Text Detection in Scene Images Based on Interest Points


Minh Hieu Nguyen, Gueesang Lee, Journal of Information Processing Systems
Vol. 11, No. 4, pp. 528-537, Aug. 2015
10.3745/JIPS.02.0026
Keywords: Connected Component, Interest Point, Tensor Voting, Text Detection
Fulltext:

Abstract

Text in images is one of the most important cues for understanding a scene. In this paper, we propose a novel approach based on interest points to localize text in natural scene images. The main ideas of this approach are as follows: first we used interest point detection techniques, which extract the corner points of characters and center points of edge connected components, to select candidate regions. Second, these candidate regions were verified by using tensor voting, which is capable of extracting perceptual structures from noisy data. Finally, area, orientation, and aspect ratio were used to filter out non-text regions. The proposed method was tested on the ICDAR 2003 dataset and images of wine labels. The experiment results show the validity of this approach.


Statistics
Show / Hide Statistics

Statistics (Cumulative Counts from November 1st, 2017)
Multiple requests among the same browser session are counted as one view.
If you mouse over a chart, the values of data points will be shown.




Cite this article
[APA Style]
Minh Hieu Nguyen and Gueesang Lee (2015). Text Detection in Scene Images Based on Interest Points. Journal of Information Processing Systems, 11(4), 528-537. DOI: 10.3745/JIPS.02.0026.

[IEEE Style]
M. H. Nguyen and G. Lee, "Text Detection in Scene Images Based on Interest Points," Journal of Information Processing Systems, vol. 11, no. 4, pp. 528-537, 2015. DOI: 10.3745/JIPS.02.0026.

[ACM Style]
Minh Hieu Nguyen and Gueesang Lee. 2015. Text Detection in Scene Images Based on Interest Points. Journal of Information Processing Systems, 11, 4, (2015), 528-537. DOI: 10.3745/JIPS.02.0026.