A Classification of Handwritten Multilingual Documents

In the Branch of computer Science handwritten script recognition is demanding part. Projected work is highlighting on the block level technique. In this wok we use the feature extraction technique for the recognition and we combined approach of Discrete Cosine Transform (DCT) and discrete wavelets Transform (DWT) for feature extraction and neural network (feed forward back propagation) classifier for classification and recognition purpose. The projected work has been experiment on three handwritten scripts Hindi, English and Urdu. For this work we create a database. That contains 9862 handwritten samples, written in three scripts.