Dissertation > Excellent graduate degree dissertation topics show

Research and Application of video object detection and tracking methods

Author: ChenBo
Tutor: LiuGuiSong
School: University of Electronic Science and Technology
Course: Computer Software and Theory
Keywords: object detection object tracking AdaBoost MeanShift
CLC: TP391.41
Type: Master's thesis
Year: 2011
Downloads: 122
Quote: 1
Read: Download Dissertation

Abstract


Moving object detection and tracking in video streams is an essential element in various computer vision applications, including motion-based object recognition, human-computer interaction, automatic vehicle navigation, and intelligent video surveillance. In the actual application of object detection and tracking, the accuracy and stability of detection and tracking depend on the design of algorithms to a great extent. The thesis is focused on the algorithms of moving object detection and tracking.Firstly, the pre-processing methods, the moving object detection methods, and the moving object tracking methods, which are commonly used in digital image processing, are studied. Then, the questions of eliminating shadow and occlusion processing are further studied.Secondly, by studying the AdaBoost algorithm and its advantage in classification, and analyzing the mean shift algorithm and its advantage in object tracking, a new object tracking algorithm, which integrate AdaBoost and mean-shift technology, is proposed in this work. The algorithm consists of a learning stage and an estimation stage. The learning stage selects the features for tracking, and the estimation stage composes a likelihood image and applies the mean shift algorithm to it to track an object. The tracking performance depends on the quality of the likelihood image. We apply two schemes to generate and integrate likelihood images, including the one based on the discrete AdaBoost (DAB) and the other based on the real AdaBoost (RAB). The DAB scheme uses tuned feature values, whereas RAB estimates class probabilities, to select the features and generate the likelihood images.Finally, in order to verify the feasibility and advantage of the algorithms and analyze the correlation performance of the detection and tracking methods, we designed and implemented an experimental platform, which is extensible and can be used to verify algorithms of detection and tracking. Through the experimental platform, we can easily select the detection and tracking methods that have been implemented and set their parameters, thus we provide an excellent analysis system for the verification and analysis of detection and tracking algorithms. We compared the performance of the proposed algorithm with the conventional mean shift tracking algorithm through the experimental platform in this thesis. The results show that by using AdaBoost method, the independence among selected features can be improved and the higher quality likelihood images can also be constructed.

Related Dissertations

  1. Research on the Classification Based on the Reconstruction of Solder Joint,TP391.41
  2. Tongue Feature Extraction and Research of Fusion Classification,TP391.41
  3. The Fatigue State Recognition of the Driver Based on Eye Detection,TP391.41
  4. Research on Visual Measurement for Spacecraft Rendezvous and Approach,TP391.41
  5. Feature Extraction, Selection and Combination in Lipreading,TP391.41
  6. The Research on Paper Currency Classification Method Based on Harr-Like Feature and Minimal Ball Including Samples,TP391.41
  7. The Research of Moving Object Tracking System Based on Embeded Image Process Unit,TP391.41
  8. The Design and Implementation of Face Detection Algorithm Based on FPGA Chip,TP391.41
  9. Research of Facial Expression Recognition Algorithm,TP391.41
  10. Study on Visual Target Detection Based on SVM,TP391.41
  11. Research on Visual Detection and Tracking of Mobile Robots,TP242.62
  12. Research on Detecting and Tracking the Moving Objects,TP391.41
  13. Research on Moving Targets Detection and Tracking in Video Sequences,TP391.41
  14. Vehicles in traffic video detection and tracking technology research and,TP391.41
  15. Mathematical morphology combined with remote sensing image invariant moments Target Detection Method,TP751
  16. Research and Implementation of Face Detection System Based on Android,TP391.41
  17. Research on the Edit Propagation for Certain Types of Images,TP391.41
  18. Research and Implementation of Human Face Detection Based on DaVinci Platform,TP391.41
  19. Research on moving target detection and tracking method,TP391.41
  20. And Cascade algorithm based on AdaBoost face detection,TP391.41
  21. The Interactive Music Play Table,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