Integrated Semantically Guided Spatial Fragmentation and Biomorphological Forest Segmentation Using Terrestrial Laser Scanning Data: Layer-wise Dynamic Connectivity Features as the Key Factor in Stem Structure Reconstruction Accuracy
( Pp. 82-94)
More about authors
Grishin Ilya A.
postgraduate student
Bauman Moscow State Technical University
Moscow, Russian Federation Afanasyev Gennady I. Cand. Sci. (Eng.), Associate Professor; associate professor; Bauman Moscow State Technical University; Moscow, Russian Federation Terekhov Valeri I. Dr. Sci. (Eng.), Associate Professor; Head, Department of Information Processing and Management Systems; Bauman Moscow State Technical University; Moscow, Russian Federation
Bauman Moscow State Technical University
Moscow, Russian Federation Afanasyev Gennady I. Cand. Sci. (Eng.), Associate Professor; associate professor; Bauman Moscow State Technical University; Moscow, Russian Federation Terekhov Valeri I. Dr. Sci. (Eng.), Associate Professor; Head, Department of Information Processing and Management Systems; Bauman Moscow State Technical University; Moscow, Russian Federation
Abstract:
The separation of a forest plot into individual trees and the automatic extraction of their stem structures from terrestrial laser scanning data are complicated by dense stands, occlusions, and the diversity of biomorphological forms. Existing algorithms usually solve scene partitioning, voxel classification, and tree growing as independent tasks, which leads to error accumulation at subsequent processing stages. This paper proposes a unified model for spatial fragmentation and biomorphological forest segmentation comprising three interrelated stages: scene partitioning by estimated stem coordinates using a Voronoi diagram, probabilistic voxel- or point-level classification, and bottom-up tree growing guided by spatial connectivity and stem membership criteria. For the semantic module, tabular, volumetric, and point-based approaches are compared: gradient boosting with layer-by-layer inference, TabNet, a three-dimensional convolutional neural network, PointNet2, and two-stage pipelines in which gradient boosting builds an initial stem mask for subsequent neural segmentation. The experiment was conducted on 567 mixed-species trees. Considering both quality and computational performance, the {CatBoost; CNN3D} pipeline was selected as the preferred solution, achieving AUC = 0.9966 and IoU = 0.9831. The obtained results show that combining interpretable layer-by-layer classification with subsequent spatial analysis improves the quality of stem structure reconstruction, which is important for automatic forest inventory tasks.
How to Cite:
Grishin I.A., Afanasiev G.I. and Terekhov V.I. Integrated semantically guided spatial fragmentation and biomorphological forest segmentation using terrestrial laser scanning data: Layer-wise dynamic connectivity features as the key factor in stem structure reconstruction accuracy. Computational Nanotechnology. 13, 2 (2026), 82–94. DOI: 10.33693/2313-223X-2026-13-2-82-94. EDN: ZMBPYF
Reference list:
Mitrofanov E.M., Mitrofanova S.A., Karminov V.N., Chumachenko S.I. Assessment of repeatability and reproducibility of tree height measurement results using a mobile laser scanner. Legislative and Applied Metrology. 2024. No. 4 (190). Pp. 11–18. (In Rus.). DOI: 10.32446/2782-5418.2024-4-ll-18. EDN: ERMOLK.
Mitrofanov E.M., Chernyakova V.V., Kabanova M.M., Ustinov S.M. Method for forming a sample to train a neural network for decoding the tree species composition of a stand using a high-resolution orthophotomap based on their 3D images. Slavic Forum. 2023. No. 2 (40). Pp. 391–399. (In Rus.). EDN: IJTKMO.
Mitrofanova S.A., Mitrofanov E.M., Karminov V.N., Chumachenko S.I. Updating of a method for measuring tree trunk diameter based on robust design. Izmeritelnaya tekhnika. 2025. Vol. 74. No. 3. Pp. 23–32. (In Rus.). DOI: 10.32446/0368-1025it.2025-3-23-32. EDN: IXIKJW.
Ustinov S.M., Mitrofanov E.M., Ustinov M.V. Study of the possibility of using a ground-based mobile laser scanner to determine the height and diameter of trees in pine plantations. Bulletin of the Buryat State Academy of Agriculture named after V.R. Filippov. 2023. No. 1 (70). Pp. 134–140. (In Rus.). DOI: 10.34655/bgsha.2023.70.1.016. EDN: AECFQE.
Chumachenko S.I., Terekhov V.I., Mitrofanov E.M., Grishin I.A. An approach for estimating trees parameters using LiDAR data. Dynamics of Complex Systems – XXI Century. 2022. Vol. 16. No. 4. Pp. 63–73. (In Rus.). DOI: 10.18127/j19997493-202204-06. EDN: XDQNDC.
Shaitura S.V., Mitrofanov E.M., Zharov V.G., Feoktistova V.M. Spatial data infrastructure for a forest ecosystem digital twin. Design and Technologies. 2022. No. 91–92 (133–134). Pp. 160–168. (In Rus.). EDN: BVTVAX.
Shaitura S.V., Shaitura N.S., Mitrofanov E.M. et al. Application of terrestrial laser scanners for forest monitoring. Environmental Engineering. 2024. No. 4. Pp. 124–132. (In Rus.). DOI: 10.26897/1997-6011-2024-4-124-132. EDN: BVDMLA.
Arik S.Ö., Pfister T. TabNet: Attentive interpretable tabular learning. In: Proceedings of the AAAI Conference on Artificial Intelligence. 2021. Vol. 35. No. 8. Pp. 6679–6687. DOI: 10.1609/aaai.v35i8.16826.
Begliarov N., Mitrofanov E.M., Kiseleva V. Generating a three-dimensional measuring scene for the forest sector as based on modern geodetic technologies. IOP Conference Series: Earth and Environmental Science. 2021. Vol. 875. No. 1. Art. 012083. DOI: 10.1088/1755-1315/875/1/012083. EDN: AWGPJS.
Calders K., Adams J., Armston J. et al. Terrestrial laser scanning in forest ecology: Expanding the horizon. Remote Sensing of Environment. 2020. Vol. 251. Art. 112102. DOI: 10.1016/j.rse.2020.112102.
Cicek Ö., Abdulkadir A., Lienkamp S.S. et al. 3D U-Net: Learning dense volumetric segmentation from sparse annotation. Lecture Notes in Computer Science. 2016. Vol. 9901. Pp. 424–432. DOI: 10.1007/978-3-319-46723-8_49.
Grishin I.A., Goryachkin B.S., Terekhov V.I. Chumachenko S.I. An efficient technique for determining tree coordinates using LiDAR data via deep learning. Proceedings of REEPE. 2024. No. 222. Pp. 1–6. DOI: 10.1109/REEPE60449.2024.10479853. EDN: WWDRDT.
Grishin I.A., Krutov T.Y., Kanev A.I. Terekhov V.I. Individual tree segmentation quality evaluation using deep learning models LiDAR based. Optical Memory and Neural Networks (Information Optics). 2023. Vol. 32. No. S2. Pp. S270–S276. DOI: 10.3103/s1060992x23060061. EDN: SUYZZK.
Grishin I.A., Terekhov V.I. Procedure for locating trees and estimating diameters using LiDAR data. Proceedings of REEPE. 2023. Pp. 1–5. DOI: 10.1109/REEPE57272.2023.10086843. EDN: SJWZYC.
Hackenberg J., Spiecker H., Calders K. et al. SimpleTree: An efficient open source tool to build tree models from TLS clouds. Forests. 2015. Vol. 6. No. 12. Pp. 4245–4294. DOI: 10.3390/f6114245.
Liang X., Kankare V., Hyyppä J. et al. Terrestrial laser scanning in forest inventories. ISPRS Journal of Photogrammetry and Remote Sensing. 2016. Vol. 115. Pp. 63–77. DOI: 10.1016/j.isprsjprs.2016.01.006.
Prokhorenkova L., Gusev G., Vorobev A., Dorogush A.V., Gulin A. CatBoost: unbiased boosting with categorical features. In: Advances in Neural Information Processing Systems (NeurIPS 2018). Vol. 31. Pp. 6638–6648. Also available as arXiv preprint. DOI: 10.48550/arXiv.1706.09516. EDN: NRYEVS.
Qi C.R., Su H., Mo K., Guibas L.J. PointNet: Deep learning on point sets for 3D classification and segmentation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2017. Pp. 652–660. DOI: 10.1109/CVPR.2017.16.
Qi C.R., Yi L., Su H., Guibas L.J. PointNet++: Deep hierarchical feature learning on point sets in a metric space. Advances in Neural Information Processing Systems. 2017. Vol. 30. Pp. 717–728. DOI: 10.48550/arXiv.1706.02413.
Raumonen P., Kaasalainen M., Åkerblom M. et al. Fast automatic precision tree models from terrestrial laser scanner data. Remote Sensing. 2013. Vol. 5. No. 2. Pp. 491–520. DOI: 10.3390/rs5020491.
Ronneberger O., Fischer P., Brox T. U-Net: Convolutional networks for biomedical image segmentation. Lecture Notes in Computer Science. 2015. Vol. 9351. Pp. 234–241. DOI: 10.1007/978-3-319-24574-4_28.
Thomas H., Qi C.R., Deschaud J.E. et al. KPConv: Flexible and deformable convolution for point clouds. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. 2019. Pp. 6411–6420. DOI: 10.1109/ICCV.2019.00651.
Wang D., Hollaus M., Pfeifer N. Feasibility of machine learning methods for separating wood and leaf points from terrestrial laser scanning data. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences. 2017. Vol. IV-2/W4. Pp. 157–164. DOI: 10.5194/isprs-annals-IV-2-W4-157-2017.
Mitrofanov E.M., Chernyakova V.V., Kabanova M.M., Ustinov S.M. Method for forming a sample to train a neural network for decoding the tree species composition of a stand using a high-resolution orthophotomap based on their 3D images. Slavic Forum. 2023. No. 2 (40). Pp. 391–399. (In Rus.). EDN: IJTKMO.
Mitrofanova S.A., Mitrofanov E.M., Karminov V.N., Chumachenko S.I. Updating of a method for measuring tree trunk diameter based on robust design. Izmeritelnaya tekhnika. 2025. Vol. 74. No. 3. Pp. 23–32. (In Rus.). DOI: 10.32446/0368-1025it.2025-3-23-32. EDN: IXIKJW.
Ustinov S.M., Mitrofanov E.M., Ustinov M.V. Study of the possibility of using a ground-based mobile laser scanner to determine the height and diameter of trees in pine plantations. Bulletin of the Buryat State Academy of Agriculture named after V.R. Filippov. 2023. No. 1 (70). Pp. 134–140. (In Rus.). DOI: 10.34655/bgsha.2023.70.1.016. EDN: AECFQE.
Chumachenko S.I., Terekhov V.I., Mitrofanov E.M., Grishin I.A. An approach for estimating trees parameters using LiDAR data. Dynamics of Complex Systems – XXI Century. 2022. Vol. 16. No. 4. Pp. 63–73. (In Rus.). DOI: 10.18127/j19997493-202204-06. EDN: XDQNDC.
Shaitura S.V., Mitrofanov E.M., Zharov V.G., Feoktistova V.M. Spatial data infrastructure for a forest ecosystem digital twin. Design and Technologies. 2022. No. 91–92 (133–134). Pp. 160–168. (In Rus.). EDN: BVTVAX.
Shaitura S.V., Shaitura N.S., Mitrofanov E.M. et al. Application of terrestrial laser scanners for forest monitoring. Environmental Engineering. 2024. No. 4. Pp. 124–132. (In Rus.). DOI: 10.26897/1997-6011-2024-4-124-132. EDN: BVDMLA.
Arik S.Ö., Pfister T. TabNet: Attentive interpretable tabular learning. In: Proceedings of the AAAI Conference on Artificial Intelligence. 2021. Vol. 35. No. 8. Pp. 6679–6687. DOI: 10.1609/aaai.v35i8.16826.
Begliarov N., Mitrofanov E.M., Kiseleva V. Generating a three-dimensional measuring scene for the forest sector as based on modern geodetic technologies. IOP Conference Series: Earth and Environmental Science. 2021. Vol. 875. No. 1. Art. 012083. DOI: 10.1088/1755-1315/875/1/012083. EDN: AWGPJS.
Calders K., Adams J., Armston J. et al. Terrestrial laser scanning in forest ecology: Expanding the horizon. Remote Sensing of Environment. 2020. Vol. 251. Art. 112102. DOI: 10.1016/j.rse.2020.112102.
Cicek Ö., Abdulkadir A., Lienkamp S.S. et al. 3D U-Net: Learning dense volumetric segmentation from sparse annotation. Lecture Notes in Computer Science. 2016. Vol. 9901. Pp. 424–432. DOI: 10.1007/978-3-319-46723-8_49.
Grishin I.A., Goryachkin B.S., Terekhov V.I. Chumachenko S.I. An efficient technique for determining tree coordinates using LiDAR data via deep learning. Proceedings of REEPE. 2024. No. 222. Pp. 1–6. DOI: 10.1109/REEPE60449.2024.10479853. EDN: WWDRDT.
Grishin I.A., Krutov T.Y., Kanev A.I. Terekhov V.I. Individual tree segmentation quality evaluation using deep learning models LiDAR based. Optical Memory and Neural Networks (Information Optics). 2023. Vol. 32. No. S2. Pp. S270–S276. DOI: 10.3103/s1060992x23060061. EDN: SUYZZK.
Grishin I.A., Terekhov V.I. Procedure for locating trees and estimating diameters using LiDAR data. Proceedings of REEPE. 2023. Pp. 1–5. DOI: 10.1109/REEPE57272.2023.10086843. EDN: SJWZYC.
Hackenberg J., Spiecker H., Calders K. et al. SimpleTree: An efficient open source tool to build tree models from TLS clouds. Forests. 2015. Vol. 6. No. 12. Pp. 4245–4294. DOI: 10.3390/f6114245.
Liang X., Kankare V., Hyyppä J. et al. Terrestrial laser scanning in forest inventories. ISPRS Journal of Photogrammetry and Remote Sensing. 2016. Vol. 115. Pp. 63–77. DOI: 10.1016/j.isprsjprs.2016.01.006.
Prokhorenkova L., Gusev G., Vorobev A., Dorogush A.V., Gulin A. CatBoost: unbiased boosting with categorical features. In: Advances in Neural Information Processing Systems (NeurIPS 2018). Vol. 31. Pp. 6638–6648. Also available as arXiv preprint. DOI: 10.48550/arXiv.1706.09516. EDN: NRYEVS.
Qi C.R., Su H., Mo K., Guibas L.J. PointNet: Deep learning on point sets for 3D classification and segmentation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2017. Pp. 652–660. DOI: 10.1109/CVPR.2017.16.
Qi C.R., Yi L., Su H., Guibas L.J. PointNet++: Deep hierarchical feature learning on point sets in a metric space. Advances in Neural Information Processing Systems. 2017. Vol. 30. Pp. 717–728. DOI: 10.48550/arXiv.1706.02413.
Raumonen P., Kaasalainen M., Åkerblom M. et al. Fast automatic precision tree models from terrestrial laser scanner data. Remote Sensing. 2013. Vol. 5. No. 2. Pp. 491–520. DOI: 10.3390/rs5020491.
Ronneberger O., Fischer P., Brox T. U-Net: Convolutional networks for biomedical image segmentation. Lecture Notes in Computer Science. 2015. Vol. 9351. Pp. 234–241. DOI: 10.1007/978-3-319-24574-4_28.
Thomas H., Qi C.R., Deschaud J.E. et al. KPConv: Flexible and deformable convolution for point clouds. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. 2019. Pp. 6411–6420. DOI: 10.1109/ICCV.2019.00651.
Wang D., Hollaus M., Pfeifer N. Feasibility of machine learning methods for separating wood and leaf points from terrestrial laser scanning data. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences. 2017. Vol. IV-2/W4. Pp. 157–164. DOI: 10.5194/isprs-annals-IV-2-W4-157-2017.
Keywords:
terrestrial laser scanning, biomorphological segmentation, voxel classification, point-based segmentation, CatBoost, PointNet2, three-dimensional convolutional neural networks, forest inventory.
