End to End System for Handwritten Text Recognition and Plagiarism Detection using CNN & BLSTM

Abstract

In recent times, plagiarism of handwritten assignments has been rampant. Deep Learning models such as Convolutional Neural Networks have proven to be resourceful for recognition tasks in computer vision. Additionally, sequential models like LSTMs have been useful for handling cursive handwriting and variations in styles of writing. In this paper, we propose an end-to-end system that performs recognition of handwritten text assignment documents and gives the similarity scores between them. This system is a conjunction of a page-to-word segmentation algorithm, a convolutional neural network (CNN) and bi-directional long short-term memory (BLSTM) network for recognition, and a plagiarism detection module.