Handwriting recognition has been one of the most intriguing problems of the modern era. It can be considered as a superset of the character recognition problem and an entry point in machine learning. Throughout the years, various approaches have been designed to tackle the problem. One such approach is formulated in this paper. We have designed our system on Artificial Neural Networks which mimics the neurons of the human brain. The network uses either of gradient feature extraction and geometric feature extraction techniques to train and to generate output. The efficiency and accuracy of the output is superior than the other approaches used in the field. It is concluded that the problem of character recognition is best tackled by Neural Networks. Neural Networks have been applied to a range of problems like stock market prediction, image processing, medicine. An elaborated view of the system is provided in the following paper.

Keywords : Handwriting Recognition; Neural Networks; Gradient feature extraction; Geometric feature extraction.