Dissertation > Excellent graduate degree dissertation topics show

Research on Surveillance Video Objects Tracking Algorithm

Author: QiPeiQing
Tutor: WuXiaoPei
School: Anhui University
Course: Applied Computer Technology
Keywords: object tracking Meanshift particle filer incremental learning Bi-2DPCA
CLC: TP391.41
Type: Master's thesis
Year: 2013
Downloads: 29
Quote: 0
Read: Download Dissertation

Abstract


Object tracking is one of the most important research topics in computer vision, with wide application in many fields such as public safety, medical diagnosis and digital multimedia. So far, there are still large quantities of challenges in the research of object tracking. Changing appearance and scene, object-to-object occlusion, object-to-scene occlusion and camera motion are all likely to result in failure of tracking. Therefore, it is of great significance to further the research of the topic. Fusion of multi algorithms and improvement to a certain algorithm can enhance the real-time feature and stability of tracking, which are hot in study of object tracking. In this thesis, based on the fruits of previous research and experiments, an adaptive object tracking algorithm has been proposed. Moverover, an object tracking module has been implemented as a crucial part of a video surveillance system. The main work of this thesis are as follows:1. Studied several existed object tracking algorithms including Meanshift based tracking, particle filter based tracking and fusion of Meanshift and particle filter based tracking. The basic principle and implementation frame of each algorithm have been described in very detail. Tracking experiments on several image and video sequences have been carried out to determine their advantages and disadvantages.2. Proposed an object tracking algorithm based on bidirectional two-dimensional principle component analysis (Bi-2DPCA). Object representation based on Bi-2DPCA has been used to generate the object image subspace. An incremental factor according to image match degree has been adopted to update the related mean matrix and covariance matrices. The comparative experiments on classical image sequences containing dynamic backgrounds have been carried out, the results showed the proposed algorithm is capable of tracking object accurately even in case of partial occlusion, and more efficient than the algorithm based on two-dimensional principle component analysis. 3. Designed and implemented an object tacking module of a video surveillance system based on fusion of particle filter and Meanshift. To decrease the cost of the module, an object blob labeling method based on the first four key pixels has been proposed. To improve the timeliness, stability and usability of the system, a module logic design based on blob labeling and tracking has been implemented.

Related Dissertations

  1. Research on Visual Measurement for Spacecraft Rendezvous and Approach,TP391.41
  2. The Research of Moving Object Tracking System Based on Embeded Image Process Unit,TP391.41
  3. Research on Visual Detection and Tracking of Mobile Robots,TP242.62
  4. Research on the Algorithms of Moving Object Detection and Tracking,TP391.41
  5. Study on Applications of Soft Sensor in Chloromethane Recovery,TQ222.214
  6. Object Detection and Tracking in Video Image,TP391.41
  7. Moving Object Detection and Tracking in Dynamic Scene,TP391.41
  8. The Research and Implementation of the Fault Management System for Triple Play,TP315
  9. Tracking Based on Sub-Block Color Histogram,Hog and Particle Filter,TP391.41
  10. The Study of College Digital Campus Web Information Filtering,TP393.09
  11. Moving Human Object Tracking and Abnormal Behavior Recognition,TP391.41
  12. The Research of Support Vector Machine Incremental Learning Algorithms,TP18
  13. Study on Algorithms of Palmprint Recognition,TP391.41
  14. Key Technologies of WEB Media Resources Intelligent Discovery,TP393.092
  15. The Interactive Music Play Table,TP391.41
  16. Based on incremental improvements in the Bayesian question classification in the field of research,TP391.3
  17. Detection of the Moving Object and Face Tracking Based on Serial Images,TP391.41
  18. Research on Identification Methods of P2P-Streaming Media Traffic Based on Incremental SVM Learning,TN919.8
  19. Research on Multiple Objects Tracking Based on Detection in Video Sequence,TP391.41
  20. Research on Target Tracking Based on Video Image Sequence,TP391.41
  21. Design and implementation of intelligent monitoring system based on face detection and tracking,TP391.41

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