Font Size: a A A

Video Retrieval Based On Color Histogram Value

Posted on:2008-11-02Degree:MasterType:Thesis
Country:ChinaCandidate:H ZhangFull Text:PDF
GTID:2208360212474346Subject:Communication and Information System
Abstract/Summary:PDF Full Text Request
With the fast development of multimedia technology and the popularization of Internet, especially on the stimulation of some information sources, such as artificial satellite, fingerprint distinguish, scientific experiments, biological medicine imaging, large-scale entertainment systems and so on, a great deal of image and video are created. Data is no longer the simplex text information, the amount of image and video information updated on the Internet everyday has reached a hundred million and it is much easier to obtain almost infinite vidoe information from the network. And then a new problem appears——how to organize and manage the multimedia information effectively? As video information is heavy and abundant, the effective retrieval will become a subject to be solved urgently.At present, the management and retrieval of video are mainly based on the description of video files, and it is not all-pervading to operate the inside details of the files, in other words, we seldom retrieval vidoe files based on their actual contents, which are more attention-getting to the users. Therefore, Content-based Video Retrieval (CBVR) becomes a hot research topic which researchers care about strongly and universally in recent years.The research of Video Retrieval Based on Color-Histogram Eigenvalue of the key frames in this paper can exactly meet the needs mentioned above, and the main structrue and content are arranged as the following text:In Chapter1 and Chapter 2 , the background and the development status of video retrieval technology are introduced. And then, the characteristic of video data and the model structure of video analysis are presented.The format of AVI video files is also explained, because it is chosen to be the research object in this paper.In Chapter3, how to detect shot transition and to extract key frames are researched. After introducing several kinds of shot types and the basic methods of shot detection, the detection of cut shot and dissolve shot is reserched emphatically, and detection based on color-histogram eigenvalue or gray-histogram eigenvalue are adopted respectively in the experiments for cut shot and dissolve shot detecting. In fact, the results show that these methods are not only simple to carry out, but...
Keywords/Search Tags:shot cutting, key frame, color-histogram, similarity matching measurement, conten -based video retrieval
PDF Full Text Request
Related items