Глибокі нейронні мережі для вирішення задач класифікації зображень звикористанням бібліотек keras та tensorflow

Abstract

This work is devoted to solving the problem of image classification based on deep learning. The task was classifyimagesinto two categories using a convolutional neural network (CNN). For this research, KerasPythonlibrary with TensorFlow library was used. As a result, convolutional neural network was created which detects the presence of a cat or a dog on the input image and assigns it the certain category. In the future, this CNN can be replaced by a pre-trained neural network from Kerasto improve classification accuracy.