Handwriting character recognition is one of the most important practical issues in pattern recognition experiments. The main objective of this project is to develop a handwritten digit recognition method with handwritten strings. In order to perform the score identification task, the scores will first be divided into individual scores. Then, the digit recognition module is used to classify each segmented digit by completing a manual digit string recognition task. Digit recognition applications include postal mail sorting, bank check processing, form data entry, etc. The heart of the problem lies in the ability to develop an efficient algorithm that can recognize handwritten digits and users scanners, tablets and other digital devices.

pattern recognition, digit recognition, artificial neural networks, Deep learning machine learning