Recognition image obtained by camera phone of character Arabic

In this paper, we proposed an offline Arabic handwriting character recognition system for isolated characters obtained by camera phone. Initially doing some pretreatments on the picture, the text is segmented into lines and then into characters. In a second phase, we have employed a several methods for extracting the features form the handwriting Arabic character, these methods are: Grey Level Co-occurrence Matrix (GLCM), Gabor Filters, Zoning, Projection Histogram, and Distance Profile. In addition, we have also tested the various combinations of Gabor Filters and Zoning. After that, for the classification stage we have used two classifiers: the Random Forest Method and Convolutional Neural Networks. However, we have presented a comparison between these classifiers. We carried out the experiments with a database containing 2800 samples collected from different writers. The experimental results show that our proposed OCR system is very efficient and provides good recognition accuracy rate of handwriting Arabic characters images acquired via camera phone.