Communication Papers of the 20th Conference on Computer Science and Intelligence Systems (FedCSIS)

DOI: http://dx.doi.org/10.15439/2025F8182

Citation: Communication Papers of the 20th Conference on Computer Science and Intelligence Systems (FedCSIS), M. Bolanowski, M. Ganzha, L. Maciaszek, M. Paprzycki, D. Ślęzak (eds). ACSIS, Vol. 45, pages 1–8 (2025)

Full text

Abstract. This study presents a cost-effective tree crown segmentation framework using a hybrid deep learning model that combines a ResNet-34 encoder with a U-Net decoder. Our approach operates on low-resolution RGB Digital Orthophotos (DOPs) collected from urban and peri-urban areas in Bochum, Germany, simulating real-world data constraints. We processed 450 orthophoto--mask pairs through a comprehensive preprocessing pipeline including resizing (from 20000×20000 to 256×256), augmentation, and noise simulation. The model was trained using 10-fold cross-validation, achieving a Dice coefficient of 0.8678, Intersection over Union (IoU) of 0.7754, precision of 0.8410, and recall of 0.9103. These results demonstrate that even with downsampled imagery, reliable segmentation of tree crowns is feasible, making our approach suitable for low-cost forest inventory and precision agroforestry applications. Unlike previous studies relying on high-resolution LiDAR, this work is among the first to show robust tree crown segmentation using low-resolution orthophotos, making it accessible for widespread use in resource-constrained settings.

References

1. D. Rajasugunasekar, A. K. Patel, K. B. Devi, A. Singh, P. Selvam, and A. Chandra, “An integrative review for the role of forests in combating climate change and promoting sustainable development,” International Journal of Environment and Climate Change, 2023.
2. R. Damaševičius and R. Maskeliūnas, “Modeling forest regeneration dynamics: Estimating regeneration, growth, and mortality rates in lithuanian forests,” Forests, vol. 16, no. 2, 2025.
3. R. Damaševičius and R. Maskeliūnas, “Adaptive sensor clustering for environmental monitoring in dynamic forest ecosystems,” Peer-to-Peer Networking and Applications, vol. 18, no. 3, 2025.
4. R. Damaševičius, G. Mozgeris, A. Kurti, and R. Maskeliūnas, “Digital transformation of the future of forestry: an exploration of key concepts in the principles behind forest 4.0,” Frontiers in Forests and Global Change, vol. 7, 2024.
5. T. Mijit, E. Firkat, X. Yuan, Y. Liang, J. Zhu, and A. Hamdulla, “Lr-seg: A ground segmentation method for low-resolution lidar point clouds,” IEEE Transactions on Intelligent Vehicles, vol. 9, no. 1, pp. 347–356, 2024.
6. D. Joshi and C. Witharana, “Vision transformer based unhealthy tree crown detection and evaluation of annotation uncertainty,” 2025.
7. L. Wallace, A. Lucieer, and C. S. Watson, “Evaluating tree detection and segmentation routines on very high resolution uav lidar data,” IEEE Transactions on Geoscience and Remote Sensing, vol. 52, no. 12, pp. 7619–7628, 2014.
8. M. D. Hossain and D. Chen, “Remote sensing image segmentation: Methods, approaches, and advances,” Remote Sensing Handbook, Volume II, pp. 117–144, 2025.
9. H. Chen, W. Li, J. Gu, J. Ren, H. Sun, X. Zou, Z. Zhang, Y. Yan, and L. Zhu, “Low-res leads the way: Improving generalization for superresolution by self-supervised learning,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024, pp. 25 857–25 867.
10. B. Koonce, “Resnet 34,” in Convolutional neural networks with swift for tensorflow: image recognition and dataset categorization. Springer, 2021, pp. 51–61.
11. J. Shen, Q. Xu, M. Gao, J. Ning, X. Jiang, and M. Gao, “Aerial image segmentation of nematode-affected pine trees with u-net convolutional neural network,” Applied Sciences, vol. 14, no. 12, p. 5087, 2024.
12. J. Reitberger, C. Schnörr, P. Krzystek, and U. Stilla, “3d segmentation of single trees exploiting full waveform lidar data,” ISPRS Journal of Photogrammetry and Remote Sensing, vol. 64, no. 6, pp. 561–574, 2009.
13. Y. Liu, A. Zhang, and P. Gao, “From crown detection to boundary segmentation: Advancing forest analytics with enhanced yolo model and airborne lidar point clouds,” Forests, vol. 16, no. 2, p. 248, Jan. 2025.
14. Z. Xi, C. Hopkinson, and L. Chasmer, “Supervised terrestrial to airborne laser scanner model calibration for 3d individual-tree attribute mapping using deep neural networks,” ISPRS Journal of Photogrammetry and Remote Sensing, vol. 209, pp. 324–343, 2024.
15. M. Freudenberg, P. Magdon, and N. Nölke, “Individual tree crown delineation in high-resolution remote sensing images based on u-net,” Neural Computing and Applications, vol. 34, no. 24, p. 22197–22207, Aug. 2022.
16. H. Hamraz, M. A. Contreras, and J. Zhang, “A robust approach for tree segmentation in deciduous forests using small-footprint airborne lidar data,” International journal of applied earth observation and geoinformation, vol. 52, pp. 532–541, 2016.
17. C. Zhang, Y. Zhou, and F. Qiu, “Individual tree segmentation from lidar point clouds for urban forest inventory,” Remote Sensing, vol. 7, no. 6, pp. 7892–7913, 2015.
18. S. Li, S. Zhao, Z. Tian, H. Tang, and Z. Su, “Individual tree segmentation based on region-growing and density-guided canopy 3d morphology detection using uav lidar data,” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2025.
19. T. Saeed, E. Hussain, S. Ullah, J. Iqbal, S. Atif, and M. Yousaf, “Performance evaluation of individual tree detection and segmentation algorithms using als data in chir pine (pinus roxburghii) forest,” Remote Sensing Applications: Society and Environment, vol. 34, p. 101178, 2024.
20. Y. Yan, J. Lei, J. Jin, S. Shi, and Y. Huang, “Unmanned aerial vehicle–light detection and ranging-based individual tree segmentation in eucalyptus spp. forests: Performance and sensitivity,” Forests, vol. 15, no. 1, p. 209, 2024.
21. M. Dalponte and D. A. Coomes, “Tree-centric mapping of forest carbon density from airborne laser scanning and hyperspectral data,” Methods in ecology and evolution, vol. 7, no. 10, pp. 1236–1245, 2016.