摘要
关键帧提取是基于内容的视频检索的一个重要的组成部分,所提取的关键帧的有效性,直接影响视频检索的结果。文中提出了一种基于非参数密度估计聚类的关键帧提取方法。首先,通过提取图像的颜色特征和运动特征,然后利用均值漂移聚类方法对融合了颜色和运动信息的特征空间进行聚类。它能自动确定类别数并具有严格的收敛性,从而大大减少了运算量,提高了运算速度。实验证明,本方法的提取结果与人的主观视觉感知系统具有良好的一致性。
Key-flame extraction has been recognized the important research issue in content-based video analysis. In this paper, an efficient key-frame extraction approach is presented based on nonparametric clustering, which provides the capability of browsing digital video sequences more efficiently. Integrating color and motion information is used to descript frame content, then key-frame extraction is accomplished by density-estimation-based nonparametric clustering, the mean shift method, which can efficiently analyze complex multimodal feature space and delineate arbitrarily shaped clusters in it. An experimental system has been build up. Experiments verify the effectiveness of the proposed approach.
出处
《计算机科学》
CSCD
北大核心
2007年第4期119-120,162,共3页
Computer Science
关键词
视频摘要
关键帧提取
均值漂移
非参数聚类
Video abstraction, Key-frames extraction, Mean shift clustering, Nonparametric clustering
