Character Recognition: A Signature Approach
Arusa Firdous · International Journal of Advanced Research in Computer Science · 2017
Abstract Optical Character Recognition includes interpretation of meaningful information pertaining to a character from a digitized image in which scanned images of handwritten, typewritten text are converted into relevant machine text. The Character Recognition of both computer typed and handwritten characters has still a long way to go in terms of research. Although significant success has been achieved in type written characters but in handwritten it is still to touch an appreciable level. Most of the methods that have been proposed in this regard have huge computational complexity. The proposed research introduces an approach of character recognition which besides producing better results based on signatures of histograms for each character to be recognized has very less computational complexity as comparted to the other methods. The proposed research provides segmentation, classification and recognition of characters which are independent in size and texture and the method proposes methods that are able to accommodate character styles which have slight variations and also does not require thinning and other preprocessing measures as is required in other approaches.