2025
Special Issue: Remote Sensing based Intelligent Visual Analytics for Real-time Environmental and Earth Monitoring Systems

Special Issue: Remote Sensing based Intelligent Visual Analytics for Real-time Environmental and Earth Monitoring Systems

An Explainable Multi-Agent Framework for Real-Time Tree Detection and Canopy Segmentation in Remote-Sensed Imagery

Published 2025-12-22

  • K.P. Swain
    • ,
  • Soumya Ranjan Nayak
    • ,
  • Tarini Prasad Pattanaik
    • ,
  • Ashish Singh

K.P. Swain


Soumya Ranjan Nayak
School of Computer Engineering, KIIT Deemed To Be University, Bhubaneswar, Odisha

Tarini Prasad Pattanaik
School of Computer Engineering, KIIT Deemed To Be University, Bhubaneswar, Odisha

Ashish Singh
School of Computer Engineering, KIIT Deemed To Be University, Bhubaneswar, Odisha

Abstract

Monitoring tree cover and canopy architecture is important for sustainable forest management, biodiversity assessments, and climate adaptation planning. However, most current methods rely on large labeled datasets or specific sensors, which limit scalability and adaptability. This study hypothesizes that a zero-shot explainable multi-agent system can successfully detect and segment trees from standard RGB satellite imagery without having to retrain on task-specific data. A new framework is proposed that combines YOLO11m for tree detection and SAM2 for crown segmentation. The system utilizes a combination of vegetation, edge, and color-based agents that work in concert under an IoU based fusion strategy to increase robustness under varying brightness, shadows, and canopy overlap. Explainability includes Grad-CAM, SHAP, and LIME-based agents to visualize model attention to establish user trust. Experiments were conducted on a dataset of 2400 high-resolution satellite imagery (0.5–1.5 m). Moreover, the framework produced a 97.3% overall accuracy score, 97.6% precision score, 97.0% recall score, and 0.92 IoU, processing each image in under 10 seconds. The results of this study demonstrate that the `multi-agent zero-shot' method achieves high accuracy, fast inference, and transparent predictions for real-time vegetation monitoring, deforestation evaluations, and the urban canopy.

Keywords

Tree Detection, Satellite Imagery, Segmentation, YOLO, Explainable AI, Environmental Monitoring

PDF

References

  1. P. Ghamisi, J. Plaza, Y. Chen, J. Li, and A. J. Plaza, “Advanced spectral classifiers for hyperspectral images: A review,” IEEE Geosci. Remote Sens. Mag., vol. 5, no. 1, pp. 8–32, Mar. 2017.
  2. B. Haq et al., “Tech-driven forest conservation: Combating deforestation with Internet of Things, artificial intelligence, and remote sensing,” IEEE Internet Things J., vol. 11, no. 14, pp. 24551–24568, Jul. 2024.
  3. L. Ma, Y. Liu, X. Zhang, Y. Ye, G. Yin, and B. A. Johnson, “Deep learning in remote sensing applications: A meta-analysis and review,” ISPRS J. Photogramm. Remote Sens., vol. 152, pp. 166–177, 2019.
  4. A. Alem and S. Kumar, “Deep learning methods for land cover and land use classification in remote sensing: A review,” in Proc. 8th Int. Conf. Reliability, Infocom Technologies and Optimization (ICRITO), Noida, India, 2020, pp. 903–908.
  5. G. Lassalle, M. P. Ferreira, L. E. Cué La Rosa, C. R. de Souza Filho, “Deep learning-based individual tree crown delineation in mangrove forests using very-high-resolution satellite imagery,” ISPRS J. Photogramm. Remote Sens., vol. 189, pp. 220–235, 2022.
  6. S. Minaee et al., “Image segmentation using deep learning: A survey,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 44, no. 7, pp. 3523–3542, Jul. 2022.
  7. Y. Xian, C. H. Lampert, B. Schiele, and Z. Akata, “Zero-shot learning—A comprehensive evaluation of the good, the bad and the ugly,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 41, no. 9, pp. 2251–2265, Sep. 2019.
  8. T. Brandtberg and F. Walter, “Automated delineation of individual tree crowns in high spatial resolution aerial images by multiple-scale analysis,” Mach. Vis. Appl., vol. 11, pp. 64–73, 1998.
  9. F. E. Fassnacht et al., “Review of studies on tree species classification from remotely sensed data,” Remote Sens. Environ., vol. 186, pp. 64–87, 2016.
  10. M. Freudenberg, P. Magdon, and N. Nölke, “Individual tree crown delineation in high-resolution remote sensing images based on U-Net,” Neural Comput. Appl., vol. 34, pp. 22197–22207, 2022.
  11. J. Xu et al., “Tree crown segmentation and diameter at breast height prediction based on BlendMask in unmanned aerial vehicle imagery,” Remote Sens., vol. 16, p. 368, 2024.
  12. S. Li et al., “Deep learning enables image-based tree counting, crown segmentation, and height prediction at national scale,” PNAS Nexus, vol. 2, no. 4, p. pgad076, 2023.
  13. Y. Zhang et al., “Individual tree detection and counting based on high-resolution imagery and the canopy height model data,” Geo-Spatial Inf. Sci., vol. 27, no. 6, pp. 2162–2178, 2024.
  14. M. Wielgosz et al., “SegmentAnyTree: A sensor and platform agnostic deep learning model for tree segmentation using laser scanning data,” Remote Sens. Environ., vol. 313, p. 114367, 2024.
  15. M. S. R. Saimun and M. M. Rahman, “A comprehensive review of tree cover mapping using satellite sensor data,” Discov. Geosci., vol. 3, p. 90, 2025.
  16. J. Holmgren, E. Lindberg, K. Olofsson, and H. J. Persson, “Tree crown segmentation in three dimensions using density models derived from airborne laser scanning,” Int. J. Remote Sens., vol. 43, no. 1, pp. 299–329, 2022.
  17. M. Walker and G. Smith et al., “Literature review of unmanned aerial systems and LIDAR with application to distribution utility vegetation management,” Arboric. Urban Forestry, vol. 49, no. 3, pp. 144–156, 2023.
  18. S. Yel and E. Gormus, “Exploiting hyperspectral and multispectral images in the detection of tree species: A review,” Front. Remote Sens., vol. 4, 2023.
  19. F. Tong and Y. Zhang, “Individual tree crown delineation in high resolution aerial RGB imagery using StarDist-based model,” Remote Sens. Environ., vol. 319, p. 114618, 2025.
  20. J. L. Vaschetti, E. Arnaudo, and C. Rossi, “TreePseCo: Scaling individual tree crown segmentation using large vision models,” Int. Arch. Photogramm., Remote Sens. Spatial Inf. Sci., vol. XLVIII-M-7-2025, pp. 275–282, 2025.
  21. L. Yao, T. Liu, J. Qin, N. Lu, and C. Zhou, “Tree counting with high spatial-resolution satellite imagery based on deep neural networks,” Ecol. Ind., vol. 125, p. 107591, 2021.
  22. A. A. Movassagh, J. A. Alzubi, M. Gheisari et al., “Artificial neural networks training algorithm integrating invasive weed optimization with differential evolutionary model,” J. Ambient Intell. Human Comput., vol. 14, pp. 6017–6025, 2023.
  23. J. A. Alzubi, R. Jain, P. Nagrath et al., “Deep image captioning using an ensemble of CNN and LSTM based deep neural networks,” J. Intell. Fuzzy Syst., vol. 40, no. 4, pp. 5761–5769, 2020.
  24. M. C. Aksoy, Trees in Satellite Imagery, Kaggle Dataset, 2021. Available: https://www.kaggle.com/datasets/mcagriaksoy/trees-in-satellite-imagery

Downloads

Download data is not yet available.

Similar Articles

1-10 of 72

You may also start an advanced similarity search for this article.