镜头边界检测中的不连续度设定与阈值判定问题研究-通信与信息系统专业论文.docxVIP

上海交通大学硕士学位论文摘要 上海交通大学硕士学位论文 摘要 第 第 PAGE IV 页 不足，提出了以下两种方法： 1. 在 MPEG 压缩域中，本文选取 I 帧的 DC 系数，分块统计直方图， χ2 比较得不连续度，与时间因子和运动差异因子所决定的自适应 阈值比较判别镜头边界。该方法无需解码视频，实时性高。 2. 在非压缩域中，设定可变窗，分块比较窗中第一帧和最后一帧的 YCbCr 直方图得不连续度，并根据其大小，动态调整下一窗口跨 度。该方法仅需解码少量的帧，特征选取更丰富，鲁棒性高。 本文通过三段具有代表性的视频，分别对比了双因子自适应阈值与 固定阈值，整体直方图与不同分块、不同权重直方图的实验结果，证明 了本文提出的自适应阈值和分块直方图的优越性。本文提出的两种镜头 边界检测算法，经过 TRECVid 权威数据库测试，在综合性能上优于一些 同类算法，达到了预期的研究目标。 关键词：镜头边界检测，时间因子，运动差异因子，可变窗 上海交通大学硕士学位论文AB 上海交通大学硕士学位论文 ABSTRACT STUDY ON DISCONTINUITY AND THRESHOLD IN SHOT BOUNDARY DETECTION PROBLEM ABSTRACT With digital technology widely used in video, videos could be repeatedly shot by digital video camera, easily edited through computer software, permanently stored by CD-ROM, rapidly transmitted through the network and vividly projected to the digital screen. Great digitalization makes it happen. However, the explosive growth of the videos (online videos especially), coupled with the video content nature, non-level and non-structural, causes video classification, indexing, archiving and searching a tremendous duty. On the other hand, YouTube and other video sites have attracted millions of Internet viewers around the world, It’s reported that in every minute about 20- hour long videos are uploaded to the YouTube, of which there is some of sex and bloody scenes. How to identify the illegal content precisely and completely from those upload videos has become a heavy burden for the video website staff. In this field, content-based video retrieval research has been raised by the academia. Similar with the method in text content analysis, we first format the video according to the content, achieve high-level semantics (content) analysis and extraction and then do video classification, indexing, archiving etc. This paper studies video shot boundary detection which is the most basic work in video content formation. In this paper, shot boundary detection problem is divided into three independent sub-problems: feature extraction, discont