Dissertation > Excellent graduate degree dissertation topics show

Research on Multi-Person Behavior Recognition and Analysis Based on Sitting Posture

Author: WuSongLin
Tutor: CuiRongYi
School: Yanbian University
Course: Applied Computer Technology
Keywords: Behavior analysis Sitting behavior recognition Moving target detection Skin color model Principal Component Analysis Face Skin Color Detection
CLC: TP391.41
Type: Master's thesis
Year: 2010
Downloads: 136
Quote: 1
Read: Download Dissertation

Abstract


Based on the visual analysis of human motion is one of the most active research topics in the field of computer vision, its core is detected from image sequence using computer vision technology to track and identify people and their behavior analysis and understanding. In recent years, the visual analysis of human motion by the extensive attention of academia and the business community at home and abroad, but as a hot and difficult in the field of computer vision, there is still a lot to be resolved theoretical and technical issues. Human behavior analysis for single and action between the larger span, the action itself is relatively simple, and study less than two interact and crowd behavior analysis, and behavior analysis also does not have a system theory and framework. Moving target detection algorithm, as a behavior analysis-based comparison due to the impact of the light changes, such as noise, resulting in behavior analysis of the result is not satisfactory. The thesis put forward posture behavior recognition method based on PCA (Principal Component Analysis), and for the identification and analysis of behavior in many environments; area boundary distance according to the adjacent face complexion judge the two interact behavior is normal or not. First of all, in a single environment, Abatement Act through the background contour extracted moving object silhouette, filled with white pixels moving object silhouette, and the use of the color area of ??the skin color model to extract moving targets. Only the video part of the frame as a whole, to the skin color region due to moving target color scale normalized such that the skin color area is full of the current video frame. Secondly, the behavior recognition method based on PCA sitting. The method utilizes posture behavior training samples constructed PCA feature space, the new behavior sample color area is projected to the feature space, and take advantage of the the cosine classifier complete identification of new posture. Meanwhile, the changes over time based on the skin color region pixel analysis of the human face orientation changes. Finally, in a multiplayer environment, the distance between the adjacent skin color region boundary analysis of the behavior of the interaction between the two. When the distance from the boundary of the skin color area of ??two adjacent faces does not exceed a pre-set threshold value, the direct determination of the two interactive behavior is abnormal; Otherwise according to the size of the skin color region of the interception of each person's body area, through a single the posture behavior recognition method based on PCA human environment to achieve the identification and analysis of individual behavior. Simulation results show that the method of the background contour abatement and skin color detection combined to provide excellent behavior analysis raw data, the light changes, shadows and other noise has good robustness. In a single environment, based on the PCA sitting behavior recognition accuracy of 85.15%. Behavior analysis method based on the adjacent skin color area boundary distance multiplayer multiplayer environment, better able to analyze the interaction between the two acts.

Related Dissertations

  1. Application of Improved Principal Component Analysis Algorithm in Course Construction,G642.4
  2. Research on Hand Tracking and Application Platform for Hand Gesture Recognition,TP391.4
  3. Research of Diagnosing Cucumber Diseases Based on Hyperspectral Imaging,S436.421
  4. The Impact of Tourism on Typical Vegetation in Luya Mountain Nature Reserve, Shanxi Province,S759.9
  5. Macaca mulatta palm morphological study of pattern ridge count,Q954
  6. Zhaoguan Lower Coal Group water inrush prediction and control techniques,TD745
  7. Research on Cultural Industrial Competitiveness of Chong Qing,F224
  8. Design and Implementation of Embedded Multi-parameter Intelligent Evironment Monitoring Systems,TP274
  9. The Research of Prairie Road Light Environment Effects on Physiological Indicators of Drivers,U491.254
  10. Research on Feature Extraction, Selection and Classification Algorithms for Pulmonary CAD,TP391.41
  11. Research on Moving Objects Detection Algorithms in Security Monitoring System,TP391.41
  12. Research Onamethod for Human Face Recognition Based on MMTD,TP391.41
  13. The Study of Moving Object Detection and Tracking Algorithm Based on Image Information,TP391.41
  14. Research of Moving Object Detection and Tracking Technology,TP391.41
  15. Research of Dynamic Association Rules,TP311.13
  16. Competitiveness’ Evaluation and Development Strategy of the Tourism Industry in Huanggang City,F592.7
  17. Research on Rural Information Promoting Urban-rural Integration in Jiangsu Southern Area,F127;F224
  18. The Research and Implementation of An Instant Messenger Monitoring System Based on DPI,TN915.09
  19. The Design of Information System of Power Equipment Current-carrying Faults Diagnosis,TP311.52
  20. The Research of E-government System Performance Evaluation Index System in Linyi City Based on the Principal Component Analysis,G206
  21. Analysis of a new bypass attack and defense policy studies,TP309

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