DEEP LEARNING ARCHITECTURES FOR AUTOMATED CLASSIFICATION OF SATELLITE IMAGERY TO MONITOR GLOBAL DEFORESTATION TRENDS AND LAND USE CHANGES
Indonesia 
China China 
Abstract
Global deforestation and land-use change require timely, accurate, and transferable monitoring systems capable of processing extensive satellite archives. This study aimed to evaluate deep learning architectures for automated classification of satellite imagery, detection of forest change, and identification of post-deforestation land uses across diverse ecological regions. A comparative experimental design was applied to 120,000 labelled image patches derived from Landsat, Sentinel-1, and Sentinel-2 data representing South America, Southeast Asia, Central Africa, North America, Europe, and Oceania. Nine architectures, including convolutional, recurrent, Siamese, Transformer-based, and hybrid models, were assessed using spatial, temporal, and cross-regional validation. The hybrid CNN–Transformer achieved the strongest overall accuracy, calibration, and transfer performance, while radar–optical fusion improved classification in persistently cloudy tropical environments. Multitemporal models reduced confusion between permanent deforestation, seasonal disturbance, harvesting, and regeneration. Geographic holdout testing produced substantial performance declines across all architectures, demonstrating that random partitioning overestimated operational reliability. The study concludes that effective global deforestation monitoring requires multimodal data, temporal context, uncertainty estimation, active learning, and geographically independent evaluation. Automated classification can strengthen forest surveillance, yet expert verification and statistically adjusted area estimation remain essential for responsible environmental decision-making and policy implementation under rapidly changing climatic, technological, and land-management conditions worldwide over time.
Full text article
References
Abbas, I., & Damaševi?ius, R. (2025). Tree Detection in RGB Satellite Imagery Using YOLO-Based Deep Learning Models. Computers, Materials and Continua, 85(1), 483–502. https://doi.org/https://doi.org/10.32604/cmc.2025.066578
Alqadhi, S., Bindajam, A. A., Mallick, J., Talukdar, S., & Rahman, A. (2024). Applying deep learning to manage urban ecosystems in arid Abha, Saudi Arabia: Remote sensing-based modelling for ecological condition assessment and decision-making. Heliyon, 10(4), e25731. https://doi.org/https://doi.org/10.1016/j.heliyon.2024.e25731
Altarez, R. D. D., Apan, A., & Maraseni, T. (2023). Deep learning U-Net classification of Sentinel-1 and 2 fusions effectively demarcates tropical montane forest’s deforestation. Remote Sensing Applications: Society and Environment, 29, 100887. https://doi.org/https://doi.org/10.1016/j.rsase.2022.100887
Amini Sedeh, M., & Sharifian, S. (2025). A serverless satellite edge computing infrastructure to detect lake shoreline change based on contrastive attention siamese transformer. Engineering Applications of Artificial Intelligence, 154, 111015. https://doi.org/https://doi.org/10.1016/j.engappai.2025.111015
Ansari, M. G. I., Safiullah, D., Cao, J., & Wu, S. (2025). Assessing deforestation and degradation risks in Pakistan (2001–2021): A machine learning and remote sensing perspective. Environmental Technology & Innovation, 40, 104539. https://doi.org/https://doi.org/10.1016/j.eti.2025.104539
B, A., Alsolai, H., Allafi, R., & Arasi, M. A. (2025). Automated detection of landslide using synergizing dual Graph Convolutional Networks, googlenet, and machine learning techniques. Journal of South American Earth Sciences, 157, 105457. https://doi.org/https://doi.org/10.1016/j.jsames.2025.105457
Babitha, M. S., Diana Andrushia, A., Anand, N., Naser, M. Z., & Pari, Y. (2025). Land use and land cover change detection from multisource satellite imagery – A hybrid convolutional neural network approach. Engineering Applications of Artificial Intelligence, 161, 112187. https://doi.org/https://doi.org/10.1016/j.engappai.2025.112187
Chouikhi, F., Abbes, A. Ben, & Farah, I. R. (2025). Enhancing change detection in multi-date images using a Multi-temporal Siamese Neural Network. Pattern Recognition Letters, 197, 65–71. https://doi.org/https://doi.org/10.1016/j.patrec.2025.07.010
Colkesen, I., Saygi, M., Ozturk, M. Y., & Altuntas, O. Y. (2025). U-shaped deep learning networks for algal bloom detection using Sentinel-2 imagery: Exploring model performance and transferability. Journal of Environmental Management, 381, 125152. https://doi.org/https://doi.org/10.1016/j.jenvman.2025.125152
Cullerton, M., Zhu, Z., Qiu, S., Rittenhouse, C. D., & Suh, J. W. (2025). Back in time: A novel time series and deep learning framework for mapping solar installations. Science of Remote Sensing, 12, 100322. https://doi.org/https://doi.org/10.1016/j.srs.2025.100322
da Silva, M. P., Correa, S. P. L. P., Schaefer, M. A. R., Reis, J. C. S., Nunes, I. M., dos Santos, J. A., & Oliveira, H. N. (2025). Advancing agricultural remote sensing: A comprehensive review of deep supervised and Self-Supervised Learning for crop monitoring. Computers & Graphics, 133, 104434. https://doi.org/https://doi.org/10.1016/j.cag.2025.104434
Dalagnol, R., Wagner, F. H., Galvão, L. S., Braga, D., Osborn, F., Sagang, L. B., da Conceição Bispo, P., Payne, M., Silva Junior, C., Favrichon, S., Silgueiro, V., Anderson, L. O., Aragão, L. E. O. e C. de, Fensholt, R., Brandt, M., Ciais, P., & Saatchi, S. (2023). Mapping tropical forest degradation with deep learning and Planet NICFI data. Remote Sensing of Environment, 298, 113798. https://doi.org/https://doi.org/10.1016/j.rse.2023.113798
Debus, A., Beauchamp, E., Kamga, J., Verhegghen, A., Zébazé, C., & Lines, E. R. (2025). Evaluating satellite data and deep learning for identifying direct deforestation drivers in Cameroon. Remote Sensing Applications: Society and Environment, 39, 101653. https://doi.org/https://doi.org/10.1016/j.rsase.2025.101653
Fuentes, I., Lopatin, J., Galleguillos, M., Ceballos-Comisso, A., Eyheramendy, S., & Carrasco, R. (2024). Is the change deforestation? Using time-series analysis of satellite data to disentangle deforestation from other forest degradation causes. Remote Sensing Applications: Society and Environment, 35, 101210. https://doi.org/https://doi.org/10.1016/j.rsase.2024.101210
Goel, R. K., & Vishnoi, S. (2025). Space-age evolution - remote sensing and IoT for productive and sustainable agricultural landscape. Sustainable Futures, 10, 101280. https://doi.org/https://doi.org/10.1016/j.sftr.2025.101280
Guimarães, U. S., Rodrigues, T. B., Vieira, A. C., Hung, E. M., Soja, M. J., Eriksson, L. E. B., & Ulander, L. M. H. (2025). Deep learning models to map deforestation based on Sentinel 1 coherent features in the southern border of Amazon. Science of Remote Sensing, 12, 100279. https://doi.org/https://doi.org/10.1016/j.srs.2025.100279
Habibie, M. I., Ramadhan, Nurda, N., Sencaki, D. B., Putra, P. K., Prayogi, H., Agustan, Sutrisno, D., & Bintoro, O. B. (2024). The development land utilization and cover of the Jambi district are examined and forecasted using Google Earth Engine and CNN1D. Remote Sensing Applications: Society and Environment, 34, 101175. https://doi.org/https://doi.org/10.1016/j.rsase.2024.101175
Hosseini, H., Atazadeh, B., & Rajabifard, A. (2025). Towards intelligent land administration systems: Research challenges, applications and prospects in AI-driven approaches. Land Use Policy, 157, 107652. https://doi.org/https://doi.org/10.1016/j.landusepol.2025.107652
Hu, X., Zhang, P., & Ban, Y. (2023). Large-scale burn severity mapping in multispectral imagery using deep semantic segmentation models. ISPRS Journal of Photogrammetry and Remote Sensing, 196, 228–240. https://doi.org/https://doi.org/10.1016/j.isprsjprs.2022.12.026
Hussain, M. F., Meng, X., Shah, S. F., & Hu, X. (2025). Ecological impact assessment and optimizing artificial neural network for land cover classification under the billion tree tsunami project. Ecological Informatics, 92, 103463. https://doi.org/https://doi.org/10.1016/j.ecoinf.2025.103463
Jansi Rani, S. V, Ioannou, I., Swetha, R., Dhivya Lakshmi, R. M., & Vassiliou, V. (2024). A novel automated approach for fish biomass estimation in turbid environments through deep learning, object detection, and regression. Ecological Informatics, 81, 102663. https://doi.org/https://doi.org/10.1016/j.ecoinf.2024.102663
Kalinaki, K., Malik, O. A., & Ching Lai, D. T. (2023). FCD-AttResU-Net: An improved forest change detection in Sentinel-2 satellite images using attention residual U-Net. International Journal of Applied Earth Observation and Geoinformation, 122, 103453. https://doi.org/https://doi.org/10.1016/j.jag.2023.103453
Kopec, E., Wijata, A. M., & Nalepa, J. (2025). Change detection in Sentinel-2 images using deep learning ensembles. Remote Sensing Applications: Society and Environment, 40, 101764. https://doi.org/https://doi.org/10.1016/j.rsase.2025.101764
Lai, P., Lv, C., Zhou, L., Yang, S., Xu, J., Dong, Q., & He, M. (2025). Improved lightweight DeepLabV3+ for bare rock extraction from high-resolution UAV imagery. Ecological Informatics, 89, 103204. https://doi.org/https://doi.org/10.1016/j.ecoinf.2025.103204
Lausch, A., Selsam, P., Heege, T., von Trentini, F., Almeroth, A., Borg, E., Klenke, R., & Bumberger, J. (2025). Monitoring and modelling landscape structure, land use intensity and landscape change as drivers of water quality using remote sensing. Science of The Total Environment, 960, 178347. https://doi.org/https://doi.org/10.1016/j.scitotenv.2024.178347
Liu, J., Kong, X., Zhu, Y., & Zhang, B. (2025). A study on land use change simulation based on PLUS model and the U-net structure: A case study of Jilin Province. Ecological Indicators, 176, 113619. https://doi.org/https://doi.org/10.1016/j.ecolind.2025.113619
Mahendra, H. N., Pushpalatha, V., Mallikarjunaswamy, S., Rama Subramoniam, S., Rao, A. S., & Sanjay Shekar, N. C. (2025). Analyzing land use land cover changes in Mysuru taluk, Karnataka state, India using vision transformers. Applied Computing and Geosciences, 28, 100308. https://doi.org/https://doi.org/10.1016/j.acags.2025.100308
Mahendra, H. N., Pushpalatha, V., Mallikarjunaswamy, S., Rama Subramoniam, S., Sunil Rao, A., & Sharmila, N. (2024). LULC change detection analysis of Chamarajanagar district, Karnataka state, India using CNN-based deep learning method. Advances in Space Research, 74(12), 6384–6408. https://doi.org/https://doi.org/10.1016/j.asr.2024.07.066
Pushpalatha, V., Mallikarjuna, P. B., Mahendra, H. N., Rama Subramoniam, S., & Mallikarjunaswamy, S. (2025). Land use and land cover classification for change detection studies using convolutional neural network. Applied Computing and Geosciences, 25, 100227. https://doi.org/https://doi.org/10.1016/j.acags.2025.100227
Quash, Y., Kross, A., & Jaeger, J. A. G. (2024). Assessing the impact of gold mining on forest cover in the Surinamese Amazon from 1997 to 2019: A semi-automated satellite-based approach. Ecological Informatics, 80, 102442. https://doi.org/https://doi.org/10.1016/j.ecoinf.2023.102442
Rahman, M., Islam, M. M., Kim, H.-J., Alam, M., Sadiq, S., Rahman, M. K., Sadir Hossan, M., Islam, M. T., Raju, M. R., Alam, M. S., Ahmad, S. I., & Dewan, A. (2024). Optimizing urban water sustainability: Integrating deep learning, genetic algorithm, and CMIP6 GCM for groundwater potential zone prediction within a social-ecological-technological framework. Advances in Space Research, 73(12), 5925–5948. https://doi.org/https://doi.org/10.1016/j.asr.2024.03.033
Ramachandran, N., Irvin, J., Sheng, H., Johnson-Yu, S., Story, K., Rustowicz, R., Ng, A. Y., & Austin, K. (2024). Automatic deforestation driver attribution using deep learning on satellite imagery. Global Environmental Change, 86, 102843. https://doi.org/https://doi.org/10.1016/j.gloenvcha.2024.102843
Ramadan, M. N. A., Ali, M. A. H., Khoo, S. Y., & Alkhedher, M. (2024). AI-powered IoT and UAV systems for real-time detection and prevention of illegal logging. Results in Engineering, 24, 103277. https://doi.org/https://doi.org/10.1016/j.rineng.2024.103277
Rashid, M. R. A., Rafi, A. Al, Islam, M. A., Sharkar, S. U., Rafi, Z. H., Hasan, M., Ali, M. S., & Khan, M. S. H. (2025). Enhancing land management policy in Bangladesh: A blockchain-based framework for transparent and efficient land management. Land Use Policy, 150, 107436. https://doi.org/https://doi.org/10.1016/j.landusepol.2024.107436
Rehman, A., Ghani, M., Ahmad, S., Benchen, F., Rehman, N., Alanazi, A. K., Hamza, M., Riaz, M. H., Rehman, T. U., & Halima, S. (2025). Influence of land use land cover on land surface temperature in the North Western, Pakistan: Observation and modelling. Trees, Forests and People, 22, 101063. https://doi.org/https://doi.org/10.1016/j.tfp.2025.101063
Safaei, M., Southworth, J., Gibbes, C., Herrero, H. V, Rahaman, M., Tefera, B. B., & Blackburn, J. K. (2025). Land-cover classification in Addo Elephant National Park: Analyzing the impact of variables, classifiers, and object-based approach. Ecological Informatics, 90, 103279. https://doi.org/https://doi.org/10.1016/j.ecoinf.2025.103279
Shaikh, I. M., Akhtar, M. N., Aabid, A., & Ahmed, O. S. (2024). Enhancing sustainability in the production of palm oil: creative monitoring methods using YOLOv7 and YOLOv8 for effective plantation management. Biotechnology Reports, 44, e00853. https://doi.org/https://doi.org/10.1016/j.btre.2024.e00853
Silpalatha, G., & Jayadeva, T. S. (2025). Accelerating fast and accurate instantaneous segmentation with YOLO-v8 for remote sensing image analysis. Remote Sensing Applications: Society and Environment, 37, 101502. https://doi.org/https://doi.org/10.1016/j.rsase.2025.101502
T S, G., C, C., T, K., & K, R. R. (2025). Enhancing earth target classification in hyperspectral imagery using graph convolutional neural networks and graph-regularized sparse coding. Remote Sensing Applications: Society and Environment, 37, 101419. https://doi.org/https://doi.org/10.1016/j.rsase.2024.101419
Xiong, S., Zhang, X., Lei, Y., Tan, G., Wang, H., & Du, S. (2024). Time-series China urban land use mapping (2016–2022): An approach for achieving spatial-consistency and semantic-transition rationality in temporal domain. Remote Sensing of Environment, 312, 114344. https://doi.org/https://doi.org/10.1016/j.rse.2024.114344
Ye, L., & Weng, Q. (2025). A hybrid neural network for mangrove mapping considering tide states using Sentinel-2 imagery. Remote Sensing of Environment, 329, 114917. https://doi.org/https://doi.org/10.1016/j.rse.2025.114917
Yu, L., Du, Z., Li, X., Gu, J., Li, X., Zhong, L., Duojiweise, Wu, H., Zhao, Q., Ma, X., Zheng, J., Yang, Y., Song, W., Wang, P., Zhao, Z., Liao, L., Long, Y., Zhang, Y., Peng, J., … Peng, D. (2025). FROM-GLC Plus 3.0: Multimodal Land Change Mapping with SAM and Dense Surface Observations. Journal of Remote Sensing, 5. https://doi.org/https://doi.org/10.34133/remotesensing.0728
Zhang, X., Zhang, S., Meng, X., Zhang, G., Zang, D., Han, Y., Ai, H., & Liu, H. (2024). Remote sensing image segmentation of gully erosion in a typical black soil area in Northeast China based on improved DeepLabV3+ model. Ecological Informatics, 84, 102929. https://doi.org/https://doi.org/10.1016/j.ecoinf.2024.102929
Zhao, J., Wang, M., Ji, X., Wang, F., Wang, M., & Cai, D. (2025). Deep learning combined with focus on important feature strategy for the detection of Dendrolimus infestation area. Computers and Electronics in Agriculture, 237, 110780. https://doi.org/https://doi.org/10.1016/j.compag.2025.110780
Authors
Copyright (c) 2026 Sa’dianoor Sa’dianoor, Chen Mei, Yang Xiang
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
