Yuze Li
ABSTRACT
The Tibetan region contains abundant cultural resources. Handwritten Tibetan numeral recognition is an important basic task for Tibetan language digitalization process. As a low-resource language, Tibetan poses more stringent robustness requirements for deep learning models, requiring them to resist interference and learn core features. This paper proposes a Spiking Fully Convolutional Neural network model SFCN based on spiking neural networks (SNNs) to accomplish handwritten Tibetan numeral recognition on the TibetanMNIST dataset. The fully convolutional neural network structure based on CNNs can effectively extract image features. Combined with SNNs as SFCN, it further enhances network robustness. Experiments show that our SFCN model maintains 98.90% accuracy while exhibiting extraordinary robustness under various interferences, compared to an artificial neural network (ANN) of the same structure. On the basis, combining interferences with training samples further improves model robustness. SFCN trained with adversarial attack method projected gradient descent (PGD) achieves the optimal robustness. SFCN-PGD model’s accuracy on the test set improved by 80.39% at the highest compared to traditional ANNs1.
- Jan Benda and Joel Tabak. 2014. Spike-Frequency Adaptation.Encyclopedia of computational neuroscience 15, 11 (2014), 1–12.Google Scholar
- Jianhao Ding, Tong Bu, Zhaofei Yu, Tiejun Huang, and Jian Liu. 2022. Snn-rat: Robustness-enhanced spiking neural network through regularized adversarial training. Advances in Neural Information Processing Systems 35 (2022), 24780–24793.Google Scholar
- Wulfram Gerstner, Werner M Kistler, Richard Naud, and Liam Paninski. 2014. Neuronal dynamics: From single neurons to networks and models of cognition. Cambridge University Press.Google ScholarDigital Library
- Ian J Goodfellow, Jonathon Shlens, and Christian Szegedy. 2014. Explaining and harnessing adversarial examples. arXiv preprint arXiv:1412.6572 (2014).Google Scholar
- Xuefeng Jiang, Yikun Wang, Wenbo Liu, Shuying Li, and Junrui Liu. 2019. Capsnet, cnn, fcn: Comparative performance evaluation for image classification. Int. J. Mach. Learn. Comput 9, 6 (2019), 840–848.Google ScholarCross Ref
- Tenzin Kaldan and Adiyillam Vijayalakshmi. 2021. TenzinNet for handwritten Tibetan numeral recognition. International Journal of Information Technology 13, 4 (2021), 1679–1682.Google ScholarCross Ref
- Mikhail Leontev, Dmitry Antonov, and Sergey Sukhov. 2021. Robustness of spiking neural networks against adversarial attacks. In 2021 International Conference on Information Technology and Nanotechnology (ITNT). IEEE, 1–6.Google ScholarCross Ref
- Jonathan Long, Evan Shelhamer, and Trevor Darrell. 2015. Fully convolutional networks for semantic segmentation. In Proceedings of the IEEE conference on computer vision and pattern recognition. 3431–3440.Google ScholarCross Ref
- Wolfgang Maass. 1997. Networks of spiking neurons: the third generation of neural network models. Neural networks 10, 9 (1997), 1659–1671.Google Scholar
- Aleksander Mądry, Aleksandar Makelov, Ludwig Schmidt, Dimitris Tsipras, and Adrian Vladu. 2017. Towards deep learning models resistant to adversarial attacks. stat 1050 (2017), 9.Google Scholar
- Nicolas Perez-Nieves, Vincent CH Leung, Pier Luigi Dragotti, and Dan FM Goodman. 2021. Neural heterogeneity promotes robust learning. Nature communications 12, 1 (2021), 5791.Google Scholar
- M Shahidur Rahman 2016. Towards optimal convolutional neural network parameters for bengali handwritten numerals recognition. In 2016 19th International Conference on Computer and Information Technology (ICCIT). IEEE, 431–436.Google Scholar
- Qianjun Shuai, Xingwen Wu, and Libiao Jin. 2019. Handwritten Tibetan Digital Words Recognition Based on improved Convolutional Neural Network. In 2019 IEEE International Conference on Signal, Information and Data Processing (ICSIDP). IEEE, 1–4.Google ScholarCross Ref
- Christian Szegedy, Wojciech Zaremba, Ilya Sutskever, Joan Bruna, Dumitru Erhan, Ian Goodfellow, and Rob Fergus. 2013. Intriguing properties of neural networks. arXiv preprint arXiv:1312.6199 (2013).Google Scholar
- Luke Taylor, Andrew J King, and Nicol Spencer Harper. 2022. Robust and accelerated single-spike spiking neural network training with applicability to challenging temporal tasks. (2022).Google Scholar
- Bing Xu, Naiyan Wang, Tianqi Chen, and Mu Li. 2015. Empirical evaluation of rectified activations in convolutional network. arXiv preprint arXiv:1505.00853 (2015).Google Scholar
- Friedemann Zenke, Sander M Bohté, Claudia Clopath, Iulia M Comşa, Julian Göltz, Wolfgang Maass, Timothée Masquelier, Richard Naud, Emre O Neftci, Mihai A Petrovici, 2021. Visualizing a joint future of neuroscience and neuromorphic engineering. Neuron 109, 4 (2021), 571–575.Google ScholarCross Ref
- Shibo Zhou, Xiaohua Li, Ying Chen, Sanjeev T Chandrasekaran, and Arindam Sanyal. 2021. Temporal-coded deep spiking neural network with easy training and robust performance. In Proceedings of the AAAI conference on artificial intelligence, Vol. 35. 11143–11151.Google ScholarCross Ref
- Zhaokun Zhou, Yuesheng Zhu, Chao He, Yaowei Wang, Shuicheng Yan, Yonghong Tian, and Li Yuan. 2022. Spikformer: When spiking neural network meets transformer. arXiv preprint arXiv:2209.15425 (2022).Google Scholar
Index Terms
- Towards Robust Handwritten Tibetan Numeral Recognition with Spiking Neural Networks
Recommendations
Handwritten numeral recognition through superimposition onto printed form
AbstractHandwriting is, eventually, a variation of the printed forms where the characters are little larger, smaller, angled and deformed than the printed forms. The small changes in handwriting define the parameters of the character to be ...
Multi-task learning network for handwritten numeral recognition
The 6th International Multi-Conference on Engineering and Technology Innovation 2017 (IMETI2017)Handwritten numeral recognition is a challenging problem in the character recognition field due to the large variation in the writing styles of different persons and high similarity in the contour of different numerals. To address this problem, an ...
Multiresolution locally expanded HONN for handwritten numeral recognition
In this paper, we propose a neural network architecture, multiresolution locally expanded high order neural network (MRLHONN) to solve the problem of handwritten numeral recognition. In this recognition scheme, the multiresolution representation of ...
Comments