Abstract
This article studies the use of a convolutional neural network for image recognition. Since neural networks at this time are widely used in recognizing computer images and have good results, further study and use of the neural network have tremendous relevance. In this article, we have made an analysis of the accuracy of image recognition using two different optimizers Adam and RMSProp, where the pictures of the authors of the article was presented as input data and several steps were performed for the implementation, which the article describes in detail. This article focuses on the RMSProp and Adam optimizers as they are adaptive and work in a large number of scenarios. The results in the form of tables and graphs were presented.