南京理工大学学报（自然科学版）2016，Vol.40Issue(1)：41-49,9.DOI:10.14177/j.cnki.32-1397n.2016.40.01.007
基于加权运动估计多层次检测的帧率转换算法
Frame rate conversion algorithm based on multi-level weighted motion estimation
摘要
Abstract
To enhance a video's frame rate conversion, a frame rate conversion algorithm based on multi-level motion estimation is proposed. Firstly, more true motion vectors are achieved by the weighted motion estimation. Secondly,we detect moving areas and complex backgrounds in the video by a codebook model. Then,a multi-level overlapped block motion compensation method is used to detect the true object's trajectory. A candidate block merging schedule is also used to protect the moving areas 'edge structure from being damaged. Finally, a reclassification schedule and a smoothing schedule are applied to the moving areas and complex backgrounds,which can effectively eliminate the blocking artifacts and ghost effects. Experiments show that the proposed algorithm perfects a higher image quality than traditional methods,and also has a better adaptability.关键词
运动矢量/加权运动估计/码本模型/多层次检测/运动补偿内插/帧率转换算法Key words
motion vector/weighted motion estimation/codebook models/multi-level detection/motion-compensated interpolation/frame rate conversion algorithm分类
信息技术与安全科学引用本文复制引用
沈娟,田绪婷,郭丹,汪萌..基于加权运动估计多层次检测的帧率转换算法[J].南京理工大学学报（自然科学版）,2016,40(1):41-49,9.基金项目
国家自然科学基金( 61305062 ) （ 61305062 ）
中国博士后科学基金( 2012M511403 ) （ 2012M511403 ）
安徽省自然科学基金(1308085QF102) （1308085QF102）
