基于内容的排球视频检索技术应用-控制科学与工程专业论文.docx

基于内容的排球视频检索技术应用 基于内容的排球视频检索技术应用 APPLICATIONS ON CONTENT-BASED VOLLEYBALL VIDEO RETRIEVAL TECHNOLOGY ABSTRACT Content-based video retrieval has been very widely used, which plays an important role in the rapid development of multimedia technology. It also takes effective management and retrieval for large quantities of multimedia data, especially for the increasing number of sports video, this has already became hot in the field of multimedia. Firstly, based on analyzing the related technologies of the original video retrieval, we propose an algorithm of adaptive dual-threshold shot boundary detection for the game video of TV. At the beginning, we convert each frame to the HSV space, set high and low ratio of threshold according to the frame difference. Then, the average value of frame difference in the sliding window is calculated to receive the dual-threshold. Finally, the shot and the gradual are detected by comparing the frame difference and threshold. Secondly, this paper classifies the state of the self-recorded game video, detects the bottom line by Canny operator and separates the zones of service and offensive, detects movements in the corresponding regions. Then, according to the characteristics of difference states of the game, the game station is judged and the video is split. Thirdly, this paper also proposes an algorithm of key frame extraction. For television video, the distance between the candidate frames was calculated by the trends in content changes, and one to three frames are extracted as key frames. For self-recorded game video clips, a key frame is chose by the states classification. This paper designed the system to achieve the above algorithms. Through the experiments of multiple videos, it comes to an expected result, verifies the algorithm’s feasibility and robustness. KEY WORDS: video retrieval, shot detection, gradual detection, adaptive dual-threshold, Canny operator ii 原创性声明 本人郑重声明：所呈交的学位论文，是本人在导师指导下，进行研究工作所取得 的成果。除文中已经注明引用的内容外