Sovremennye problemy distantsionnogo zondirovaniya Zemli iz kosmosa, 2023, Vol. 20, No. 6, pp. 144-154
Enhancement of Earth remote sensing data processing by artificial intelligence technologies utilizing system analysis of the signal transmission chain
V.V. Eremeev
1 , N.A. Egoshkin
1 , A.A. Makarenkov
1 , V.A. Ushenkin
1 , O.V. Postylyakov
2 1 Ryazan State Radio Engineering University named after V.F. Utkin, Ryazan, Russia
2 A.M. Obukhov Institute of Atmospheric Physics RAS, Moscow, Russia
Accepted: 28.11.2023
DOI: 10.21046/2070-7401-2023-20-6-144-154
The general scheme of the optical Earth remote sensing systems signal transmission chain (STC) is examined. System analysis of the STC is performed. Results of the analysis were applied to the process of object classification utilizing artificial intelligence (AI) approaches in order to enhance their performance. The impact of STC components on registered radiance by Earth remote sensing systems is examined. Three approaches of STC model application for enhancement of AI tools are proposed: application of STC model for removal of irrelevant information by correction of source Earth remote sensing data; application of STC model for the generation of learning data by synthesizing images of the Earth; application of STC model to generate additional information transferred to the input of AI tools in addition to images of the Earth. The results of experimental studies of STC application for cloud detection in the Earth remote sensing data by convolution neural networks are presented. The paper demonstrates that increase of the artificial intelligence methods performance can be achieved by inclusion of additional images (processed using approximate STC model) in the input of convolutional neural networks.
Keywords: signal transmission chain, Earth remote sensing system, artificial intelligence, classification, atmospheric transmission model, spectral exitance
Full textReferences:
- Voronin A. A., Egoshkin N. A., Eremeev V. V., Moskatinev I. V., Geometric data processing from earth global observing space systems, Vestnik Ryazanskogo gosudarstvennogo radiotekhnicheskogo universiteta, 2009, No. 1, Issue 27, pp. 12–17 (in Russian).
- Egoshkin N. A., Satellite images blurring and defocus correction in the case of geometric distortion, Digital signal processing, 2016, Issue 3, pp. 37–41 (in Russian).
- Egoshkin N. A., Eremeev V. V., Moskvitin A. E., Geostationary satellite images matching based on using the Earth disk edge points, Vestnik Ryazanskogo gosudarstvennogo radiotekhnicheskogo universiteta, 2007, No. 1, Issue 22, pp. 10–17 (in Russian).
- Egoshkin N. A., Eremeev V. V., Makarenkov A. A., Fusion of hyperspectral images of the Earth acquired in different spectral ranges, Digital signal processing, 2015, Issue 3, pp. 3–7 (in Russian).
- Moskvitin A. E., Ushenkin V. A., Laryukov S. A., Algorithm and software of high-speed neural network cloud segmentation of Resurs-P satellites’s panchromatic images, Digital signal processing, 2023, Issue 3, pp. 8–17 (in Russian).
- Postylyakov O. V., Spherical radiative transfer model with computation of layer air mass factors and some of its applications, Izvestiya, Atmospheric and Oceanic Physics, 2004, Vol. 40, No 3, pp. 276–290.
- Sovremennye tekhnologii obrabotki dannykh distantsionnogo zondirovaniya Zemli (Modern technologies of remote sensing data processing), V. V. Eremeev (ed.), Moscow: Fizmatlit, 2015, 460 p. (in Russian).
- Schowengerdt R. A., Remote Sensing: Models and Methods for Image Processing, 3rd ed., San Diego, CA: Academic Press, 2006, 558 p.
- Metz C. E., Basic Principles of ROC Analysis, Seminars in Nuclear Medicine, 1978, Vol. 8, No. 4, pp. 283–298, DOI: 10.1016/S0001-2998(78)80014-2.
- Postylyakov O. V., Linearized vector radiative transfer model MCC++ for a spherical atmosphere, J. Quantitative Spectroscopy and Radiative Transfer, 2004, Vol. 88, Issues 1–3, pp. 297–317, DOI: 10.1016/j.jqsrt.2004.01.009.