Image Classification with MNIST data set using Convolutional Neural Network

Abstract

Image classification is the assignment of taking an information picture and yield as a class or a likelihood of classes that best depicts the picture. Convolutional neural system (CNN) is a class of profound neural systems, most normally applied to breaking down visual symbolism. These were enlivened by natural procedures in that the network design between neurons looks like the association of the creature visual cortex. In this work, picture classification is performed on the MNIST informational collection utilizing the convolutional neural system. Since, the information is pictures with numbers from 0-9, the yield is multi class with ten classes.