Improving the Perceptual Quality of Document Images Using Deep Neural Network

Pandey, RK and Ramakrishnan, AG (2019) Improving the Perceptual Quality of Document Images Using Deep Neural Network. In: 16th International Symposium on Neural Networks, ISNN 2019, 10 - 12 July 2019, Moscow, pp. 448-459.

PDF ISNN_2019.pdf - Published Version Restricted to Registered users only Download (1MB) | Request a copy

Abstract

Given a low-resolution binary document image, we aim to improve its perceptual quality for enhanced readability. We have proposed a simple, deep learning based model, that uses convolution with transposed convolution and sub-pixel layers in the best possible way to construct the high-resolution image. The proposed architecture scales across the three different scripts tested, namely Tamil, Kannada and Roman. To show that the reconstructed output has enhanced readability, we have used the objective criterion of optical character recognizer (OCR) character level accuracy. The reported results by our CTCS architecture shows significant improvement in terms of the subjective criterion of human readability and objective criterion of OCR character level accuracy. © Springer Nature Switzerland AG 2019.

Item Type:	Conference Paper
Publication:	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Publisher:	Springer Verlag
Additional Information:	The copyright for this article belongs to Springer Verlag.
Keywords:	Binary images; Convolution; Deep learning; Deep neural networks; Network architecture; Optical character recognition, Binary document image; High resolution image; Learning Based Models; Objective criteria; Perceptual quality; Proposed architectures; Readability; Super resolution, Image enhancement
Department/Centre:	Division of Electrical Sciences > Electrical Engineering
Date Deposited:	07 Dec 2022 07:15
Last Modified:	07 Dec 2022 07:15
URI:	https://eprints.iisc.ac.in/id/eprint/78292

Actions (login required)

View Item