000			01617nam a2200169Ia 4500
999			_c47013 _d47013
005			20200308073501.0
008			190927s9999 xx 000 0 und d
082			_aT Ab64 2011
100			_aAblanquee, Maria Fatima
245		0	_aOffline handwritten character recognition using artificial neural network
260			_a3
260			_b602
260			_c2011
520			_aThis project is entitled OFFLINE HANDWRITTEN CHARACTER RECOGNITION USING NEURAL NETWORK. Neural Networks grew out of research in Artificial Intelligence; specifically, attempts to mimic the way human brain thinks. They process information the way the human brain does. They learn the input-output relationship through training. They cannot be programmed to perform a specific task. Neural Networks nowadays are gaining much attention since they are very useful in applications that involve solving very complex problems where the conventional algorithms will not work. In this project, a system is presented to recognize handwritten non-cursive English UPPERCASE LETTERS using Neural network. The system will accept scanned image containing handwritten non-cursive English uppercase letters which will first undergo a preprocessing stage to remove noise and will be segmented after into suspected relevant characters. The detected characters will be used for training and recognition stages wherein a learning rule used by neural network is implemented. At the end of the training, when the network has learned, the network should be able to recognize the inputs correctly or with minimal errors.
526			_a000-099
942			_cRB _2ddc