Handwritten character recognition usingnaïve bayesclassifier method for desktop and mobile application

Abstract

In Recent years, character recognition has gained more importance in the area of pattern recognition owning to its application in various domains. Many OCRs systems are been applied, but less interest have been given to document images obtained by camera phone.Off-line recognition of handwritten words is a difficult task due to the high variability and uncertainty of human writing. This paper presents a complete offline handwritten recognition system which describes the implementation of a desktop application and an android application. Our system includes five stages namely: pre-processing, segmentation, feature extraction, classification and postprocessing. Input for this system is a photo of handwritten text captured by a camera phone. Then it was directed to above stages and finally the output is produced. Naïve Bayes (NB) classification algorithm is used as classifier. In classification process we cut the image in several blocks. For each block, we compute a vector of descriptors. Then, we use K-means to cluster the low-level features including Zernike and Hu moments. Finally, we apply Bayesian networks classifier to classify the whole image of words. Experiments were performed with handwritten and machine-printed character images. The results indicate that the proposed system is very effective and yields good recognition rate for character images obtained by camera