[Home ] [Archive]   [ فارسی ]  
:: Main :: About :: Current Issue :: Archive :: Search :: Submit :: Contact ::
Main Menu
Home::
Browse::
Journal Info::
Guide for Authors::
Submit Manuscript::
Articles archive::
For Reviewers::
Contact us::
Site Facilities::
Reviewers::
::
Search in website

Advanced Search
..
Receive site information
Enter your Email in the following box to receive the site news and information.
..
:: Volume 16, Issue 1 (9-2026) ::
JGST 2026, 16(1): 81-90 Back to browse issues page
Zero-Shot Tree Crown Delineation in Semi-Arid Urban Forests Using UAV Imagery and the Segment Anything Model
MAHNAZ BABAEI , Zahra Azizi * , Siavash Bakhtiarvand Bakhtiari
Abstract:   (11 Views)
Urban forests play a critical role in mitigating environmental stresses in semi-arid cities by improving air quality, reducing urban heat island effects, enhancing carbon sequestration, and supporting ecosystem resilience. Accurate delineation of individual tree crowns (ITCs) is fundamental for urban forest inventory, biomass estimation, ecological monitoring, and sustainable green-space management. Recent advances in foundation models have created new opportunities for automated image segmentation; however, their applicability for tree crown delineation in semi-arid urban forests remains insufficiently investigated. In particular, the performance of zero-shot segmentation models under varying crown architectures and canopy densities is still poorly understood. This study evaluated the zero-shot capability of the Segment Anything Model (SAM) for ITC delineation using high-resolution UAV-derived RGB imagery acquired over a 1,600-ha urban afforestation area in southwestern Isfahan, Iran. A total of 5,974 reference trees representing three dominant species with contrasting crown structures, including Cercis siliquastrum, Robinia pseudoacacia, and Pinus eldarica, were used for quantitative evaluation. Model performance was assessed using Precision, Recall, F1-score, and Intersection over :union: (IoU) metrics. The results revealed substantial differences among species. SAM achieved its highest performance for Cercis siliquastrum, characterized by well-separated and relatively uniform crowns, with an F1-score of 0.78 and an IoU of 0.52. In contrast, performance declined considerably for Robinia pseudoacacia (F1 = 0.39, IoU = 0.23) and Pinus eldarica (F1 = 0.35, IoU = 0.18), where crown overlap, irregular canopy geometry, and dense vegetation structure increased segmentation uncertainty. Overall, 3,487 trees were correctly delineated by the model. The findings demonstrate that SAM provides a promising training-free framework for large-scale urban tree crown mapping. Nevertheless, segmentation accuracy remains strongly dependent on canopy structure, crown morphology, and canopy density. This study provides one of the first comprehensive evaluations of the zero-shot Segment Anything Model in semi-arid urban afforestation environments and highlights both the opportunities and limitations of foundation models for operational urban forest monitoring, inventorying, and management. The results further emphasize the need for integrating structural information and advanced prompting strategies to improve segmentation performance in complex urban canopy conditions.
Article number: 6
Keywords: Tree crown delineation, Zero-shot learning, UAV imagery, Semi-arid urban forests, Deep learning
Full-Text [PDF 845 kb]   (4 Downloads)    
Type of Study: Research | Subject: Photo&RS
Received: 2026/05/19 | Accepted: 2026/08/28
References
1. E. G. McPherson, N. van Doorn, and J. de Goede, "Structure, function and value of street trees in California, USA," Urban Forestry & Urban Greening, vol. 17, pp. 104-115, 2016. [DOI:10.1016/j.ufug.2016.03.013]
2. A. Harikumar, F. Bovolo, and L. Bruzzone, "A local projection-based approach to individual tree detection and 3-D crown delineation in multistoried coniferous forests using high-density airborne LiDAR data," IEEE Transactions on Geoscience and Remote Sensing, vol. 57, no. 2, pp. 1168-1182, Feb. 2019, doi: 10.1109/TGRS.2018.2865014. [DOI:10.1109/TGRS.2018.2865014]
3. F. E. Fassnacht, E. Schmidt-Riese, T. Kattenborn, and J. Hernández, "Explaining Sentinel-2-based dNBR and RdNBR variability with reference data from the bird's eye (UAS) perspective," International Journal of Applied Earth Observation and Geoinformation, vol. 95, Art. no. 102262, 2021, doi: 10.1016/j.jag.2020.102262. [DOI:10.1016/j.jag.2020.102262]
4. S. A. U. V. Natesan, "Individual tree species identification using Dense Convolutional Network (DenseNet) on multitemporal RGB images from UAV," Journal of Unmanned Vehicle Systems, vol. 8, no. 4, pp. 310-333, Dec. 2020, doi: 10.1139/juvs-2020-0007. [DOI:10.1139/juvs-2020-0007]
5. J. Talebi and Z. Azizi, "Enhancing land feature classification with the BTR Extractor: A novel software package for high-accuracy analysis of aerial laser scan data," MethodsX, vol. 14, Art. no. 103090, 2025, doi: 10.1016/j.mex.2025.103090. [DOI:10.1016/j.mex.2024.103090]
6. Z. Azizi and M. Miraki, "Individual urban trees detection based on point clouds derived from UAV-RGB imagery and local maxima algorithm: A case study of Fateh Garden, Iran," Environment, Development and Sustainability, vol. 26, no. 1, pp. 2331-2344, Jan. 2024, doi: 10.1007/s10668-022-02853-3.. [DOI:10.1007/s10668-022-02820-7]
7. Z. Azizi, A. Hosseini, and Y. Iranmanesh, "Estimating biomass of single oak trees using terrestrial photogrammetry," Journal of Environmental Science and Technology, vol. 19, no. 4, pp. 81-93, 2017.
8. H. Huang, X. Li, and C. Chen, "Individual tree crown detection and delineation from very-high-resolution UAV images based on bias field and marker-controlled watershed segmentation algorithms," IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 11, no. 7, pp. 2256-2269, Jul. 2018, doi: 10.1109/JSTARS.2018.2830410. [DOI:10.1109/JSTARS.2018.2830410]
9. K. Korznikov, D. Kislov, T. Petrenko, V. Dzizyurova, J. Doležal, P. Krestov, and J. Altman, "Unveiling the potential of drone-borne optical imagery in forest ecology: A study on the recognition and mapping of two evergreen coniferous species," Remote Sensing, vol. 15, no. 18, Art. no. 4394, 2023, doi: 10.3390/rs15184394. [DOI:10.3390/rs15184394]
10. M. Miraki and H. Sohrabi, "Integrating UAV imagery, manual, and automatic crown delineation for estimating green space water requirements in semi-arid climate," SSRN Electronic Journal, 2024, doi: 10.2139/ssrn.4953253. [DOI:10.2139/ssrn.4962387]
11. A. Daryaei, H. Sohrabi, C. Atzberger, and M. Immitzer, "Fine-scale detection of vegetation in semi-arid mountainous areas with focus on riparian landscapes using Sentinel-2 and UAV data," Computers and Geosciences, vol. 147, Art. no. 105686, 2021, doi: 10.1016/j.cageo.2020.105686. [DOI:10.1016/j.compag.2020.105686]
12. H. Zhong, Z. Zhang, H. Liu, J. Wu, and W. Lin, "Individual tree species identification for complex coniferous and broad-leaved mixed forests based on deep learning combined with UAV LiDAR data and RGB images," Forests, vol. 15, no. 2, Art. no. 293, 2024, doi: 10.3390/f15020293. [DOI:10.3390/f15020293]
13. L. P. Osco, Q. Wu, E. L. de Lemos, W. N. Gonçalves, A. P. M. Ramos, J. Li, and J. Marcato Junior, "The Segment Anything Model (SAM) for remote sensing applications: From zero to one shot," International Journal of Applied Earth Observation and Geoinformation, vol. 124, Art. no. 103540, 2023, doi: 10.1016/j.jag.2023.103540. [DOI:10.1016/j.jag.2023.103540]
14. R. Ravindran, Y. Treitz, and D. Iwaszczuk, "Investigation on dimensionality reduction methods for tree-crown segmentation in hyperspectral imagery using Segment Anything Model," ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, vol. X-G-2025, pp. 705-712, 2025, doi: 10.5194/isprs-annals-X-G-2025-705-2025. [DOI:10.5194/isprs-annals-X-G-2025-705-2025]
15. A. Kirillov, E. Mintun, N. Ravi, H. Mao, C. Rolland, L. Gustafson, T. Xiao, S. Whitehead, A. C. Berg, W.-Y. Lo, P. Dollár, and R. Girshick, "Segment Anything," in Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), Paris, France, 2023, pp. 4015-4026, doi: 10.1109/ICCV51070.2023.00371. [DOI:10.1109/ICCV51070.2023.00371]
16. M. Teng, A. Ouaknine, E. Laliberté, Y. Bengio, D. Rolnick, and H. Larochelle, "Assessing SAM for tree crown instance segmentation from drone imagery," Computer Vision and Pattern Recognition, 2025. https://arxiv.org/abs/2503.20199.
17. T. Kattenborn, J. Leitloff, F. Schiefer, and S. Hinz, "Review on convolutional neural networks (CNN) in vegetation remote sensing," ISPRS Journal of Photogrammetry and Remote Sensing, vol. 173, pp. 24-49, Mar. 2021, doi: 10.1016/j.isprsjprs.2020.12.010. [DOI:10.1016/j.isprsjprs.2020.12.010]
18. C. Hu, S. Dai, S. Wu, Q. Ye, and H. Yan, "Per-SAM-MCPA: A lightweight framework for individual tree crown segmentation from UAV imagery," Remote Sensing, vol. 18, no. 10, Art. no. 1559, 2026, doi: 10.3390/rs18101559. [DOI:10.3390/rs18101559]
19. J. Karthigesu, T. Owari, S. Tsuyuki, and T. Hiroshima, "Improving individual tree crown detection and species classification in a complex mixed conifer-broadleaf forest using two machine learning models with different combinations of metrics derived from UAV imagery," Geomatics, vol. 5, no. 3, Art. no. 32, 2025, doi: 10.3390/geomatics5030032. [DOI:10.3390/geomatics5030032]
20. J. L. Vaschetti, E. Arnaudo, and C. Rossi, "TreePseCo: Scaling individual tree crown segmentation using large vision models," The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, vol. XLVIII-M-7, pp. 275-282, 2025, doi: 10.5194/isprs-archives-XLVIII-M-7-2025-275-2025. [DOI:10.5194/isprs-archives-XLVIII-M-7-2025-275-2025]
21. A. Carraro, M. Sozzi, and F. Marinello, "The Segment Anything Model (SAM) for accelerating the smart farming revolution," Smart Agricultural Technology, vol. 6, Art. no. 100367, 2023, doi: 10.1016/j.atech.2023.100367. [DOI:10.1016/j.atech.2023.100367]
Send email to the article author

Add your comments about this article
Your username or Email:

CAPTCHA


XML   Persian Abstract   Print


Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:

BABAEI M, Azizi Z, Bakhtiarvand Bakhtiari S. Zero-Shot Tree Crown Delineation in Semi-Arid Urban Forests Using UAV Imagery and the Segment Anything Model. , JGST. 2026; 16 (1) : 6
URL: http://jgst.issgeac.ir/article-1-1260-en.html


Rights and permissions
Creative Commons License This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
Volume 16, Issue 1 (9-2026) Back to browse issues page
نشریه علمی علوم و فنون نقشه برداری Journal of Geomatics Science and Technology