Handwritten Character Recognition Using Directed Acyclic Graph

In this paper, a Directed Acyclic Graph (DAG) neural network is proposed for the recognition of handwritten Manipuri characters. Handwritten character recognition is one of the exciting and challenging research areas for pattern recognition. A particular method may work on some datasets; however, there is no absolute solution to this problem, and researchers are still finding solutions, especially for regional languages. It may be due to unavailability of the standard dataset, compounding characters, presence of similar-looking symbols, or curvature of characters in these languages. The objective of this paper is to recognize handwritten Meitei Mayek (Manipuri character) using Directed Acyclic Graph (DAG) deep neural network.