A lot of experiments on fuzzy video show that the proposed algorithm has better performance than other methods.关键词
视频去模糊/语义分割/像素模糊/非线性核Key words
video deblurring/semantic segmentation/pixel blur/nonlinear kernel分类
信息技术与安全科学引用本文复制引用
董飞,马源源..基于语义分割和像素非线性核的视频去模糊∗[J].计算机与数字工程,2019,47(10):2589-2596,8.基金项目
陕西铁路工程职业技术学院基金项目"基于OpenCV的目标跟踪算法研究与应用"(编号:Ky2017-082)资助. （编号:Ky2017-082）
