Received 18.03.2025, Revised 10.06.2025, Accepted 28.08.2025

Use of artificial intelligence and machine learning for automated detection of methane emissions on satellite images

German Krauklit

The aim of this study was to analyse modern machine learning methods and their integration into the processing of satellite data for the automated detection of methane emission sources. The research examined such methods as convolutional neural networks (CNN), support vector machines (SVM), random forest (RF), and k-nearest neighbours (KNN). The main findings demonstrated the efficiency of CNNs in the automated analysis of satellite images, particularly their ability to detect spatial patterns characteristic of methane emission sources with high accuracy and resilience to noise. It was established that the implementation of this technology for monitoring in Azerbaijan confirmed its capability to promptly identify anthropogenic threats based on satellite imagery. Meanwhile, the SVM method, implemented in Python, achieved a classification result with a 90% probability in favour of methane emission presence. At the same time, the R-based programme successfully classified images into the categories “methane emissions” and “no methane emissions”, effectively enabling the localisation of emission sources. The random forest method also proved effective in identifying methane sources using spectral characteristics, vegetation indices, and temperature data from satellite images. The R programme displayed the geographic distribution of methane sources, while the Python-based code validated the method’s effectiveness by processing an image of the city of Baku, identifying potential methane sources even at low concentrations. In turn, the KNN method showed potential in classifying pixels by spatial and spectral features, allowing for rapid detection of hazardous zones in new satellite images. The obtained results affirm the feasibility of integrating machine learning methods into satellite monitoring systems to ensure accurate, prompt, and automated detection of methane emission sources, which is a significant step towards enhancing environmental safety and effective natural resource management in Azerbaijan and beyond

convolutional neural network; support vector machine; random forest; k-nearest neighbours’ algorithm; implementation in Python and R; satellite monitoring
144-156
Krauklit, G. (2025). Use of artificial intelligence and machine learning for automated detection of methane emissions on satellite images. Information Technologies and Computer Engineering, 22(2), 144-156. https://doi.org/10.31649/vitce/2.2025.144

References

[1] Alshammari, A.F. (2024). Implementation of classification using K-nearest neighbors (KNN) in Python. International Journal of Computer Applications, 186(33), 19-24.

[2] Bairy, J., Shah, O.R., & Zhang, J. (2025). Data-driven solutions and artificial intelligence in the energy sector: Leveraging open GHGRP data for emissions forecasting for U.S. petroleum and natural gas assets. In SPE Oklahoma City oil and gas symposium (article number SPE-224384-MS). Oklahoma: Society of Petroleum Engineers. doi: 10.2118/224384-MS.

[3] Baku Satellite image from Pleiades Neo. (2021). Retrieved from https://space-solutions.airbus.com/newsroom/ satellite-image-gallery/pleiades-neo/pleiades-neo-baku/.

[4] Bi, H., & Neethirajan, S. (2024). Utilizing satellite data and machine learning for benchmarking methane emissions in the Canadian dairy industry. SSRN. doi: 10.2139/ssrn.4939071.

[5] Braik, A.M., & Koliou, M. (2024). Automated building damage assessment and large-scale mapping by integrating satellite imagery, GIS, and deep learning. Computer-Aided Civil and Infrastructure Engineering, 39(15), 2389-2404. doi: 10.1111/mice.13197.

[6] Chen, Y., Gao, J., & Zhang, K. (2020). R-CNN-based satellite components detection in optical images. International Journal of Aerospace Engineering, 2020(1), article number 8816187. doi: 10.1155/2020/8816187.

[7] Conway, T.J. (2024). To cut emissions, Azerbaijan prioritizes decarbonization of oil and gas. Retrieved from https://rmi. org/to-cut-emissions-azerbaijan-prioritizes-oil-and-gas/.

[8] Duan, H., et al. (2023). Quantification of diffusive methane emissions from a large eutrophic lake with satellite imagery. Environmental Science and Technology, 57(36), 13520-13529. doi: 10.1021/acs.est.3c05631.

[9] EOS Land Viewer. (n.d.). Passive sensors (day). Retrieved from https://eos.com/ landviewer/?day=true&s=Sentinel2&lat=40.42866&lng=49.52911&z=10.

[10] Goddard, L. (2024). Solving methane mysteries with satellite imagery. Retrieved from https://blog.datadesk.eco/p/ methane-mysteries.

[11] Ibrahimova, S., & Bağırzadə, M. (2024). Digital economy and artificial intelligence. The Caucasus: Economic and Social Analysis Journal of Southern Caucasus, 58(12(1)), 102-112. doi: 10.36962/cesajsc5801024-102.

[12] Ismayilov, R., & Mustafayev, Y. (2024). Leak detection in gas distribution networks. PAHTEI Refered & Reviewed Journal, 48(5(8)), 193-203. doi: 10.36962/PAHTEI4308022024-21.

[13] Karim, A., Alkholeify, A., Choi, J., Dhall, J., Huda, T., Ranjekar, A., Yassine, Y., & Wischert, D. (2024). Spiking neural network design for on-board detection of methane emissions through neuromorphic computing. In IAF earth observation symposium: Heald at the 75th international astronautical congress (pp. 1212-1223). Milan: International Astronautical Federation. doi: 10.52202/078362-0152.

[14] Koshariya, A.K., Alok, Hiremath, S.S., Nigam, R., Shankar, U., Kansara, S.S., Sinha, S.S., & Pratibha. (2025). Detection and management of plant disease using artificial intelligence and machine learning applications: A review. Journal of Scientific Research and Reports, 31(3), 281-298. doi: 10.9734/jsrr/2025/v31i32902.

[15] Kumar, D.V.R., Lakshman, V.V.V., Trylokya, M., Rao, M.L., & Sri, K.B. (2025). Land-water classification for satellite images. International Journal for Research in Applied Science and Engineering Technology, 13(3), 1945-1948. doi: 10.22214/ijraset.2025.67731.

[16] Kumar, S., Setia, R., & Singh, K. (2023). Artificial intelligence and machine learning in satellite data processing and services. Singapore: Springer. doi: 10.1007/978-981-19-7698-8.

[17] Mammadli, H., & Murtuzov, T. (2024). Preparation of geological and structural maps using artificial intelligence. In Conference “Study of the Earth’s subsoil and forecasting of mineral raw materials resources”. Baku: Baku State University.

[18] Marcon, Y., Stetzler, M., Ferré, B., Kopiske, E., & Bohrmann, G. (2025). Deep learning-based characterization of underwater methane bubbles using simple dual camera platform. Limnology and Oceanography: Methods, 23(3), 155-175. doi: 10.1002/lom3.10672.

[19] Marzouk, O.A. (2023). Adiabatic flame temperatures for oxy-methane, oxy-hydrogen, air-methane, and air-hydrogen stoichiometric combustion using the NASA CEARUN tool, GRI-Mech 3.0 reaction mechanism, and Cantera Python package. Engineering, Technology & Applied Science Research, 13(4), 11437-11444. doi: 10.48084/etasr.6132.

[20] Mexwell. (2023). Aerial imagery capturing oil refineries. Retrieved from https://www.kaggle.com/datasets/mexwell/ aerial-imagery-capturing-oil-refineries.

[21] Radman, A., Mahdianpari, M., Varon, D.J., & Mohammadimanesh, F. (2023). S2MetNet: A novel dataset and deep learning benchmark for methane point source quantification using Sentinel-2 satellite imagery. Remote Sensing of Environment, 295, article number 113708. doi: 10.1016/j.rse.2023.113708.

[22] Rashidov, T. (2019). Assessments of gas emission from the mud volcanoes in Azerbaijan. Geophysical Research Abstracts, 21, EGU2019-399.

[23] Rouet-Leduc, B., & Hulbert, C. (2024). Automatic detection of methane emissions in multispectral satellite imagery using a vision transformer. Nature Communications, 15, article number 3801. doi: 10.1038/s41467-024-47754-y.

[24] Sable, A.V., Patil, A., Rathi, M., & Shriwas, A. (2024). Classification of image using convolutional neural network (CNN). International Journal for Research in Applied Science and Engineering Technology, 12(4), 3357-3362. doi: 10.22214/ ijraset.2024.60650.

[25] Schillaci, G., et al. (2024). Analyzing methane emissions using satellite imagery and artificial intelligence. In 2024 Abu Dhabi international conference and exhibition (article number: SPE-222006-MS). Abu Dhabi: Society of Petroleum Engineers. doi: 10.2118/222006-MS.

[26] Schuit, B.J., et al. (2023). Automated detection and monitoring of methane super-emitters using satellite data. Atmospheric Chemistry and Physics, 23(16), 9071-9098. doi: 10.5194/acp-23-9071-2023.

[27] Sloan, S., Talkhani, R.R., Huang, T., Engert, J., & Laurance, W.F. (2024). Mapping remote roads using artificial intelligence and satellite imagery. Remote Sensing, 16(5), article number 839. doi: 10.3390/rs16050839.

[28] Sugeng, & Praminiarto, H. (2024). Detection of drowsiness in drivers using image processing and support vector machine (SVM) classification. Informatics Online Journal, 9(2), 238-248. doi: 10.15575/join.v9i2.1076.

[29] Xing, Y., Han, G., Mao, H., He, H., Bo, Z., & Gong, R. (2025). MAM-YOLOv9: A multiattention mechanism network for methane emission facility detection in high-resolution satellite remote sensing images. IEEE Transactions on Geoscience and Remote Sensing, 66. doi: 10.1109/TGRS.2025.3545034.

[30] Xu, P., Wang, J., Jiang, Y., & Gong, X. (2024). Applications of artificial intelligence and machine learning in image processing. Frontiers in Materials, 11, article number 1431179. doi: 10.3389/fmats.2024.1431179.