Handwritten Digit Recognition Research Of Receipts

Receipt processing develops with the development of financial market. Receipt processing is an arduous work .So in order to using computer to instead people to do these arduous work, processing financial document automatically is necessary. Automatic handwritten digit recognition is the key.Handwritten digit recognition system involved the area of image processing and pattern recognition. This paper describes the process of handwriting digit recognition based on the handwritten digit of receipts .The emphases are the segmentation of handwritten numeral string and the choice and training of classifier.The process of handwritten digit recognition involves preprocess, feature extraction and segmentation and recognition of handwritten digit string. Firstly receipt image is preprocessed through the methods, which are reducing the noise, making the image smooth and binary methods; confirm the interesting region. By document analysis; get rid of the preprinted form frame lines and get the handwritten numeral string by Hough transform; put forward the recognition-based Method to solve the segmentation problem; design a classifier based on the BP neural networks to recognize the handwritten number, to get classifier with better performance, negative data must be the necessary trained samples and the rate of negative data and positive data must be reasonable. The classifier gets higher recognition accuracy through the increase of refuse rate.The experiment results show that the classifier trained with negative data and positive data get lower rate of misrecognition and better recognition accuracy.