|
Eleshkevich A.D.
|
nipineft@tomsknipi.ru |
“TomskNIPIneft” JSC Tomsk |
|
Eremenko M.S.
|
|
“TomskNIPIneft” JSC Tomsk |
|
Saibel E.G.
|
|
“TomskNIPIneft” JSC Tomsk |
|
Khristolubov I.A.
|
|
“TomskNIPIneft” JSC Tomsk |
|
Chernov A.G.
|
|
“TomskNIPIneft” JSC Tomsk |
DOI: 10.24412/2076-6785-2023-7-127-131
Abstract
This article discusses the experience of creating and implementing methods of automated decoding using machine learning algorithms to solve the problem of identifying areas of anthropogenic impact in oil and gas fields. The options for using various remote sensing data are described to identify such areas, using examples of forest clearings, along with their advantages and disadvantages. An original approach based on neural networks for decoding aerospace imagery is proposed, and its prospects for use are considered.
Materials and methods
Numerical modeling method using Earth remote sensing data.
Keywords
pattern recognition, remote sensing data, anthropogenic zones, computer vision, machine learning, forest cuttings, environmental monitoring, neural networks
