声明
CHAPTER 1:INTRODUCTION
1.1 Motivation
1.2 Recent Achievements
1.3 Main Contribution
1.4 Thesis Structure
CHAPTER 2:RELATED WORK
2.1 From Perceptron to Deep Neural Networks
2.2 Opening the“Black Box”
2.3 Offline Handwritten Chinese Character Recognition
CHAPTER 3:PROPOSED METHODOLOGY
3.1 Melnyk-Net:A High-Performance CNN for Offline HCCR
3.2 Obtaining Class Activation Maps
CHAPTER 4:EXPERIMENTS AND RESULTS
4.1 Datasets
4.2 Training Strategy
4.3 Results
4.4 Comparison with Other Methods
4.5 Will it work on MNIST?
CONCLUSION AND FUTURE WORK
参考文献
APPENDIX A:LIST OF PUBICATIONS
致谢
湖南大学;