10-10-2014, 10:50 AM
Abstracts: Text extraction from image is a well-known pattern recognition problem. Extraction of text involves pre-processing, detection, binarization, enhancement and recognition of the text from a given image. However, text is varied due to differences in size, style, orientation, and alignment, as well as low image contrast and complex background make the problem of automatic text extraction extremely challenging. This has been implemented by using a technique which extracts text in two phases. First phase uses a simple, method based on edge detection. Second phase involves development of an Optical Character Recognizer. It could be possibly being performed in real time software.