{"title":"Detection of artificial fragments embedded in remote sensing images by adversarial neural networks","authors":"M. Gashnikov, A. Kuznetsov","doi":"10.18287/2412-6179-co-1064","DOIUrl":null,"url":null,"abstract":"We investigate algorithms for detecting artificial fragments of remote sensing images generated by adversarial neural networks. We consider a detector of artificial images based on the detection of a spectral artifact of generative-adversarial neural networks that is caused by a layer for enhancing the resolution. We use the detecting algorithm to detect artificial fragments embedded in natural remote sensing images using an adversarial neural network that includes a contour generator. We use remote sensing images of various types and resolutions, whereas the substituted areas, some being not simply connected, have different sizes and shapes. We experimentally prove that the investigated spectral neural network detector has high efficiency in detecting artificial fragments of remote sensing images.","PeriodicalId":46692,"journal":{"name":"Computer Optics","volume":null,"pages":null},"PeriodicalIF":1.1000,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Computer Optics","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.18287/2412-6179-co-1064","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"OPTICS","Score":null,"Total":0}
引用次数: 0
Abstract
We investigate algorithms for detecting artificial fragments of remote sensing images generated by adversarial neural networks. We consider a detector of artificial images based on the detection of a spectral artifact of generative-adversarial neural networks that is caused by a layer for enhancing the resolution. We use the detecting algorithm to detect artificial fragments embedded in natural remote sensing images using an adversarial neural network that includes a contour generator. We use remote sensing images of various types and resolutions, whereas the substituted areas, some being not simply connected, have different sizes and shapes. We experimentally prove that the investigated spectral neural network detector has high efficiency in detecting artificial fragments of remote sensing images.
期刊介绍:
The journal is intended for researchers and specialists active in the following research areas: Diffractive Optics; Information Optical Technology; Nanophotonics and Optics of Nanostructures; Image Analysis & Understanding; Information Coding & Security; Earth Remote Sensing Technologies; Hyperspectral Data Analysis; Numerical Methods for Optics and Image Processing; Intelligent Video Analysis. The journal "Computer Optics" has been published since 1987. Published 6 issues per year.