MarkerTeV

Text Lines Extraction from Complex Scene

A brief description
A step by step example
References


Examples of text lines extraction. Each color correspond to a text line candidate to be feed to an OCR system

Input image

First set of candidate lines
(components darker than their background)

second set of candidate lines
(components lighter than their background)
input image: Template Matching Techniques in Computer Vision: Theory and Practice, Roberto Brunelli - Roby's book
set of candidates 1
set of candidates 2
input image: a book cover
set of candidates 1
set of candidates 2
input image: syntetic web image
set of candidates 1
set of candidates 2
input image: variable message sign
set of candidates 1
set of candidates 2
input image: traffic sign
set of candidates 1
set of candidates 2



Other examples of text lines extraction from generic scenes. Here all the candidate text lines are superimposed with different colors.
input image: tev team result
input image: tev team result
input image: city result
input image: snow result

References

S. Messelodi, C.M. Modena
Scene Text Recognition and Tracking to Identify Athletes in Sport Videos
Multimedia Tools and Applications, Special Issue on Automated Information Extraction in Media Production (on-line 2011) [doi]

S. Messelodi, C. M. Modena
Automatic Identification and Skew Estimation of Text Lines in Real Scene Images
Pattern Recognition, Vol. 32, No. 5, pp. 789-808, May 1999 [abstract] [doi]

S. Messelodi, C. M. Modena
Context Driven Text Segmentation and Recognition
Pattern Recognition Letters, Vol. 17, No. 1, pp. 47-56, 1996 [abstract] [doi]

---------------

TeV | research | technologies | people | links | publications | students | private