Dissertation > Excellent graduate degree dissertation topics show

Intelligent Video Surveillance Target Detection Technology

Author: ChenJingDong
Tutor: SangNong
School: Huazhong University of Science and Technology
Course: Pattern Recognition and Intelligent Systems
Keywords: Offline learning Background Modeling Target Detection Cascade classifier Support Vector Machine
CLC: TP391.41
Type: Master's thesis
Year: 2011
Downloads: 105
Quote: 0
Read: Download Dissertation

Abstract


Images and video of the target detection of computer vision research focus, but also intelligent video surveillance system in key technologies. This goal is achieved by common video surveillance environments target detection algorithm, to obtain a specific environment in real time, accurate target detection algorithm for intelligent video surveillance system upper layer applications provide an important foundation support. This paper consists of two modules: (1) Based off machine learning target detection section. Based on machine learning target detection common practice is through certain target a large number of samples to learn to get against such targets classifier, and then use this classifier in the image and video detection of such targets. Machine learning for target detection process is more similar to humans in the external environment understanding and awareness of the process. In this section, the main research offline learning conditions of target detection algorithms. Briefly two features (Haar and HOG) calculation method introduced support vector machine (SVM) and Adaboost learning algorithm is the basic principle. Through experiments detailed analysis Haar, HOG feature extraction algorithm and SVM, Adaboost learning algorithm in a specific surveillance environment the best combination of problems, the design of the monitoring environment, fast, accurate target detection system. (2) based on background modeling target detection section. In this section we briefly introduce Gaussian mixture model (GMM), codebook (Codebook) and texture-based background modeling (LBP) algorithm is the basic principle, through experimental comparative analysis of its application environment and its advantages and disadvantages. Here combines Gaussian mixture model (GMM) and codebook (Codebook) background model algorithm framework, proposes an improved background modeling algorithm. This algorithm can be as GMM as accurately estimated pixel point sampling probability distribution, and can be like Codebook same algorithm has less experience parameters and efficient real-time calculation of performance. The traditional LBP coding noise by larger changes, LBP-encoded on each of the should equal weight considerations, changes LBP coding method, remove the bit-weighted this step, the direct use of the binary encoding of image statistics . Proposed based on Hamming distance measure the histogram method, in addition to artificial binary code weighting steps to improve the statistical histogram interference performance, thereby to obtain a more stable matching performance. Finally, the text do summarized. Proposed offline learning and background modeling calculation phase combines design automatic target detection system of thought, in order to reduce human-annotated samples of work, while improving target detection system wide adaptability.

Related Dissertations

  1. Research on Automatic Detection Algorithm for Substructure Distress of Highway Pavement Based on SVM,U418.6
  2. Research on Autamatic Music Structrue Analysis,TN912.3
  3. Research on Joint Target Detection for Dual-Sensor Image and System Implementation,TP391.41
  4. Research on Transductive Support Vector Machine and Its Application in Image Retrieval,TP391.41
  5. Fault Diagnosis Method Based on Support Vector Machine,TP18
  6. Process Support Vector Machine and Its Application to Satellite Thermal Equilibrium Temperature Prediction,TP183
  7. Research for Infrared Image Target Identification and Tracking Technology,TP391.41
  8. Study on the Road Condition Monitoring Based on Vehicular 3D Acceleration Sensor,TP274
  9. Research of Diagnosing Cucumber Diseases Based on Hyperspectral Imaging,S436.421
  10. Moving target trajectory analysis based Intelligent Traffic Monitoring System,TP277
  11. Research and Application on Detection and Tracking of Object Based on Video Stream,TP391.41
  12. The Research on Intrusion Detection System Based on Machine Learning,TP393.08
  13. Research on Improved K Neighbor Support Vector Machine Algorithm Faced Text Classification,TP391.1
  14. Design and Implementation of Embedded Multi-parameter Intelligent Evironment Monitoring Systems,TP274
  15. Research on Face Recognition Based on AdaBoost Algorithm,TP391.41
  16. Research on Feature Extraction, Selection and Classification Algorithms for Pulmonary CAD,TP391.41
  17. Research on Subimage Selection and Mathching Method for Synthetic Aperture Radar(SAR) Target Recognition,TN957.52
  18. Research on Moving Object Retrieval for Video Surveillance,TP391.41
  19. Research on Moving Objects Detection Algorithms in Security Monitoring System,TP391.41
  20. Research on Specific Target Recognition Algorithms in Video Surveillance,TP391.41
  21. Research of Visual Attention Model for Complex Object,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