Urdu is widely used and reputed language all over the world. In Pakistan, most of the government official work has to be done in Urdu. All the court orders and government official documents are prepared in Urdu. But unfortunately the people there in the government offices are not It professionals. They feel easy with getting their work done by handwriting. No one ever thought of raising the status of their own language. Several techniques have been tried in order to recognize hand-written scripts namely segmentation based approaches, template matching, holistic methods, artificial neural networks, HMM based techniques.

1. Introduction 1.1 Problem statement 1.2 Key issues 1.3 Difficulties during research 2. Literature Review 2.1 Segmentation based approach 2.2 Template matching 2.3 Holistic methods 3. Research Methodology 3.1 Artificial neural networks 3.2 HMM based recognition 4. Conclusion .