Deep Learning for Roman Handwritten Character Recognition

The advantage of deep learning is that the analysis and learning of massive amounts of unsupervised data make it a beneficial tool for Big Data analysis. Convolution Neural Network (CNN) is a deep learning method that can be used to classify image, cluster them by similarity, and perform image recognition in the scene. This paper conducts a comparative study between three deep learning models, which are simple-CNN, AlexNet and GoogLeNet for Roman handwritten character recognition using Chars74K dataset. The produced results indicate that GooleNet achieves the best accuracy but it requires a longer time to achieve such result while AlexNet produces less accurate result but at a faster rate.

[1]  Graham Kendall,et al.  Document zone content classification for technical document images using Artificial Neural Networks and Support Vector Machines , 2009, 2009 Second International Conference on the Applications of Digital Information and Web Technologies.

[2]  Himanshu Sharma,et al.  A Deep Learning Hybrid CNN Framework Approach for Vegetation Cover Mapping Using Deep Features , 2017, 2017 13th International Conference on Signal-Image Technology & Internet-Based Systems (SITIS).

[3]  Shuzlina Abdul Rahman,et al.  Towards emotional control recognition through handwriting using fuzzy inference , 2008, 2008 International Symposium on Information Technology.

[4]  D. Singh,et al.  Handwritten English Character Recognition Using Neural Network , 2010 .

[5]  Chris Rowen,et al.  Using Convolutional Neural Networks for Image Recognition By , 2015 .

[6]  Marc Chaumont,et al.  Camera model identification with the use of deep convolutional neural networks , 2016, 2016 IEEE International Workshop on Information Forensics and Security (WIFS).

[7]  M. N. Masrek,et al.  Comparison of Machine Learning algorithms performance in detecting network intrusion , 2010, 2010 International Conference on Networking and Information Technology.

[8]  Gong Cheng,et al.  Scene classification of high resolution remote sensing images using convolutional neural networks , 2016, 2016 IEEE International Geoscience and Remote Sensing Symposium (IGARSS).

[9]  Ricardo Matsumura de Araújo,et al.  On the Performance of GoogLeNet and AlexNet Applied to Sketches , 2016, AAAI.