Master Thesis

Master Thesis

Master thesis To obtain a Master of Science Degree in Informatics and Communication Systems from the Merseburg University of Applied Sciences Subject: Tunisian truck license plate recognition using an Android Application based on Machine Learning as a detection tool Author: Supervisor: Achraf Boussaada Prof.Dr.-Ing. Rüdiger Klein Matr.-Nr.: 23542 Prof.Dr. Uwe Schröter Table of contents Chapter 1: Introduction ................................................................................................................................. 1 1.1 General Introduction: ................................................................................................................................... 1 1.2 Problem formulation: ................................................................................................................................... 1 1.3 Objective of Study: ........................................................................................................................................ 4 Chapter 2: Analysis ........................................................................................................................................ 4 2.1 Methodological approaches: ........................................................................................................................ 4 2.1.1 Actual approach: ................................................................................................................................... 4 2.1.2 Image Processing with OCR: ................................................................................................................. 6 2.1.3 Chosen approach: ................................................................................................................................. 7 Chapter 3: Artificial Intelligence & Machine Learning ..................................................................................... 8 3.1 Introduction: ................................................................................................................................................. 8 3.2 Types: ........................................................................................................................................................... 8 3.2.1 Supervised learning: ............................................................................................................................. 8 3.2.2 Unsupervised Learning: ........................................................................................................................ 9 3.2.3 Semi-supervised Learning: .................................................................................................................. 10 3.2.4 Reinforcement Learning: .................................................................................................................... 10 3.3 Techniques: ................................................................................................................................................. 12 3.3.1 SVM: ................................................................................................................................................... 12 3.3.2 Random Forest: .................................................................................................................................. 15 3.3.2.1 Decision Tree: ............................................................................................................................. 15 3.3.2.2 Random Forest: ........................................................................................................................... 17 3.3.2.3 Feature Importance: ................................................................................................................... 18 3.3.3 Deep Learning: .................................................................................................................................... 19 3.3.3.1 Deep Learning and Neural Networks: ......................................................................................... 20 3.3.3.2 Perceptron: ................................................................................................................................. 25 3.3.3.3 Feed Forward: ............................................................................................................................. 25 3.3.3.4 Recurrent Neural Network: ........................................................................................................ 25 3.3.3.5 Deep Convolutional Network: .................................................................................................... 26 3.3.4 Convolutional Neural Network: .......................................................................................................... 33 3.3.4.1 The Convolution operation: ........................................................................................................ 33 3.3.4.2 Motivation: ................................................................................................................................. 35 3.3.4.3 Pooling: ....................................................................................................................................... 40 3.3.4.4 Normalization: ............................................................................................................................ 41 3.3.4.5 Random or unsupervised Features: ............................................................................................ 42 3.3.4.6 Regularization: ............................................................................................................................ 42 3.3.4.7 Probability conversion: ............................................................................................................... 43 3.3.5 Transfer Learning: ............................................................................................................................... 46 3.3.5.1 Applications of Transfer Learning: .............................................................................................. 49 3.4 Recapitulation: ........................................................................................................................................... 51 Chapter 4: Implementation .......................................................................................................................... 52 4.1 Software & Tools: ....................................................................................................................................... 52 4.2 Object Detection API:.................................................................................................................................. 54 4.3 Model Training: .......................................................................................................................................... 54 4.4 Mobile implementation: ............................................................................................................................. 58 4.5 Optimization for mobile usage: .................................................................................................................. 62 4.5.1 Minimum Device Requirement: ......................................................................................................... 62 4.5.2 Removing training-only nodes: ........................................................................................................... 62 4.5.3 Recompiling TensorFlow inference library: ........................................................................................ 64 4.5.4 Retrain with mobile data: ................................................................................................................... 65 4.5.5 Reduce model loading time or memory footprint & improve RAM usage: ........................................ 66 4.5.6 Reduce model size: ............................................................................................................................. 67 4.5.7 Exploring Quantized Calculations: ...................................................................................................... 67 4.6 Recapitulation: ........................................................................................................................................... 68 Chapter 5: Summary .................................................................................................................................... 69 5.1 Discussion: .................................................................................................................................................. 69 5.1.1 Results: ............................................................................................................................................... 69 5.1.2 Future Improvements: ........................................................................................................................ 72 5.1.2.1 TensorFlow Lite: .......................................................................................................................... 72 5.1.2.2 Pre-trained models: .................................................................................................................... 74 5.1.2.3 Training and serving with Cloud TPUs: ....................................................................................... 74 5.2 Conclusion: ................................................................................................................................................

View Full Text

Details

  • File Type
    pdf
  • Upload Time
    -
  • Content Languages
    English
  • Upload User
    Anonymous/Not logged-in
  • File Pages
    92 Page
  • File Size
    -

Download

Channel Download Status
Express Download Enable

Copyright

We respect the copyrights and intellectual property rights of all users. All uploaded documents are either original works of the uploader or authorized works of the rightful owners.

  • Not to be reproduced or distributed without explicit permission.
  • Not used for commercial purposes outside of approved use cases.
  • Not used to infringe on the rights of the original creators.
  • If you believe any content infringes your copyright, please contact us immediately.

Support

For help with questions, suggestions, or problems, please contact us