Improved Recognition of Hand-Written Digits Using Convolutional Neural Network (CNN)
Keywords:
Natural Language Processing, Hand-Written Digit Recognition, Convolutional Neural Network, Adam Optimizer, SGDM OptimizerAbstract
Traditional handwriting recognition systems depend on various features and prior knowledge making its practical implementation more challenging. Handwriting recognition system has focused deep learning algorithms in recent times and achieved good performance accuracy but still, needs to be improved in terms of recognition accuracy due to rapid growth of data and massive computations. Deep learning is remarkably being employed in various fields because of diverse range of applications. In this paper, we have employed CNN (Convolutional Neural Network) with two different optimizers and five different epoch number to observe the variations of accuracy for classifying different handwritten digits. We have trained seven-layered CNN with Adam and SGDM optimizer on MNIST (Modified National Institute of Standards and Technology) dataset and found that for maximum number of iterations, Adam optimizer outperform the SGDM optimizer in terms of accuracy which is found to be 99.50%, respectively.
References
Xiao, J., Zhu, X., Huang, C., Yang, X., Wen, F., & Zhong, M. (2019). A new approach for stock price analysis and prediction based on SSA and SVM. International Journal of Information Technology & Decision Making, 18(01), 287-310.
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017). ImageNet classification with deep convolutional neural networks. Communications of the ACM, 60(6), 84-90.
Lee, S., Son, K., Kim, H., & Park, J. (2017, July). Car plate recognition based on CNN using embedded system with GPU. In 2017 10th International Conference on Human System Interactions (HSI) (pp. 239-241). IEEE.
Wu, Y. C., Yin, F., & Liu, C. L. (2017). Improving handwritten Chinese text recognition using neural network language models and convolutional neural network shape models. Pattern Recognition, 65, 251-264.
Cohen, G., Afshar, S., Tapson, J., & Van Schaik, A. (2017, May). EMNIST: Extending MNIST to handwritten letters. In 2017 International Joint Conference on Neural Networks (IJCNN) (pp. 2921-2926). IEEE.
Sarkhel, R., Das, N., Das, A., Kundu, M., & Nasipuri, M. (2017). A multi-scale deep quad tree based feature extraction method for the recognition of isolated handwritten characters of popular indic scripts. Pattern Recognition, 71, 78-93.
Boufenar, C., Kerboua, A., & Batouche, M. (2018). Investigation on deep learning for off-line handwritten Arabic character recognition. Cognitive Systems Research, 50, 180-195.
Kavitha, B. R., & Srimathi, C. (2019). Benchmarking on offline Handwritten Tamil Character Recognition using convolutional neural networks. Journal of King Saud University-Computer and Information Sciences.
Husnain, M., Saad Missen, M. M., Mumtaz, S., Jhanidr, M. Z., Coustaty, M., Muzzamil Luqman, M., ... & Sang Choi, G. (2019). Recognition of Urdu handwritten characters using convolutional neural network. Applied Sciences, 9(13), 2758.
Tavanaei, A., & Maida, A. S. (2017, May). Multi-layer unsupervised learning in a spiking convolutional neural network. In 2017 international joint conference on neural networks (IJCNN) (pp. 2023-2030). IEEE.
Jin, J., Fu, K., & Zhang, C. (2014). Traffic sign recognition with hinge loss trained convolutional neural networks. IEEE Transactions on Intelligent Transportation Systems, 15(5), 1991-2000.
Tabik, S., Peralta, D., Herrera-Poyatos, A., & Herrera, F. (2017). A snapshot of image pre-processing for convolutional neural networks: case study of MNIST. International Journal of Computational Intelligence Systems, 10(1), 555-568.
Siddique, M. A. B., Khan, M. M. R., Arif, R. B., & Ashrafi, Z. (2018, September). Study and observation of the variations of accuracies for handwritten digits recognition with various hidden layers and epochs using neural network algorithm. In 2018 4th International Conference on Electrical Engineering and Information & Communication Technology (iCEEiCT) (pp. 118-123). IEEE.
Enriquez, E. A., Gordillo, N., Bergasa, L. M., Romera, E., & Huélamo, C. G. (2018, November). Convolutional neural network vs traditional methods for offline recognition of handwritten digits. In Workshop of Physical Agents (pp. 87-99). Springer, Cham.
Lopez, B., Nguyen, M. A., & Walia, A. (2019). Modified MNIST.
Nguyen, C. T., Khuong, V. T. M., Nguyen, H. T., & Nakagawa, M. (2020). CNN based spatial classification features for clustering offline handwritten mathematical expressions. Pattern Recognition Letters, 131, 113-120.
Zhao, H. H., & Liu, H. (2020). Multiple classifiers fusion and CNN feature extraction for handwritten digits recognition. Granular Computing, 5(3), 411-418.
Ge, D. Y., Yao, X. F., Xiang, W. J., Wen, X. J., & Liu, E. C. (2019, October). Design of high accuracy detector for MNIST handwritten digit recognition based on convolutional neural network. In 2019 12th International Conference on Intelligent Computation Technology and Automation (ICICTA) (pp. 658-662). IEEE.
Younis, K. S., & Alkhateeb, A. A. (2017). A new implementation of deep neural networks for optical character recognition and face recognition. Proceedings of the new trends in information technology, 157-162.
Jana, R., & Bhattacharyya, S. (2019). Character recognition from handwritten image using convolutional neural networks. In Recent Trends in Signal and Image Processing (pp. 23-30). Springer, Singapore.
Hossain, M. A., & Ali, M. M. (2019). Recognition of handwritten digit using convolutional neural network (CNN). Global Journal of Computer Science and Technology.
Ahlawat, S., Choudhary, A., Nayyar, A., Singh, S., & Yoon, B. (2020). Improved handwritten digit recognition using convolutional neural networks (CNN). Sensors, 20(12), 3344.
https://www.kaggle.com/oddrationale/mnist-in-csv
Murphy, K. P. (2012). Machine learning: a probabilistic perspective. MIT press.
Ali, S., Shaukat, Z., Azeem, M., Sakhawat, Z., Mahmood, T., & ur Rehman, K. (2019). An efficient and improved scheme for handwritten digit recognition based on convolutional neural network. SN Applied Sciences, 1(9), 1-9.