Handwritten digit recognitionUsingencryptionmethods

Abstract

Over the past years, deep learning, a field of study of multi-layer technical neural networks, has had a strong impact on many areas of technical intelligence, including handwriting recognition. Handwriting recognition is a major research problem in the fields of image analysis and pattern recognition. Even today, digit recognition plays an important role in many areas, such as authenticating bank checks, exchanging remote computer files, and recognizing student notes. In this thesis, we present several contributions from the fields of deep learning and handwritten digit recognition, where we used encryption methods to encrypt images in order to improve the performance of recognizing handwritten digits. Each encryption method was studied separately. After that, several proposed protocols were used to monitor the accuracy of handwritten digit recognition. In this study, we used Convolutional Neural Networks CNN to train our model, where we conducted an empirical study on the CVL dataset and achieved great results and high accuracy rates that were compared to the state of the art in this subject