摘要
分析了视频图像现有块匹配运动估计的技术特点,通过实验数据定量评价了各算法的优缺点,提出的一种改进的自适应运动估计算法:基于起始搜索点预测的准十字菱形搜索算法.该算法根据序列图像中运动矢量的十字中心偏置分布特性和运动矢量间的时空相关性,设计了一种准十字菱形搜索模板,并对静止块设定阈值,直接中止搜索;结合起始搜索点预测,并根据运动类型自适应选择搜索策略,使本文算法在保证了搜索准确性的同时,提高了运动估计的速度.
The characteristic of block match motion estimation is analyzed by evaluating their advan- tages and disadvantages through the data got from experiments. An improved motion estimation fast algorithm is proposed, which is adaptive quasi-cross-diamond search algorithm based on predictive initial point. Based on the cross-center-bias property and spatial-temporal correlation of the motion vector field, a quasi-cross-diamond search pattern is designed, and the stop criteria is introduced to detect the stationary block, thus terminates the current search immediately. Combined with the prediction of initial search point, and according to the type of local motion, different strategies are adaptively utilized. The algorithm in this paper substantially improves the motion estimation speed, while ensures the search accuracy.
出处
《北京交通大学学报》
CAS
CSCD
北大核心
2008年第6期20-25,共6页
JOURNAL OF BEIJING JIAOTONG UNIVERSITY
基金
国家自然科学基金资助项目(60772040
60772042)
关键词
视频图像处理
菱形搜索
准十字菱形搜索
块匹配
运动估计
video image processing
diamond search
quasi-coss-diamond search
block matching
motion estimation
