Skip to Main Content
This paper proposes a two stage approach for extracting and recognizing text in images. This approach uses Morphological operation for text localization, and, Support Vector Machine for character recognition, along with some preprocessing and post processing steps. This algorithm is tested with natural Images as well as Images taken from ICDAR 2003 Robust Reading Competition dataset. The proposed work is evaluated using ICDAR 2003 competition metrics specification. Experimental results shows a promising improvement in the overall performance of the proposed technique when compared with that of the results using methods discussed in the related work. Also, the proposed feature set improves the discriminating ability during recognition stage against the features used in the previous methods discussed in the related work.
Date of Conference: 30-31 March 2012