Dissertation > Excellent graduate degree dissertation topics show

Based on quantum genetic algorithm BP neural network face recognition technology

Author: WuYan
Tutor: QuBo
School: Suzhou University
Course: Electronics and Communication Engineering
Keywords: Face Recognition BP neural network Quantum Genetic Algorithm (QGA) Independent Component Analysis
CLC: TP391.41
Type: Master's thesis
Year: 2008
Downloads: 571
Quote: 1
Read: Download Dissertation

Abstract


In a biometric recognition, face recognition occupies an extremely important position, it is in access control, justice, commerce, and video surveillance has a very wide range of applications. Face recognition is a very active field of computer pattern recognition research topics. After research staff years of effort, this area has been made more results. However, due to the complexity of the problem itself face recognition, to achieve universal application there are many critical issues to be resolved. This paper describes the overview of the development and face recognition face recognition the primary method, and then using the M-FastICA method for face feature extraction, and QGA-BP neural network classifier as face recognition, simulation experiment with a more good recognition results, which show that the method is a feasible approach for face recognition. In the face feature extraction: In this paper, in the overall algebraic feature extraction algorithm PCA and FastICA, results show that using FastICA algorithm for feature extraction of facial feature obtained more efficiently. But for the face recognition for such online data processing, FastICA algorithm shortcomings exist computationally intensive, so, we use the algorithm based on M-FastICA face feature extraction method. The algorithm simplifies the process of Newton Jacobian matrix calculations. Simulation results show that the algorithm not only inherits the FastICA algorithm extracted facial features effective advantages, but compared with the FastICA algorithm can further reduce the number of iterations and convergence time. In the face recognition: In this paper, BP neural network as the face recognition classifier. Focuses on the hidden layer nodes and network convergence speed and recognition rate between the experimental results, the simulation experiment with some recognition effect, but the training time is longer, can not meet face recognition online real-time requirements. To solve these problems, we propose a quantum genetic algorithm with BP neural network connection weights to optimize the results show that the algorithm can greatly reduce the neural network weights search the global optimal solution of the time. Using quantum genetic algorithm to optimize BP neural network connection weights, designed based on quantum genetic algorithm BP neural network classifier. With this classifier on ORL face database for face recognition classification experiment, and achieved an average recognition rate of 95.83%. Experimental results show that the proposed design based on quantum genetic algorithm BP neural network classifier performance than BP neural network classifier. Finally, the paper and quantum genetic algorithm for face recognition are analyzed and discussed.

Related Dissertations

  1. Research on Algorithms of 2D Face Template Protection,TP391.41
  2. Research on Feature Extraction and Classification of Tongue Shape and Tooth-Marked Tongue in TCM Tongue Diagnosis,TP391.41
  3. Research on Visual Servo System of Mechanical ARM,TP242.6
  4. Municipal tourism land use planning environmental impact assessment,X820.3
  5. Study on Taste Characteristic of Taste Peptide Enzymatic Production from Oyster Base on A Neural Network Method,TS254.4
  6. The Research on Evaluation of Living Status Systems of Expressway Relocated People,D523
  7. Mine Risk Information Integration and Intelligent Early Warning,X936
  8. Research of Orange Quality Classification Technology Based on Computer Vision,TP391.41
  9. Research of Video Face Recognition Based on Weighted Voting and Key-Frame Extraction,TP391.41
  10. Face Recognition Method Based on DE,TP391.41
  11. Optimization Study on Gating System and Molding Process Parameters of Injection Mold Based on Simulation,TQ320.662
  12. Research on Face Recognition Methods Based on Flexible Neural Tree,TP391.41
  13. Study on Luohe Technical Supervision Bureau of Food Safety Early Warrning System Based on Neural Network,F203
  14. Research of Adaptive Active Noise Control Based on Neural Network,TP183
  15. Research on Face Recognition Based on AdaBoost Algorithm,TP391.41
  16. Research on Automatic Reading System for Digital Meters,TP391.41
  17. Research on Feature Extraction, Selection and Classification Algorithms for Pulmonary CAD,TP391.41
  18. The Research of Evaluation Method in Connect6 Based on BP-TD Learning,TP18
  19. Research on State Diagnosis on Fan Based on Factor Analysis and BP Neural Network,F426.61
  20. The Research and Design of Converter Steelmaking Endpoint Guiding System,TF345
  21. Analysis on Water Ecological Carrying Capacity of Jiangxi Province,TV213.4

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