基于深度学习的刚果假钞检测

声明

致谢

Preface

1 Introduction

1.1 Background

1.2 Motivation

1.3 Problem Statement

1.4 Objective of this thesis

1.5 Organization of this thesis

2 Overview

2.1 Counterfeit Detection Pen

2.2 Ultraviolet Light/Scanner

2.3 Image processing

2.3.1 Colour Model

2.3.2 RGB (RED, GREEN, BLUE)

2.3.3 HSV (HUE,SATURATION,VALUE)

2.3.4 Colour Space

2.3.5 Colour Scheme

2.3.6 Image Segmentation

2.4 CNN/ANN

2.4.1 Convolutional neural network (CNN)

2.4.2 Artificial Neural Network (ANN)

2.4.3 Similarities and Difference between CNN and ANN

2.5 Related Work

2.6 Summary

3 Contribution

3.1 The Architecture for Banknote Classification

3.2 Image acquisition

4 Experiment

4.1 Image Acquisition and Preprocessing

4.1.1 Image Acquisition

4.1.2 Pre-processing

4.1.3 Region of Interest retrieval

4.2 Feature Extraction, Result and Analysis

4.2.1 Feature Extraction

4.2.2 Feature obtained using CNN model

4.2.3 The Architecture of CNN

4.3 Banknote Classification and Verification Result and Analysis

4.4 Summary

5 Conclusions

Strength of the study

Scope and limitation of the study

参考文献

Author Profile and Research Achievements Obtained during the Study for A Master’s Degree

Declaration of Originality

Dataset for the Master’s Thesis