Task-driven single-image super-resolution reconstruction of document scans

TL;DR

This paper explores using task-driven deep learning-based super-resolution as a preprocessing step to enhance optical character recognition accuracy on low-resolution document scans.

Contribution

It introduces a multi-task loss function combining text detection and image similarity to improve super-resolution for document images.

Findings

Super-resolution improves OCR accuracy on low-resolution scans.

Task-driven training enhances super-resolution relevance for text recognition.

Encouraging results support real-world application of document super-resolution.

Abstract

Super-resolution reconstruction is aimed at generating images of high spatial resolution from low-resolution observations. State-of-the-art super-resolution techniques underpinned with deep learning allow for obtaining results of outstanding visual quality, but it is seldom verified whether they constitute a valuable source for specific computer vision applications. In this paper, we investigate the possibility of employing super-resolution as a preprocessing step to improve optical character recognition from document scans. To achieve that, we propose to train deep networks for single-image super-resolution in a task-driven way to make them better adapted for the purpose of text detection. As problems limited to a specific task are heavily ill-posed, we introduce a multi-task loss function that embraces components related with text detection coupled with those guided by image similarity. The obtained results reported in this paper are encouraging and they constitute an important step towards real-world super-resolution of document images.