| Video and other multimedia files are more and more used in various fields, because of their vitality and diversity, as well as the capacity to store large amounts of information. Traditional page-oriented Web search engine is usually only concerned with pure text message and when it comes to video content, it can only deal with text-form information added manually; meanwhile normal people detect programs usually simply analyze the color information of each individual frame of a video to extract content information, which means lack of contextual overall information and accuracy. To achieve the purpose of more accurately extracting the video information and searching, this system can automatically extract the content features of video files, and quickly find video files according to the keywords describing the features of video content, through the co-ordination between the analyzing process of contextual color features of video files based on OpenCV (an open source computer vision library), and the index-and-search process of feature metadata based on Lucene (an open source search engine library).It can be verified that this system has the advantages of high efficiency, high accuracy, and ease of use, etc. |
PDF Full Text Request