Dissertation > Excellent graduate degree dissertation topics show

License Plate Recognition Algorithm Based on Wavelet Packet Analysis and BP Neural Networks

Author: ZhouXiShou
Tutor: ChenTianXing
School: Southwest Jiaotong University
Course: Mechanical Design and Theory
Keywords: Wavelet packet transform BP neural network License Plate Location Character Recognition
CLC: TP391.41
Type: Master's thesis
Year: 2010
Downloads: 283
Quote: 6
Read: Download Dissertation

Abstract


Automatic license plate recognition technology is a very important technology in modern intelligent transportation systems, the research focus in recent years. Intelligent transportation systems in the vehicle tracking, automated highway toll collection, parking automatic billing and urban traffic counts plays an extremely important role, while the license is the only sign of the vehicle, so the automatic license plate recognition technology is at the core of the entire system the status of research significance. Researchers at home and abroad has begun on automatic license plate recognition technology to carry out in-depth research, put forward a number of algorithms and programs, while some products have been put into use, but because of the effect of failing to meet the requirements of people expected, from the real practical and General requirements, there is a gap. Therefore, how to improve the correctness of the license plate location and recognition algorithms and real-time, there is also a larger space. In this paper, on the basis of inherited the results of previous studies, the advanced theoretical tools wavelet packet transform applied to the vehicle license plate recognition, mainly working in the following areas: (1) The license plate positioning; (2) character segmentation; (3) character recognition. License plate location, license plate image gray-scale transformation, smoothing, by analyzing the traditional image enhancement method is proposed based on wavelet packet transform enhancement algorithms; combination of edge detection, morphological processing, projection method algorithm license plate positioning algorithm is proposed based on the license plate texture features. The experiments show that the algorithm achieved the desired results, the accuracy of the extracted license plate more than 92.9% from the original image. Character segmentation, multi-resolution wavelet packet transform, license plate image denoising, and then use the method of combining vertical projection and prior knowledge of a single character region segmentation. The accuracy and anti-jamming algorithm is much better than the traditional projection algorithm. Character recognition, the first analysis of the two commonly used method of character recognition - template matching and neural network; proposed as a neural network input vector feature extraction using wavelet packet energy vector, variable learning rate momentum BP neural network for character recognition, the BP neural network convergence, speed training. Proved by experiments that this whole program is feasible and effective in its recognition rate, speed and other aspects of wavelet packet analysis and BP neural network-based license plate recognition technology has unique advantages and broad prospects.

Related Dissertations

  1. Research on Feature Extraction and Classification of Tongue Shape and Tooth-Marked Tongue in TCM Tongue Diagnosis,TP391.41
  2. Research on Visual Servo System of Mechanical ARM,TP242.6
  3. Municipal tourism land use planning environmental impact assessment,X820.3
  4. Study on Taste Characteristic of Taste Peptide Enzymatic Production from Oyster Base on A Neural Network Method,TS254.4
  5. The Research on Evaluation of Living Status Systems of Expressway Relocated People,D523
  6. Mine Risk Information Integration and Intelligent Early Warning,X936
  7. Research of Orange Quality Classification Technology Based on Computer Vision,TP391.41
  8. Optimization Study on Gating System and Molding Process Parameters of Injection Mold Based on Simulation,TQ320.662
  9. Research on Feature Extraction, Selection and Classification Algorithms for Pulmonary CAD,TP391.41
  10. Research on State Diagnosis on Fan Based on Factor Analysis and BP Neural Network,F426.61
  11. Study on Agricultural Machinery Injury in North of China and Related Risk Factors,R139
  12. Doppler weather radar based windshear prediction,P415.2
  13. 2205 hot workability and mechanical properties of stainless steel clad plate and Prediction,TG142.71
  14. Number of feature-based recognition of the banknote,TP391.41
  15. Data mining and decision support technology applied research in hospitals,TP311.13
  16. HD image license plate location based vehicle tracking and violation detection,TP391.41
  17. Dynamic Simulation and Prediction of Urban Sprawl Based on BP Neural Network,P208
  18. Research on the Inspection and the Application of GPS Kenimatic Surveying Technology in the Project of FAST,P228.4
  19. The Study about the Optimization of Insulation Structure and Grading Ring of Tap Changer,TM403.3
  20. Automatic license plate location and character segmentation,TP391.41
  21. Study and Implementation of Multi-License-Plate Recognition System under Complex Background,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