Using automl technologies to solve monitoring problems

Abstract

Purpose: to substantiate the relevance of using the approach and technologies of automated machine learning (AutoML) to solve monitoring problems. The methodology for using this approach was described using the example of problems of monitoring animals on aerial photographs. Materials and methods: the article describes the architecture of a software application for automated machine learning of convolutional neural networks based on a genetic algorithm, as well as the methodology for its use in solving monitoring problems. Results: testing (validation) of the presented system was carried out using the example of solving problems of animal recognition on aerial photographs. Conclusions: the advantages and disadvantages, implementation features, areas of application of the AutoML approach and technologies are shown.