A comparative Study of Handwritten Devanagari Script Character Recognition Techniques

In the discipline of pattern recognition, optical character recognition is a critical task. A significant amount of research has been done on character recognition in the English language but in the Indian context, the research has been limited. Devanagari is a commonly used Indian script that is the foundation of languages like Hindi, Sanskrit, Kashmiri, and Marathi. Several researchers have published their work on this topic in recent years with some promising results. To expand upon the existing work and to provide a benchmark for future studies, a comparative study of four different classifiers and two different feature extraction techniques have been proposed in this paper. Multi-Layer Perceptron, K-Nearest Neighbor, Support Vector Machine, and Random Forest algorithms are used as classifiers whereas Convolutional Neural Network and Histogram of Oriented Gradients are used as feature extraction techniques.