ePrints@IIScePrints@IISc Home | About | Browse | Latest Additions | Advanced Search | Contact | Help

Multi-script and multi-oriented text localization from scene images

Kasar, Thotreingam and Ramakrishnan, Angarai G (2012) Multi-script and multi-oriented text localization from scene images. In: 4th International Workshop, CBDAR 2011, September 22, 2011, Beijing, China.

[img]
Preview
PDF
CBDAR11_Kasar_LNCS.pdf - Accepted Version

Download (16MB) | Preview
Official URL: http://dx.doi.org/10.1007/978-3-642-29364-1_1

Abstract

This paper describes a new method of color text localization from generic scene images containing text of different scripts and with arbitrary orientations. A representative set of colors is first identified using the edge information to initiate an unsupervised clustering algorithm. Text components are identified from each color layer using a combination of a support vector machine and a neural network classifier trained on a set of low-level features derived from the geometric, boundary, stroke and gradient information. Experiments on camera-captured images that contain variable fonts, size, color, irregular layout, non-uniform illumination and multiple scripts illustrate the robustness of the method. The proposed method yields precision and recall of 0.8 and 0.86 respectively on a database of 100 images. The method is also compared with others in the literature using the ICDAR 2003 robust reading competition dataset.

Item Type: Conference Paper
Publisher: Springer
Additional Information: Copyright of this article belongs to Springer Berlin Heidelberg.
Keywords: Text Detection; Scene Text; Multi-Script Documents; Multi-Oriented Text; Camera-Based Document Analysis
Department/Centre: Division of Electrical Sciences > Electrical Engineering
Date Deposited: 23 Apr 2013 05:26
Last Modified: 23 Apr 2013 05:26
URI: http://eprints.iisc.ac.in/id/eprint/46230

Actions (login required)

View Item View Item