Extraction and Recognition of Text From Digital English Comic Image Using Median Filter

Text extraction from image is one of the complicated areas in digital image processing. Text characters entrenched in image represents a rich source of information for text retrieval application. It isa complex process to detect and recognize the text from comic image due to their various size, gray scale values, complex backgrounds and different styles of font. Text extraction process from comic image helps to preserve the text and formatting during conversion process and provide high quality of text from the printed document. Automatic text extraction from comic images receives a growing attention because of prospective application in image retrieval. In existing work, Japanese text is extracted vertically from Manga Comic Image using Blob extraction functions. At the same time, text is extracted from multiple constraints using optical character recognition (OCR) and make translation of Japanese language of Manga into some other languages in conventional way to share the enjoyment of reading Manga through the Internet. This paper talks about English text extraction from blob comic image using various methods.