基于视频片段的视频检索

来源期刊:中南大学学报(自然科学版)2010年第3期

论文作者:夏利民 胡振兴

文章页码:1009 - 1014

关键词:视频片段检索;高维索引结构;k近邻查询;相似度度量;空间和纹理特征

Key words:video clip retrieval; high-dimensional index structure; k-nearest neighbor query; similarity measure; spatial and texture features

摘    要:为提高视频检索的查询效率,提出一种基于视频片段的视频检索方法。该方法利用相邻帧之间的HIS(Hue, Saturation, Intensity)颜色信息特征将视频流分割成子片段,并采用高维索引结构Vector-Approximation Trie (VA-Trie)来组织视频子片段,然后,利用空间和纹理特征定义视频片段的相似度模型,在此基础上采用基于限定性滑动窗口的高效视频检索算法进行视频片段检索。研究结果表明:与其他检索方法相比,该方法能有效地提高视频检索的查全率和查询率,适合用于运动视频检索。

Abstract: A video retrieval method based on video clip was presented to enhance the retrieval efficiency. The video stream was segmented into segments by HSI (Hue, Saturation, Intensity) color information between neighboring frames, and the high-dimensional index structure Vector-Approximation Trie (VA-Trie) was adopted to organize the segments. The new similarity model was defined using spatial and texture features, which can fully take into account the temporal order among video segments. Furthermore, the new query algorithm was used for video retrieval based on restricted sliding window to improve the retrieval accuracy. Experimental results indicate that the proposed approach can improve the recall ratio and precision ratio effectively, and suits the retrieval of sports videos.

基金信息:国家自然科学基金资助项目

有色金属在线官网  |   会议  |   在线投稿  |   购买纸书  |   科技图书馆

中南大学出版社 技术支持 版权声明   电话:0731-88830515 88830516   传真:0731-88710482   Email:administrator@cnnmol.com

互联网出版许可证:(署)网出证(京)字第342号   京ICP备17050991号-6      京公网安备11010802042557号