| 基于相关反馈的视频检索算法 |
| |
| 引用本文： | 曹建荣,蔡安妮.基于相关反馈的视频检索算法[J].吉林大学学报(信息科学版),2006,24(2):138-143. |
| |
| 作者姓名： | 曹建荣 蔡安妮 |
| |
| 作者单位： | 北京邮电大学,电信工程学院,北京,100876;山东建筑工程学院,信息与电气工程系,济南,250101;北京邮电大学,电信工程学院,北京,100876 |
| |
| 摘 要： | 针对信息检索中如何提高检索的精度问题,提出了一个基于相关反馈的视频检索算法。利用信号在神经网络中来回传递实现自动相关反馈,提高了检索精度,且最大限度地减少了相关反馈中人的参与,最后检索的结果是与查询最接近的一段视频片段。给出了神经网络的构造和具体的检索算法,并与基于关键帧的视频检索方法进行了比较。实验结果表明,用该算法检索的准确度比用基于关键帧的视频检索方法平均提高6%以上。
|
| 关 键 词： | 视频检索 相关反馈 神经网络 |
| 文章编号： | 1671-5896(2006)02-0138-06 |
| 修稿时间： | 2005年4月1日 |
Algorithm Based on Relevance Feedback for Video Retrieval |
| |
| CAO Jian-rong,CAI An-ni.Algorithm Based on Relevance Feedback for Video Retrieval[J].Journal of Jilin University:Information Sci Ed,2006,24(2):138-143. |
| |
| Authors: | CAO Jian-rong CAI An-ni |
| |
| Abstract: | Improving the accuracy of retrieval is very important in information indexing and retrieval.An algorithm of video retrieval based on relevance feedback was presented.It implemented the automatic relevance feedback using the neural network and minimized user interactions.The result of retrieval was the video clip that is the most similar to query.The architecture of the neural network and the strategies of the retrieval had been instructed.The results of experiment had been given and compared with KFVR(Key Frame-based Video Retrieval).The retrieval precision of our algorithm is higher than KFVR about 6%. |
| |
| Keywords: | video retrieval relevance feedback neural network |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
