APPLICATION OF GOOGLE EARTH IMAGERY AND YOLO FOR URBAN TREE DETECTION, SPATIAL MAPPING, AND CARBON STOCK ESTIMATION
DOI:
https://doi.org/10.70169/VJFS.1336Abstract
Effective management of urban green spaces plays an important role in maintaining ecosystem services and improving environmental quality in urban areas. This study aimed to evaluate the applicability of the YOLO26 model for detecting and mapping the spatial distribution of urban trees using Google Earth imagery in Tan Binh ward, Ho Chi Minh City. A total of 41 plant species belonging to 20 families were recorded, with mean diameter and height of 26.44 cm and 7.52 m, respectively. Mean biomass and carbon stock were 0.37 and 0.17 tons tree⁻¹, respectively. Mimusops elengi was the dominant species, accounting for 39.4% of the total trees, followed by Lagerstroemia speciosa (10.6%) and Cassia fistula (8.6%). Sapotaceae had the highest proportion of trees (40.2%), followed by Fabaceae (21.6%) and Lythraceae (10.6%). After 200 epochs, the YOLO26m model achieved the following results on the test dataset, with 95% confidence intervals of 0.502 - 0.681 for Precision, 0.576 - 0.765 for Recall, 0.533 - 0.687 for mAP50, and 0.228 - 0.342 for mAP50 - 95. These results indicate that the YOLO26m has potential for urban tree detection; however, its localization accuracy remains limited at higher IoU thresholds. The YOLO26m model estimated an average tree density of 13.5 trees ha-1, with biomass and carbon storage of 4.98 t ha-1 and 2.29 t C ha-1, respectively. Across the entire study area of 2.12 km² (approximately 212 ha), approximately 2856 trees were estimated, corresponding to 1056.72 tonnes of biomass and 485.52 tonnes of carbon stock. These findings demonstrate the potential of combining Google Earth imagery with YOLO26m for urban tree detection, mapping, and assessment of biomass and carbon stock.
References
1. Abbas I, Damaševičius R, 2025. Tree detection in RGB satellite imagery using YOLO-based deep learning models. Computers, Materials and Continua 85: 483-502. DOI: 10.32604/cmc.2025.066578.
2. Alfarhood M, Alabdullah N, 2026. Satellite-based tree detection for urban green space monitoring in arid environments. Frontiers in Environmental Science 14(2026):1-11. DOI: 10.3389/fenvs.2026.1888242.
3. Bao B, 2026. Crown-level, tree-only mapping of urban aboveground carbon stocks in dense city cores using deep learning. Urban Forestry & Urban Greening 121: 129477. DOI: 10.1016/j.ufug.2026.129477.
4. Barbierato E, Bernetti I, Capecchi I, Saragosa C, 2020. Integrating remote sensing and street view images to quantify urban forest ecosystem services. Remote Sensing 12: 329. DOI: 10.3390/rs12020329.
5. Brown S, 2002. Measuring carbon in forests: current status and future challenges. Environmental pollution (Barking, Essex : 1987). 116(3): 363-72. DOI: 10.1016/s0269-7491(01)00212-3.
6. Bộ Nông nghiệp và Môi trường, 2026. Bản đồ tra cứu các đơn vị hành chính Việt Nam. Nhà xuất bản Tài nguyên - Môi trường và Bản đồ Việt Nam. https://sapnhap.bando.com.vn/ (truy cập ngày 22/8/2026).
7. Eggleston HS, Buendia Leandro, Miwa Kyoko, Ngara Todd, Tanabe Kiyoto, 2006. IPCC guidelines for national greenhouse gas inventories. Institute for global environmental strategies: Hayama, Kanagawa, Japan.
8. Guo C, Hu Y, Qin J, Xu L, Chu M, Wang H, 2023. Image-based estimation of crown volume of individual street trees by plane calculation of angle disparity. Urban Forestry & Urban Greening 86: 128029. DOI: 10.1016/j.ufug.2023.128029.
9. Jocher G, Qiu J, Liu M, Lyu S, Akyon FC, Kalfaoglu ME, 2026. Ultralytics YOLO26: unified real-time end-to-end vision models. arXiv preprint arXiv. DOI: 10.48550/arXiv.2606.03748.
10. Lian X, Liu Z, Jiao L, Liu D, 2026. Mapping individual urban trees in arid cities: linking canopy coverage to land surface temperature variability using remote sensing. GIScience & Remote Sensing 63: 2640262. DOI: 10.1080/15481603.2026.2640262.
11. Nguyễn Minh Cảnh, Võ Trường Sơn, Mạc Văn Chăm, Nguyễn Thị Minh Hải, 2025. Đánh giá thực trạng và khả năng hấp thụ CO2 của cây xanh tại Công viên Gia Định, thành phố Hồ Chí Minh. Tạp chí Khoa học và Công nghệ Lâm nghiệp, 14: 13-24.
12. Redmon J, Divvala S, Girshick R, Farhadi A. 2016. You Only Look Once: Unified, Real-Time Object Detection. Pages 779-788. DOI: 10.55250/Jo.vnuf.14.5.2025.013-024
13. Sengun E, Aksoy S, Sertel E, Fransson JES, 2025. Comparative analysis of YOLOv8 and YOLOv11 on tree detection using UAV RGB and Laser scanning data. ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci. X-2/W2-2025: 173-179. DOI: 10.5194/isprs-annals-X-2-W2-2025-173-2025.
14. Stubbings P, Peskett J, Rowe F, Arribas-Bel D, 2019. A hierarchical urban forest index using street-level imagery and deep learning. Remote Sensing 11: 1395. DOI: 10.3390/rs11121395.
15. Tang Y, Chen A, Zhao S, 2016. Carbon storage and sequestration of urban street trees in Beijing, China. Frontiers in Ecology and Evolution 4: .1-8. DOI: 10.3389/fevo.2016.00053.
16. Trần Hợp, 2002. Tài nguyên cây gỗ Việt Nam. Hà Nội: Nhà xuất bản Nông nghiệp.
17. Tran N-S, Doan CC, 2024. Above ground biomass and carbon sequestration of urban green spaces in Danang City, Vietnam. Asian Journal of Biology 20: 1-8. DOI: 10.9734/ajob/2024/v20i6409.
18. Trần Ngọc Huyền T, 2025. Tích hợp ảnh UAV, Deep Learning và WebGIS phục vụ công tác quản lý cây xanh đô thị -Trường hợp nghiên cứu tại TP.HCM. Xây dựng 7: 6.
19. Tuan NT, Tuan PV, Hung NT, Duc Duong T, 2026. Individual tree detection and crown segmentation for Pinus kesiya using UAV–LiDAR and deep learning models. Dendrobiology 95: 156-169. DOI: 10.12657/denbio.095.011.
20. Velasquez-Camacho L, Etxegarai M, de-Miguel S, 2023. Implementing deep learning algorithms for urban tree detection and geolocation with high-resolution aerial, satellite, and ground-level images. Computers, Environment and Urban Systems 105: 102025. DOI: 10.1016/j.compenvurbsys.2023.102025.
21. Velasquez-Camacho L, Cardil A, Mohan M, Etxegarai M, Anzaldi G, de-Miguel S, 2021. Remotely sensed tree characterization in urban areas: A Review. Remote Sensing 13: 4889. DOI: 10.3390/rs13234889.
22. Yang M, Mou Y, Liu S, Meng Y, Liu Z, Li P, Xiang W, Zhou X, Peng C, 2022. Detecting and mapping tree crowns based on convolutional neural network and Google Earth images. International Journal of Applied Earth Observation and Geoinformation 108: 102764. DOI: 10.1016/j.jag.2022.102764.
23. Yoo J, Qi Y, Ashe-McNalley I, MacDonald B, Wilson JP, 2026. An open and transferable deep learning framework for mapping urban tree canopy using NAIP imagery. Remote Sensing 18: 1899. DOI: 10.3390/rs18121899
24. Zanne AE, 2009. Global wood density database. Dryad.
25. Zheng Y, Wu G, 2022. YOLOv4-Lite-Based urban plantation tree fetection and positioning with high-resolution remote sensing imagery. Frontiers in Environmental Science 9: 1-12. DOI: 10.3389/fenvs.2021.756227.
