Retrospective information about the state of the urbanized environment can be obtained by analyzing images located in the Internet space. Artificial intelligence-based methods allow recognizing objects of a given class in images, while models based on neural networks demonstrate high recognition accuracy. In this paper, a neural network model based on its own dataset for recognizing objects of an urbanized environment is proposed using the example of a class of construction works an creation of geoinformation system with coordinate time reference. The pixellib library was used for image segmentation. Photos found on the Internet were used to create the dataset. During the testing of the model, 15 versions of the neural network were trained. The best results were demonstrated by the model mask_rcnn_model.160-0.460171.h5 (version 15). The main reason for increasing the recognition accuracy of this model was more accurate contouring of objects and the addition of self-captured images to the dataset. The authors plan to improve the recognition accuracy of the neural network and train it to recognize other classes of objects in an urbanized environment.

Egor A. Koryukin
Lab-researcher
Ural Federal University named after the first President of Russia B.N. Yeltsin.
Institut of radioelectronics and information technologies RTF
32, Mira str., Ekaterinburg, 620002, Russia
e-mail: e.a.koryukin@gmail.com
ORCID: 0000-0003-4757-0871
SPIN-код: 1353-3812
AuthorID: 1217338

Veniamin S. Bobakov
Lab-researcher
Ural Federal University named after the first President of Russia B.N. Yeltsin
19, Mira str., Ekaterinburg, 620002, Russia
е-mail: veniabobakov97@gmail.com

Anastasia S. Butorova
Junior Researcher
Ural Federal University named after the first President of Russia B.N. Yeltsin
19, Mira str., Ekaterinburg, 620002, Russia
Junior Researcher
Institute of industrial ecology UB RA
20, S. Kovalevskoy str., Ekaterinburg, 620990, Russia
e-mail: a.s.butorova@urfu.ru
ORCID: 0000-0002-1570-6642
Scopus Author ID: 57959209000
SPIN-код: 5129-3990
AuthorID: 944608

Aleksandr P. Sergeev
Candidate of Physical and Mathematical Sciences,
Assistant Professor, Leading Researcher
Institute of Industrial Ecology UB RAS,
20, S. Kovalevskoy str., Ekaterinburg, 620990, Russia
e-mail: sergeev@ecko.uran.ru
ORCID: 0000-0001-7883-6017
ResearcherID: B-1565-2018
Scopus Author ID: 56227321200
SPIN-код: 8617-8873
AuthorID: 185393

1. Ilminskikh N.G. The urbanized environment. Vestnik Kurganskogo gosudarstvennogo universiteta. Seriya: Estestvennye nauki. 2012;(3):39–45.
2. Vershinin V.V., Nartov A.S. Differentiated system for calculating of payments for technologically polluted territories, based on the indicative method of assessing the impact of polyaromatic hydrocarbons. Geopolitics and Ecogeodynamics of regions. 2021;7(2):158–169. DOI: 10.37279/ 2309-7663-2021-7-2-154-166.
3. Beshentsev A.N., Kuklina E.E., Kalashnikov K.I., Baldanov N.D. Monitoring of the urbanized territory: methods, technologies, results. Vestnik of SSUGT. 2020;25(2):169–182. DOI: 10.33764/2411-1759-2020-25-2-169-182.
4. Sazonova L.E. Overview of information resources used for the analysis of green spaces in urban areas. Interexpo Geo-Siberia. 2022;6:213–219. DOI: 10.33764/2618-981X-2022-6-213-219.
5. Golubnichii A.A., Nedelina D.O. Predvaritel’noe raionirovanie goroda Chernogorska po priznaku vizual’noi zagryaznennosti. Evraziiskii nauchnyi zhurnal. 2015;(8):23–25.
6. Zaharov K.V., Medvedkov A.A., Borisov V.F. Landscape fragmentation and park landscaping as factors of accumulation heavy metals in birch leaves. Ekologiya i stroitelstvo. 2020;(1):4–13. DOI: 10.35688/2413-8452-2020-01-001.
7. Vasilyev A.V. Research of emissions to the atmosphere from automobile transport and of peculiarities of distribution of automobile transport flows on the territory of city district Togliatti of Russia. Izvestia of Samara Scientific Center of the Russian Academy of Sciences. 2021;23(5):39–46. DOI: 10.37313/1990-5378-2021-23-5-33-46.
8. Korotkiy A.A., Maslov V.B., Maslov D.V., Panfilov A.V., Kirsanov M.V. New kind of public transport for the urbanized environment – city rope-ways “Rope Metro”. Bulletin of higher educational institutions. North Caucasus region. Technical sciences. 2010;(4):73–77.
9. Seleznev A.A., Yarmoshenko I.V., Sergeev A.P. 137Cs in puddle sediments as timescale tracer in urban environment. Journal of Environmental Radioactivity. 2015;142:9–13. DOI: 10.1016/j.jenvrad.2015.01.001.
10. Seleznev A.A., Yarmoshenko I.V., Sergeev A.P. Method for reconstructing the initial baseline relationship between potentially harmful element and conservative element concentrations in urban puddle sediment. Geoderma. 2018;326:1–8. DOI: 10.1016/j.geoderma.2018.04.003.
11. Sarić R., Ulbricht M., Krstić M., Kevrić J., Jokić D. Recognition of Objects in the Urban Environment using R-CNN and YOLO Deep Learning Algorithms. In: 2020 9th Mediterranean Conference on Embedded Computing (MECO) (Budva, Montenegro, 2020). 2020. pp. 1–4. DOI: 10.1109/MECO49872.2020.9134080.
12. Najafizadeh L., Froehlich J.E. A Feasibility Study of Using Google Street View and Computer Vision to Track the Evolution of Urban Accessibility. In: ASSETS ’18: Proceedings of the 20th International ACM SIGACCESS Conference on Computers and Accessibility. 2018. pp. 340–342. DOI : 10.1145/3234695.3240999.
13. Pixellib’s official documentation. 2020. Available at: https://pixellib.readthedocs.io/en/latest/ (accessed 16.11.2022).
14. Ginsburg I. Issleduem arkhitektury svertochnykh neironnykh setei s pomoshch’yu fast.ai [Exploring convolutional neural network architectures using fast.ai]. Proglib. 28.12.2020. Available at: https://proglib.io/p/issleduem-arhitektury-svertochnyh-neyronnyh-setey-s-pomoshchyu-fast-ai-2020-12-28 (accessed 28.11.2022).
15. Jung A. Imgaug. Image augmentation for machine learning experiments. Github. 06.02.2020. Available at: https://github.com/aleju/imgaug (accessed 16.11.2022).
16. Kumar H. Evaluation metrics for object detection and segmentation: mAP. Technical Fridays. 20.09.2019. Available at: https://kharshit.github.io/blog/2019/09/20/evaluation-metrics-for-object-detection-and-segmentation (accessed 16.11.2022).