Multispectral airborne laser scanning for tree species classification: a benchmark of machine learning and deep learning algorithms

Climate-smart and biodiversity-preserving forestry demands precise information on forest resources, extending to the individual tree level. Multispectral airborne laser scanning (ALS) has shown promise in automated point cloud processing and tree segmentation, but challenges remain in identifying rare tree species and leveraging deep learning techniques. This study addresses these gaps by conducting a comprehensive benchmark of machine learning and deep learning methods for tree species classification. For the study, we collected high-density multispectral ALS data (>1000 pts/m$^2$) at three wavelengths using the FGI-developed HeliALS system, complemented by existing Optech Titan data (35 pts/m$^2$), to evaluate the species classification accuracy of various algorithms in a test site located in Southern Finland. Based on 5261 test segments, our findings demonstrate that point-based deep learning methods, particularly a point transformer model, outperformed traditional machine learning and image-based deep learning approaches on high-density multispectral point clouds. For the high-density ALS dataset, a point transformer model provided the best performance reaching an overall (macro-average) accuracy of 87.9% (74.5%) with a training set of 1065 segments and 92.0% (85.1%) with 5000 training segments. The best image-based deep learning method, DetailView, reached an overall (macro-average) accuracy of 84.3% (63.9%), whereas a random forest (RF) classifier achieved an overall (macro-average) accuracy of 83.2% (61.3%). Importantly, the overall classification accuracy of the point transformer model on the HeliALS data increased from 73.0% with no spectral information to 84.7% with single-channel reflectance, and to 87.9% with spectral information of all the three channels.

@article{taher2025_2504.14337,
  title={ Multispectral airborne laser scanning for tree species classification: a benchmark of machine learning and deep learning algorithms },
  author={ Josef Taher and Eric Hyyppä and Matti Hyyppä and Klaara Salolahti and Xiaowei Yu and Leena Matikainen and Antero Kukko and Matti Lehtomäki and Harri Kaartinen and Sopitta Thurachen and Paula Litkey and Ville Luoma and Markus Holopainen and Gefei Kong and Hongchao Fan and Petri Rönnholm and Antti Polvivaara and Samuli Junttila and Mikko Vastaranta and Stefano Puliti and Rasmus Astrup and Joel Kostensalo and Mari Myllymäki and Maksymilian Kulicki and Krzysztof Stereńczak and Raul de Paula Pires and Ruben Valbuena and Juan Pedro Carbonell-Rivera and Jesús Torralba and Yi-Chen Chen and Lukas Winiwarter and Markus Hollaus and Gottfried Mandlburger and Narges Takhtkeshha and Fabio Remondino and Maciej Lisiewicz and Bartłomiej Kraszewski and Xinlian Liang and Jianchang Chen and Eero Ahokas and Kirsi Karila and Eugeniu Vezeteu and Petri Manninen and Roope Näsi and Heikki Hyyti and Siiri Pyykkönen and Peilun Hu and Juha Hyyppä },
  journal={arXiv preprint arXiv:2504.14337},
  year={ 2025 }
}