Fast Glare Detection in Document Images

TL;DR

This paper presents a neural network-based method for detecting glare in document images captured by mobile devices, effectively identifying glare regions to improve text recognition accuracy.

Contribution

It introduces a novel approach combining luminance and histogram features with CNNs for accurate glare detection in mobile-captured document images.

Findings

High recall and F-score in glare detection

Effective use of luminance and histogram features

CNN-based heatmap generation for glare localization

Abstract

Glare is a phenomenon that occurs when the scene has a reflection of a light source or has one in it. This luminescence can hide useful information from the image, making text recognition virtually impossible. In this paper, we propose an approach to detect glare in images taken by users via mobile devices. Our method divides the document into blocks and collects luminance features from the original image and black-white strokes histograms of the binarized image. Finally, glare is detected using a convolutional neural network on the aforementioned histograms and luminance features. The network consists of several feature extraction blocks, one for each type of input, and the detection block, which calculates the resulting glare heatmap based on the output of the extraction part. The proposed solution detects glare with high recall and f-score.