Automatic Text Area Segmentation in Natural Images

TL;DR

This paper introduces a hierarchical approach for segmenting text areas in natural images by identifying uniform backgrounds with holes, effectively detecting text regardless of language or script, demonstrated on diverse image datasets.

Contribution

The method uniquely focuses on background detection before text verification, handling multiple languages and symbols without prior language-specific assumptions.

Findings

Correctly segmented text in 63 out of 65 images

Minimal false positives with only 4 non-text areas segmented

Effective across different languages and symbol types

Abstract

We present a hierarchical method for segmenting text areas in natural images. The method assumes that the text is written with a contrasting color on a more or less uniform background. But no assumption is made regarding the language or character set used to write the text. In particular, the text can contain simple graphics or symbols. The key feature of our approach is that we first concentrate on finding the background of the text, before testing whether there is actually text on the background. Since uniform areas are easy to find in natural images, and since text backgrounds define areas which contain "holes" (where the text is written) we thus look for uniform areas containing "holes" and label them as text backgrounds candidates. Each candidate area is then further tested for the presence of text within its convex hull. We tested our method on a database of 65 images including English and Urdu text. The method correctly segmented all the text areas in 63 of these images, and in only 4 of these were areas that do not contain text also segmented.