Image Classification Using Convolutional Neural Networks

Abstract

At present, many different tools can be used to classify images, each of which is intended to solve tasks in a certain range. This article provides a brief overview of libraries and technologies with respect to image classification. The architecture of a simple convolutional neural network for image classification is built. Experiments on image recognition have been conducted with popular neural networks such as VGG 16 and ResNet 50. Both neural networks have shown good results. However, ResNet 50 overfitted due to the fact that the dataset contained the same type of images for training, as this neural network has more layers that allow reading the attributes of objects in the images. A comparative analysis of image recognition that was particularly prepared for this experiment was carried out using the trained models.