Assessing the Performance of Deep Learning Networks for Real-Time Deforestation Segmentation in the Amazon and Atlantic Biomes
Abstract
Deforestation poses significant ecological and societal challenges, while advances in satellite imagery and deep learning have enhanced monitoring precision and scalability. This study evaluates ten deep learning models for deforestation segmentation, including U-Net, ResNet, FCN, and YOLO variants, assessing accuracy and computational efficiency, focusing on the Amazon and Atlantic Forest biomes. The results highlight that U-Net and ResNet50 achieve the highest accuracy, while YOLOv8 and YOLOv11 offer an optimal balance between speed and performance. The findings contribute to the model selection for real-time deforestation detection, supporting conservation and environmental decision-making in underexplored areas like the Atlantic Forest.
References
Bragagnolo, L., da Silva, R. V., and Grzybowski, J. M. V. (2021). Amazon and atlantic forest image datasets for semantic segmentation. DOI: 10.5281/zenodo.4498086
Cabral, B. F., Yanai, A. M., de Alencastro Graça, P. M. L., Escada, M. I. S., de Almeida, C. M., and Fearnside, P. M. (2024). Amazon deforestation: A dangerous future indicated by patterns and trajectories in a hotspot of forest destruction in brazil. Journal of Environmental Management, 354:120354.
Careli, A., Vilas Boas, E., Teixeira, E., Silva, E., Aquino, G., and Pereira de Figueiredo, F. (2024). Deep learning segmentation models evaluation for deforestation monitoring embedded systems. In 2024 International Conference on Intelligent Cybernetics Technology & Applications, pages 274–278.
de Bem, P. P., de Carvalho Junior, O. A., Fontes Guimarães, R., and Trancoso Gomes, R. A. (2020). Change detection of deforestation in the brazilian amazon using landsat data and convolutional neural networks. Remote Sensing, 12(6).
He, K., Zhang, X., Ren, S., and Sun, J. (2016). Deep residual learning for image recognition. In 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 770–778.
Jocher, G., Chaurasia, A., and Qiu, J. (2023). Ultralytics yolov8.
Jocher, G. and Qiu, J. (2024). Ultralytics yolo11.
Kovačovič, P., Pirnik, R., Kafková, J., Michálik, M., Kanalikova, A., and Kuchár, P. (2025). Satellite-based forest stand detection using artificial intelligence. IEEE Access, 13:10898–10917.
Li, X., Fu, C., Wang, Q., Zhang, W., Sham, C.-W., and Chen, J. (2024). Dmsa-unet: Dual multi-scale attention makes unet more strong for medical image segmentation. Knowledge-Based Systems, 299:112050.
Long, J., Shelhamer, E., and Darrell, T. (2015). Fully convolutional networks for semantic segmentation. In 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 3431–3440.
Maurenza, D., Crouzeilles, R., Prevedello, J. A., and Almeida-Gomes, e. a. (2024). Effects of deforestation on multitaxa community similarity in the brazilian atlantic forest. Conservation Biology, n/a(n/a):e14419.
Md Jelas, I., Zulkifley, M. A., Abdullah, M., and Spraggon, M. (2024). Deforestation detection using deep learning-based semantic segmentation techniques: a systematic review. Frontiers in Forests and Global Change, 7:1300060.
Ortega Adarme, M., Queiroz Feitosa, R., Nigri Happ, P., Aparecido De Almeida, C., and Rodrigues Gomes, A. (2020). Evaluation of deep learning techniques for deforestation detection in the brazilian amazon and cerrado biomes from remote sensing imagery. Remote Sensing, 12(6).
Pathak, A., Kamani, M., and Priyanka, R. (2024). Satellite image segmentation via image quality enhancement and modified unet architecture. In 2024 2nd International Conference on Advancement in Computation & Computer Technologies (InCACCT), pages 646–651.
Ribas, L. C., Casaca, W., and Fares, R. T. (2025). Conditional generative adversarial networks and deep learning data augmentation: A multi-perspective data-driven survey across multiple application fields and classification architectures. AI, 6(2):32.
Rodrigues, J., Dias, M. A., Negri, R., Hussain, S. M., and Casaca, W. (2024). A robust dual-mode machine learning framework for classifying deforestation patterns in amazon native lands. Land, 13(9):1427.
Ronneberger, O., Fischer, P., and Brox, T. (2015). U-net: Convolutional networks for biomedical image segmentation. In Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015, volume 9351 of Lecture Notes in Computer Science, pages 234–241. Springer.
Silpalatha, G. and Jayadeva, T. (2025). Accelerating fast and accurate instantaneous segmentation with yolo-v8 for remote sensing image analysis. Remote Sensing Applications: Society and Environment, 37:101502.
Torres, D. L., Turnes, J. N., Soto Vega, P. J., Feitosa, R. Q., Silva, D. E., Marcato Junior, J., and Almeida, C. (2021). Deforestation detection with fully convolutional networks in the amazon forest from landsat-8 and sentinel-2 images. Remote Sensing, 13(24).
Wang, C.-Y. and Liao, H.-Y. M. (2024). Yolov9: Learning what you want to learn using programmable gradient information.
Cabral, B. F., Yanai, A. M., de Alencastro Graça, P. M. L., Escada, M. I. S., de Almeida, C. M., and Fearnside, P. M. (2024). Amazon deforestation: A dangerous future indicated by patterns and trajectories in a hotspot of forest destruction in brazil. Journal of Environmental Management, 354:120354.
Careli, A., Vilas Boas, E., Teixeira, E., Silva, E., Aquino, G., and Pereira de Figueiredo, F. (2024). Deep learning segmentation models evaluation for deforestation monitoring embedded systems. In 2024 International Conference on Intelligent Cybernetics Technology & Applications, pages 274–278.
de Bem, P. P., de Carvalho Junior, O. A., Fontes Guimarães, R., and Trancoso Gomes, R. A. (2020). Change detection of deforestation in the brazilian amazon using landsat data and convolutional neural networks. Remote Sensing, 12(6).
He, K., Zhang, X., Ren, S., and Sun, J. (2016). Deep residual learning for image recognition. In 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 770–778.
Jocher, G., Chaurasia, A., and Qiu, J. (2023). Ultralytics yolov8.
Jocher, G. and Qiu, J. (2024). Ultralytics yolo11.
Kovačovič, P., Pirnik, R., Kafková, J., Michálik, M., Kanalikova, A., and Kuchár, P. (2025). Satellite-based forest stand detection using artificial intelligence. IEEE Access, 13:10898–10917.
Li, X., Fu, C., Wang, Q., Zhang, W., Sham, C.-W., and Chen, J. (2024). Dmsa-unet: Dual multi-scale attention makes unet more strong for medical image segmentation. Knowledge-Based Systems, 299:112050.
Long, J., Shelhamer, E., and Darrell, T. (2015). Fully convolutional networks for semantic segmentation. In 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 3431–3440.
Maurenza, D., Crouzeilles, R., Prevedello, J. A., and Almeida-Gomes, e. a. (2024). Effects of deforestation on multitaxa community similarity in the brazilian atlantic forest. Conservation Biology, n/a(n/a):e14419.
Md Jelas, I., Zulkifley, M. A., Abdullah, M., and Spraggon, M. (2024). Deforestation detection using deep learning-based semantic segmentation techniques: a systematic review. Frontiers in Forests and Global Change, 7:1300060.
Ortega Adarme, M., Queiroz Feitosa, R., Nigri Happ, P., Aparecido De Almeida, C., and Rodrigues Gomes, A. (2020). Evaluation of deep learning techniques for deforestation detection in the brazilian amazon and cerrado biomes from remote sensing imagery. Remote Sensing, 12(6).
Pathak, A., Kamani, M., and Priyanka, R. (2024). Satellite image segmentation via image quality enhancement and modified unet architecture. In 2024 2nd International Conference on Advancement in Computation & Computer Technologies (InCACCT), pages 646–651.
Ribas, L. C., Casaca, W., and Fares, R. T. (2025). Conditional generative adversarial networks and deep learning data augmentation: A multi-perspective data-driven survey across multiple application fields and classification architectures. AI, 6(2):32.
Rodrigues, J., Dias, M. A., Negri, R., Hussain, S. M., and Casaca, W. (2024). A robust dual-mode machine learning framework for classifying deforestation patterns in amazon native lands. Land, 13(9):1427.
Ronneberger, O., Fischer, P., and Brox, T. (2015). U-net: Convolutional networks for biomedical image segmentation. In Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015, volume 9351 of Lecture Notes in Computer Science, pages 234–241. Springer.
Silpalatha, G. and Jayadeva, T. (2025). Accelerating fast and accurate instantaneous segmentation with yolo-v8 for remote sensing image analysis. Remote Sensing Applications: Society and Environment, 37:101502.
Torres, D. L., Turnes, J. N., Soto Vega, P. J., Feitosa, R. Q., Silva, D. E., Marcato Junior, J., and Almeida, C. (2021). Deforestation detection with fully convolutional networks in the amazon forest from landsat-8 and sentinel-2 images. Remote Sensing, 13(24).
Wang, C.-Y. and Liao, H.-Y. M. (2024). Yolov9: Learning what you want to learn using programmable gradient information.
Published
2025-07-20
How to Cite
BENVENUTO, Giovana A.; CASACA, Wallace; NEGRI, Rogério; COLNAGO, Marilaine.
Assessing the Performance of Deep Learning Networks for Real-Time Deforestation Segmentation in the Amazon and Atlantic Biomes. In: INTEGRATED SOFTWARE AND HARDWARE SEMINAR (SEMISH), 52. , 2025, Maceió/AL.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2025
.
p. 227-238.
ISSN 2595-6205.
DOI: https://doi.org/10.5753/semish.2025.8270.
