Approach for Document Detection by Contours and Contrasts

TL;DR

This paper introduces a modified contour-based document detection method that ranks contour hypotheses by contrast, significantly reducing errors and achieving state-of-the-art results on benchmark datasets for mobile device applications.

Contribution

A novel contour ranking modification that improves document detection accuracy and efficiency on mobile devices, outperforming existing methods.

Findings

Reduced ordering errors by 40%

Decreased overall detection errors by 10%

Achieved state-of-the-art performance on MIDV-500 dataset

Abstract

This paper considers arbitrary document detection performed on a mobile device. The classical contour-based approach often fails in cases featuring occlusion, complex background, or blur. The region-based approach, which relies on the contrast between object and background, does not have application limitations, however, its known implementations are highly resource-consuming. We propose a modification of the contour-based method, in which the competing contour location hypotheses are ranked according to the contrast between the areas inside and outside the border. In the experiments, such modification allows for the decrease of alternatives ordering errors by 40% and the decrease of the overall detection errors by 10%. The proposed method provides unmatched state-of-the-art performance on the open MIDV-500 dataset, and it demonstrates results comparable with state-of-the-art performance on the SmartDoc dataset.