Share Email Print
cover

Proceedings Paper

A maximally stable extremal region based scene text localization method
Author(s): Chengqiu Xiao; Lixin Ji; Chao Gao; Shaomei Li
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

Text localization in natural scene images is an important prerequisite for many content-based image analysis tasks. This paper proposes a novel text localization algorithm. Firstly, a fast pruning algorithm is designed to extract Maximally Stable Extremal Regions (MSER) as basic character candidates. Secondly, these candidates are filtered by using the properties of fitting ellipse and the distribution properties of characters to exclude most non-characters. Finally, a new extremal regions projection merging algorithm is designed to group character candidates into words. Experimental results show that the proposed method has an advantage in speed and achieve relatively high precision and recall rates than the latest published algorithms.

Paper Details

Date Published: 6 July 2015
PDF: 7 pages
Proc. SPIE 9631, Seventh International Conference on Digital Image Processing (ICDIP 2015), 96311Y (6 July 2015); doi: 10.1117/12.2196999
Show Author Affiliations
Chengqiu Xiao, National Digital Switching System Engineering and Technological Research Ctr. (China)
Lixin Ji, National Digital Switching System Engineering and Technological Research Ctr. (China)
Chao Gao, National Digital Switching System Engineering and Technological Research Ctr. (China)
Shaomei Li, National Digital Switching System Engineering and Technological Research Ctr. (China)


Published in SPIE Proceedings Vol. 9631:
Seventh International Conference on Digital Image Processing (ICDIP 2015)
Charles M. Falco; Xudong Jiang, Editor(s)

© SPIE. Terms of Use
Back to Top