Dissertation > Excellent graduate degree dissertation topics show

Based on chi-square test and macroblock type statistics of the video shot boundary detection algorithm

Author: ChenZuo
Tutor: JiangXingHao；SunZuoFeng
School: Shanghai Jiaotong University
Course: Communication and Information System
Keywords: Shot boundary detection DC histogram Chi-square test Machine Learning Macroblock type statistics
CLC: TP391.41
Type: Master's thesis
Year: 2009
Downloads: 25
Quote: 1
Read: Download Dissertation

Abstract


With online video time in recent years the rapid development of the vast amounts of video information retrieval and classification , as the current research in the field of video analysis and processing hot issue . To retrieve the video , the video extracting the relevant information necessary to the video camera into a plurality of basic units of the set , the final extract keyframes and motion information . Therefore, the video shot segmentation , shot boundary detection is the lens that is the most basic analysis of one of the most important elements , but also the essential preparatory work. This paper draws on the work of previous achievements in this area is proposed based on the video I-frame brightness histogram of the DC component of the chi-square (χ2) test and analysis of statistical characteristics of the macroblock type of shot boundary detection algorithm. The algorithm first groups to obtain all the images compressed video GOP (Group of Pictures) in the I-frame inter luminance histogram of the DC component of the chi-square value of the chi-square value indicates an adjacent I- frame difference between frames , frames the greater the difference , indicating that the adjacent I- frames in the GOP shot boundary is more likely to exist , so an algorithm to determine the threshold value is set to the adjacent I- frames whether the shot change . Through the proposed ideas based on machine learning algorithms can be dynamically set the threshold level in the preliminary judging GOP shot change position. As for the type, format and image quality parameters are not the same video segment , the chi-square value of the frame difference threshold different, so a large number of experimental data by the dynamic training threshold , the experimental data to identify the most similar with the actual situation , the accuracy of and recall rates highest threshold as the optimal threshold value . After calculating a GOP may transform each of the lens frame and the previous frame similarity and the similarity difference , by setting an appropriate threshold to determine whether there is conversion lens , the lens is the shear transformation or gradual , and ultimately complete MPEG- 2 video shot boundary detection purposes. Simulation results show that : the algorithm either shear or gradient have a high accuracy rate , and because the algorithm does not MPEG-2 video decompression process , but through the GOP calculated directly , so the shot segmentation , high efficiency, consumption less when calculated is greatly reduced .

Related Dissertations

  1. Macaca mulatta palm pattern morphology,Q954
  2. Based on Data Distribution Characteristics of Text Classification,TP391.1
  3. Research and Implement of Chinese Word Segment Techniques Based on the Conditional Random Field,TP391.1
  4. Semi-supervised Learning and Active Learning of Sentiment Classification Coupled with Domain Knowledge,TP181
  5. A Static Behavior-Based Method to Detect Malware on Android,TP309
  6. Supervision topic model research and application,TP391.1
  7. Distortion effects on image quality evaluation and classification,TP391.41
  8. Learning-based human motion synthesis inverse kinematics,TP391.41
  9. Content-based malicious code detection Research and Implementation,TP393.092
  10. Home Academic Information Extraction System,TP393.092
  11. Social network algorithm and its application in recommendation,TP391.3
  12. Application of Data Mining in Email Anti-Spam System,TP393.098
  13. Based on self-learning social relation extraction research,TP391.1
  14. Graph model based on statistical and machine learning algorithm and its application of certain,TP181
  15. P2P Traffic Identification Method,TP393.06
  16. Based on rough sets and SVM national defense Comprehensive Quality Assessment Methods,E075
  17. SVM Based on SIFT and scene classification,TP391.41
  18. Machine learning based on sparse coding and image content recognition algorithm,TP391.41
  19. Random Forests Feature Selection,TP311.13
  20. Research on Techniques of Collecting Internet Traffic Ground Truth,TP393.06
  21. Research of Protein-Protein Interaction Extraction Based on Rich Feature and Multiple Kernels Learning,Q51

CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer applications > Information processing (information processing) > Pattern Recognition and devices > Image recognition device
© 2012 www.DissertationTopic.Net  Mobile