An off-line handwritten alphabetical character recognition system usingmultilayer feed forward neural network is described in the paper. A new method,called, diagonal based feature extraction is introduced for extracting thefeatures of the handwritten alphabets. Fifty data sets, each containing 26alphabets written by various people, are used for training the neural networkand 570 different handwritten alphabetical characters are used for testing. Theproposed recognition system performs quite well yielding higher levels ofrecognition accuracy compared to the systems employing the conventionalhorizontal and vertical methods of feature extraction. This system will besuitable for converting handwritten documents into structural text form andrecognizing handwritten names.
展开▼
