摘要
关键帧提取是基于内容的视频检索中的重要一步,为了能够有效地提取出不同类型视频的关键帧,提出一种基于粒子群的关键帧提取算法。该方法首先提取出视频中每帧的全局运动和局部运动特征,然后通过粒子群算法自适应地提取视频关键帧。实验结果表明,采用该算法对不同类型的视频提取出的关键帧具有较好的代表性。
Key frame extraction was an important step in video retrieval.In order to effectively extract key frames of different video types,a key frame extraction algorithm based on particle swarm was proposed in this paper.This method first extracted the global motion and local motion features in each frame,and video key frame was extracted by Particle Swarm Optimization（PSO） adaptively.The experimental results show that the key frame extraction algorithm for different types of video is more representative.
出处
《计算机应用》
CSCD
北大核心
2011年第2期358-361,共4页
journal of Computer Applications
基金
国家自然科学基金资助项目(60673190)
江苏省自然科学基金资助项目(BK2009199)
关键词
视频检索
关键帧提取
粒子群
运动特征
video retrieval
key frame extraction
Particle Swarm Optimization（PSO）
motion characteristic
