Low-complexity text extraction in Korean signboards for mobile applications

Text extraction is a very important step in text understanding. Natural scene images bring new challenges to correctly extract interested text regions. In the paper, an efficient and low-complexity text extraction method is used to extract Korean text from signboards of shops. By combining the edge-based approach and some knowledge of the Korean text specified in our application, our method successfully extracts text and can be used in the mobile devices which have low capability.