摘要
针对珍贵视频影像中的斑点识别与修复问题,提出了一种基于边缘检测的斑点识别算法.首先利用边缘检测算法提取当前帧的大致边缘,然后依据边缘像素,利用提出的正反扫描方法和局部运动估计的SROD算法识别出斑点全部像素点.实验表明,提出的算法时间复杂度是SROD算法和Gullu MK算法的1/4到1/2,而斑点识别效果有较明显的改善,为后续斑点修复提供了快速准确的定位信息.
For solving blotch identification and restoration problems in precious archival films,an edge detection-based blotch identification algorithm is proposed.Firstly,edge detection algorithm is adopted to extract approximate edges of the current frame.Then,according to edge points,all blotch pixels are identified through the proposed sequential-reverse scanning method and the local compensation-based simplified rank ordered differences（SROD） algorithm.Experimental results show that the time complexity of the proposed algorithm is 1/4 to 1/2 of the SROD algorithm and Gullu MK algorithm;and if achieves significant improvement in identification results.The proposed algorithm provides fast and accurate blotch location information for the following blotch restoration process.
出处
《三峡大学学报（自然科学版）》
CAS
2010年第6期84-88,共5页
Journal of China Three Gorges University：Natural Sciences
基金
国家自然科学基金(60572048
60972162)
湖北省教育厅科学基金项目(Q20101202)
关键词
斑点识别
边缘检测
简化的等级排列差分算法
影视资料
blotch identification
edge detection
simplified rank ordered differences（SROD）
archival film
