Neural network technology to search for targets in remote sensing images of the Earth N S Abramov 1 , А А Talalayev 1 , V P Fralenko 1 , O G Shishkin 1 and V M Khachumov 1,2 1 Aylamazyan Program Systems Institute of Russian Academy of Sciences, Peter the First Street, 4 “a”, Veskovo Village, Yaroslavl Region, Russia 2 The Peoples' Friendship University of Russia, Miklukho-Maklaya Street, 6, Moscow, Russia e-mail: shishkinog@mail.ru, nikolay.s.abramov@gmail.com Abstract. The paper introduces how multi-class and single-class problems of searching and classifying target objects in remote sensing images of the Earth are solved. To improve the recognition efficiency, the preparation tools for training samples, optimal configuration and use of deep learning neural networks using high-performance computing technologies have been developed. Two types of CNN were used to process ERS images: a convolutional neural network from the nnForge library and a network of the Darknet type. A comparative analysis of the results is obtained. The research showed that the capabilities of convolutional neural networks allow solving simultaneously the problems of searching (localizing) and recognizing objects in ERS images with high accuracy and completeness. 1. Introduction Today, there is an upsurge of activity in the field of Earth remote sensing (ERS) data processing: new software systems are being created, high-resolution image processing methods are being modernized. The current situation is characterized by the improvement of the equipment of spacecraft (SC) and ground control stations, the expansion of the functionality and spectrum of the image processing tasks performed. The scope of application of these spacecraft includes monitoring of forest, agricultural and arctic zones, analysis of natural disasters, environmental protection, public safety, etc. The growing volumes of evolving ERS data have significantly increased the requirements for speed and quality of information processing. Recently, artificial neural networks (ANN) and high-performance computing technologies have been increasingly used. The analysis of modern work on the application of ANN has shown that neural networks are mainly used for searching and recognizing targets that are related to the category of nonrigid [1,2]. The authors of this paper created a scientific and practical groundwork in solving various problems based on intelligent processing of ERS images (multispectral, panchromatic, color) search for rigid objects and zones of interest using the developed spectrographic approach and the generalized metric (fires, inundations, ice conditions assessment, etc.) [3-8]. The proposed paper presents the results of new in-depth studies related to the use of modern convolutional neural networks (CNN) for processing panoramic full-color ERS images obtained from unmanned aerial vehicles (UAVs); some methods and V International Conference on "Information Technology and Nanotechnology" (ITNT-2019)