首页--工业技术论文--自动化技术、计算机技术论文--计算技术、计算机技术论文--计算机的应用论文--信息处理(信息加工)论文--模式识别与装置论文

Robust Chinese License Plate Recognition System Using Deep Learning Techniques

ABSTRACT第6页
摘要第7-11页
List of abbreviation/symbols第11-13页
Chapter 1 Introduction第13-18页
    1.1 Background第13-15页
    1.2 Chosen Plate Format and Used Techniques第15-16页
    1.3 Motivation, contribution, and organization of thesis第16-18页
Chapter 2 Literature Review and Key Technologies第18-27页
    2.1 Literature Review第18-20页
    2.2 Key technologies used in our research work第20-23页
        2.2.1 Introduction to Artificial Intelligence (AI)第20-21页
        2.2.2 Introduction to Optical Character Recognition (OCR)第21页
        2.2.3 Introduction to Machine Learning (ML)第21-22页
        2.2.4 Introduction to Deep Learning (DL)第22-23页
        2.2.5 Introduction to Convolutional Neural Networks (CNNs)第23页
    2.3 Software Programs and Tools Used in research work第23-27页
        2.3.1 Python Programming Language第24页
        2.3.2 Tensorflow Deep Learning Library第24-25页
        2.3.3 OpenCV Computer Vision Library第25页
        2.3.4 MATLAB Software Environment第25-26页
        2.3.5 C++ Programming Language第26-27页
Chapter 3 License Plate Recognition System Architectures第27-36页
    3.1 License plate recognition (LPR) system第27-28页
    3.2 LPR systems with Three Components Architecture (3CA)第28-31页
        3.2.1 Preprocessing component第28-29页
        3.2.2 Segmentation component第29-30页
        3.2.3 Character Recognition component第30页
        3.2.4 Advantages of LPR-3CA system第30-31页
        3.2.5 Disadvantages of LPR-3CA system第31页
    3.3 LPR systems with Single Components Architecture (SCA)第31-33页
        3.3.1 Preprocessing component (Optional)第31-32页
        3.3.2 Plate Recognition component第32页
        3.3.3 Advantages of LPR-SCA systems第32页
        3.3.4 Disadvantages of LPR-SCA systems第32-33页
    3.4 Drawbacks of traditional DIP algorithms used in LPR systems第33-35页
    3.5 Advantages of machine learning techniques第35-36页
Chapter 4 Proposed LPR System Components第36-47页
    4.1 Deep learning and CNNs第36-37页
    4.2 Preparing license plate images for segmentation第37页
    4.3 Segmentation CNN第37-42页
        4.3.1 Description of segmentation CNN operation concept第37-38页
        4.3.2 Segmentation CNN structure第38-40页
        4.3.3 Training of segmentation CNN第40-41页
        4.3.4 Advantages of segmentation method第41页
        4.3.5 Disadvantages of segmentation method第41-42页
    4.4 Character recognition CNN第42-44页
        4.4.1 Preparations before character recognition第42页
        4.4.2 Structure of character recognition CNN第42页
        4.4.3 Training of character recognition CNN第42-43页
        4.4.4 Advantages of character recognition method第43-44页
        4.4.5 Disadvantages of character recognition method第44页
    4.5 Structure of whole LPR system第44-47页
        4.5.1 Advantages of proposed LPR system第45-46页
        4.5.2 Disadvantages of whole system第46-47页
Chapter 5 Experimental Results第47-61页
    5.1 Segmentation CNN dataset details第47-48页
    5.2 Character recognition CNN dataset details第48-53页
    5.3 Segmentation CNN experimental results第53-55页
        5.3.1 Evaluating accuracy of the segmentation CNN第53页
        5.3.2 Achieved accuracy第53页
        5.3.3 Performance and size第53-55页
    5.4 Character recognition CNN experimental results第55-56页
        5.4.1 Achieved accuracy第55-56页
        5.4.2 Performance and size第56页
    5.5 Whole LPR system experimental results第56-57页
        5.5.1 Accuracy achieved第56页
        5.5.2 Performance and size第56-57页
    5.6 Comparison results第57-59页
        5.6.1 Training of SCO-CNN model第57页
        5.6.2 Accuracy of SCO-CNN model第57页
        5.6.3 Performance and size of SCO-CNN第57-59页
    5.7 Notes about preprocessing step第59页
    5.8 Analyzing errors reasons in our LPR system第59-61页
        5.8.1 Lack (shortage) of Chinese characters samples第59-60页
        5.8.2 Errors in license plate detection systems第60页
        5.8.3 Lack (shortage) of dataset第60-61页
Chapter 6 Conclusion and Future Work第61-62页
    6.1 Conclusion第61页
    6.2 Future work第61-62页
References第62-66页
Appendix 1: List of Chinese Characters of License Plates and Corresponding Provinces第66-67页

论文共67页,点击 下载论文
上一篇:基于深度学习的公路裂缝识别研究
下一篇:基于深度学习的人脸特征表示与人脸检索算法研究