Developing Handwriting Recognition System Using Neural Network Model

The goal of this paper is to solve the recognition problem with the development potential of machine learning, computer vision, and pattern recognition. It recognizes letters and handwritten characters. The paper focuses on the development of handwriting recognition algorithms. Information will be converted into digital form using a camera to capture handwritten letters. They are then pre-processed and extracted to obtain information in text form. This paper selects 14,541 and 260 samples for training and testing the network, respectively. We use 10 letters each in the database to create the test samples. As a result, the model achieves an accuracy of up to 90 %. The results show that the proposal system is feasible for practical application.