HAND DRAWN CHARACTER RECOGNITION USING CONVOLUTIONAL NEURAL NETWORK









Abstract

The living beings with vision and intelligence can visualize and sense different surrounding things as well as text patterns and can make decisions according to the same. But for making computer systems that much capable of performing visualization and decision making, some special ways are being used that are known as algorithms. Convolutional Neural Network is a special algorithm which makes computer systems capable of performing various intelligent tasks. The goal of this paper is to build a CNN model that will be able to extract and recognize the character from the image provided to the model with less complexity and better accuracy. We intent to complete this by using CNN along with MNIST Dataset and other python libraries which are needful. Though the aim is to build a model which can recognize a single character based result, we also aim to enhance its functionality for a sentence recognition and then towards person’s handwriting. Throughout this work, we aim to learn CNN and other things and apply it as a logic on our model.


Modules


Algorithms


Software And Hardware