Individual tree species classification based on UAV hyperspectral and LiDAR data

doi:10.14088/j.cnki.issn0439-8114.2025.04.001

Abstract

Abstract: UAV LiDAR data were used for individual tree segmentation to obtain tree crown boundaries, and UAV hyperspectral data within the crown boundaries were extracted. Feature fusion schemes were constructed based on band reflectance, vegetation indices, and texture indices, including scheme 1 (band reflectance), scheme 2 (vegetation indices), scheme 3 (texture indices), scheme 4 (band reflectance + vegetation indices), scheme 5 (band reflectance + texture indices), scheme 6 (vegetation indices + texture indices), and scheme 7 (band reflectance + vegetation indices + texture indices). The random forest algorithm was applied to classify individual tree species in the Gudao shelterbelt of the Yellow River Delta, achieving classification of four species: Robinia pseudoacacia, Sophora japonica, Ulmus pumila, and Fraxinus chinensis. The results showed that using only texture indices yielded the worst classification accuracy (0.333) and Kappa coefficient (0.056). Scheme 7 achieved the best classification results, with an accuracy of 0.917 and a Kappa coefficient of 0.887. Scheme 5 achieved a classification accuracy of 0.916 and a Kappa coefficient of 0.886. While maintaining classification accuracy, scheme 5 significantly reduced feature dimensionality. Therefore, the combination of spectral reflectance and spatial texture features was recommended as the optimal scheme.

Key words: UAV, hyperspectral, LiDAR, individual tree level, tree species classification

CLC Number:

S771.8

ZHANG Meng, WANG Hong, LIU Si-si, YANG Wen-cai. Individual tree species classification based on UAV hyperspectral and LiDAR data[J]. HUBEI AGRICULTURAL SCIENCES, 2025, 64(4): 1-6.

References

[1] BONAN G B.Forests and climate change: Forcings, feedbacks, and the climate benefits of forests[J]. Science, 2008, 320: 1444-1449.
[2] 李增元,陈尔学. 中国林业遥感发展历程[J]. 遥感学报,2021, 25(1): 292-301.
[3] LATIFI H, HEURICH M.Multi-scale remote sensing-assisted forest inventory: A glimpse of the state-of-the-art and future prospects[J]. Remote sensing, 2019, 11: 1260.
[4] KHAN A, GUPTA S, GUPTA S K.Multi-hazard Disaster Studies: Monitoring, detection, recovery, and management, based on emerging technologies and optimal techniques[J]. International journal of disaster risk reduction, 2020, 47: 101642.
[5] KOLANUVADA S R, ILANGO K K.Automatic extraction of tree crown for the estimation of biomass from UAV imagery using neural networks[J]. Journal of the indian society of remote sensing, 2021, 49(3): 651-658.
[6] 刘洪麟. 机载高光谱成像仪光谱定标关键技术研究[D].北京:中国科学院大学(中国科学院上海技术物理研究所),2020.
[7] MA Y, ZHAO Y, IM J, et al.A deep-learning-based tree species classification for natural secondary forests using unmanned aerial vehicle hyperspectral images and LiDAR[J]. Ecological indicators, 2024, 159: 111608.
[8] QIN H, ZHOU W, YAO Y, et al.Individual tree segmentation and tree species classification in subtropical broadleaf forests using UAV-based LiDAR, hyperspectral, and ultrahigh-resolution RGB data[J]. Remote sensing of environment, 2022, 280: 113143.
[9] QUAN Y, LI M, HAO Y, et al.Tree species classification in a typical natural secondary forest using UAV-borne LiDAR and hyperspectral data[J]. GIScience & remote sensing, 2023, 60(1):300-304.
[10] JANNE M, SARITA K S, SONJA K, et al.Tree species classification from airborne hyperspectral and LiDAR data using 3D convolutional neural networks[J]. Remote sensing of environment, 2021, 256: 112322.
[11] LI X, LI Y, WANG H, et al.Ecological restoration evaluation of afforestation in Gudao Oilfield based on multi-source remote sensing data[J]. Ecological engineering, 2023, 197: 107107.
[12] SHAO B,LUO J,HE M, et al.Ecological risk assessment at the food web scale:A case study of a mercury contaminated oilfield[J]. Chemosphere, 2020, 260: 127599.
[13] PIIROINEN R, FASSNACHT FE, HEISKANEN J, et al.Invasive tree species detection in the Eastern Arc Mountains biodiversity hotspot using one class classification[J]. Remote sensing of environment, 2018, 218: 119-131.
[14] BALTA H,VELAGIC J,BOSSCHAERTS W,et al.Fast statistical outlier removal based method for large 3D point clouds of outdoor environments[J]. IFAC-papers on line,2018,51(22):348-353.
[15] ZHAO X,GUO Q,SU Y,et al.Improved progressive TIN densification filtering algorithm for airborne LiDAR data in forested areas[J]. ISPRS journal of photogrammetry and remote sensing, 2016, 117: 79-91.
[16] CHEN L, GU Y, ZHANG X, et al.A Normalized spatial-spectral supervoxel segmentation method for multispectral point cloud data[J]. IEEE transactions on geoscience and remote sensing, 2023, 61: 1-11.
[17] SHI Y, SKIDMORE A K, WANG T, et al.Tree species classification using plant functional traits from LiDAR and hyperspectral data[J]. International journal of applied earth observation and geoinformation, 2018, 73: 207-219.
[18] FAUVEL M, BENEDIKTSSON J A, CHANUSSOT J, et al.Spectral and spatial classification of hyperspectral data using SVMs and morphological profiles[J]. IEEE transactions on geoscience and remote sensing, 2008, 46(11):3804-3814.
[19] LU B, HE Y.Evaluating empirical regression, machine learning, and radiative transfer modelling for estimating vegetation chlorophyll content using bi-seasonal hyperspectral images[J]. Remote sensing, 2019, 11: 1979.
[20] BALLANTI L, BLESIUS L, HINES E, et al.Tree species classification using hyperspectral imagery: A comparison of two classifiers[J]. Remote sensing, 2016, 8(6):445.
[21] BLANCO S R, HERAS D B, ARGÜELLO F. Texture extraction techniques for the classification of vegetation species in hyperspectral imagery: Bag of words approach based on superpixels[J]. Remote sensing,2020, 12(16): 2633.
[22] JIAO Z, ZHANG A, SUN G, et al.Hyperspectral image based vegetation index (HSVI): A new vegetation index for urban ecological research[J]. International journal of applied earth observation geoinformation,2021,103:102529.