Classification of Typical Tree Species in Laser Point Cloud Based on Deep Learning

We propose the Point Cloud Tree Species Classification Network (PCTSCN) to overcome challenges in classifying tree species from laser data with deep learning methods. The network is mainly composed of two parts: a sampling component in the early stage and a feature extraction component in the later stage. We used geometric sampling to extract regions with local features from the tree contours since these tend to be species-specific. Then we used an improved Farthest Point Sampling method to extract the features from a global perspective. We input the intensity of the tree point cloud as a dimensional feature and spatial information into the neural network and mapped it to higher dimensions for feature extraction. We used the data obtained by Terrestrial Laser Scanning (TLS) and Unmanned Aerial Vehicle Laser Scanning (UAVLS) to conduct tree species classification experiments of white birch and larch. The experimental results showed that in both the TLS and UAVLS datasets, the input tree point cloud density and the highest feature dimensionality of the mapping had an impact on the classification accuracy of the tree species. When the single tree sample obtained by TLS consisted of 1024 points and the highest dimension of the network mapping was 512, the classification accuracy of the trained model reached 96%. For the individual tree samples obtained by UAVLS, which consisted of 2048 points and had the highest dimension of the network mapping of 1024, the classification accuracy of the trained model reached 92%. TLS data tree species classification accuracy of PCTSCN was improved by 2–9% compared with other models using the same point density, amount of data and highest feature dimension. The classification accuracy of tree species obtained by UAVLS was up to 8% higher. We propose PCTSCN to provide a new strategy for the intelligent classification of forest tree species.

[1]  Zhengjun Liu,et al.  Tree species classification of LiDAR data based on 3D deep learning , 2021 .

[2]  T. Kneib,et al.  Predicting Tree Species From 3D Laser Scanning Point Clouds Using Deep Learning , 2021, Frontiers in Plant Science.

[3]  C. Hopkinson,et al.  See the forest and the trees: Effective machine and deep learning algorithms for wood filtering and tree species classification from terrestrial laser scanning , 2020 .

[4]  Julia Armesto,et al.  Individual Tree Detection in a Eucalyptus Plantation Using Unmanned Aerial Vehicle (UAV)-LiDAR , 2020, Remote. Sens..

[5]  Sanjeev Arora,et al.  An Exponential Learning Rate Schedule for Deep Learning , 2019, ICLR.

[6]  Wen Xiao,et al.  Mean Shift Segmentation Assessment for Individual Forest Tree Delineation from Airborne Lidar Data , 2019, Remote. Sens..

[7]  Dacheng Tao,et al.  Control Batch Size and Learning Rate to Generalize Well: Theoretical and Empirical Evidence , 2019, NeurIPS.

[8]  Lin Cao,et al.  An Automated Hierarchical Approach for Three-Dimensional Segmentation of Single Trees Using UAV LiDAR Data , 2018, Remote. Sens..

[9]  Wei Wu,et al.  PointCNN: Convolution On X-Transformed Points , 2018, NeurIPS.

[10]  Tiziano Ghisu,et al.  An automated approach for wood-leaf separation from terrestrial LIDAR point clouds using the density based clustering algorithm DBSCAN , 2018, Agricultural and Forest Meteorology.

[11]  Laura E. Chasmer,et al.  Filtering Stems and Branches from Terrestrial Laser Scanning Point Clouds Using Deep 3-D Fully Convolutional Networks , 2018, Remote. Sens..

[12]  Miro Govedarica,et al.  Model of Point Cloud Data Management System in Big Data Paradigm , 2018, ISPRS Int. J. Geo Inf..

[13]  Cewu Lu,et al.  PointSIFT: A SIFT-like Network Module for 3D Point Cloud Semantic Segmentation , 2018, ArXiv.

[14]  Lei Wang,et al.  MSNet: Multi-Scale Convolutional Network for Point Cloud Classification , 2018, Remote. Sens..

[15]  Leslie N. Smith,et al.  A disciplined approach to neural network hyper-parameters: Part 1 - learning rate, batch size, momentum, and weight decay , 2018, ArXiv.

[16]  Hamid Hamraz,et al.  Deep learning for conifer/deciduous classification of airborne LiDAR 3D point clouds representing individual trees , 2018, ISPRS Journal of Photogrammetry and Remote Sensing.

[17]  Xi Zhu,et al.  Foliar and woody materials discriminated using terrestrial LiDAR in a mixed natural forest , 2018, Int. J. Appl. Earth Obs. Geoinformation.

[18]  P. Knapp,et al.  Divergent growth rates of alpine larch trees (Larix lyallii Parl.) in response to microenvironmental variability , 2018 .

[19]  Quoc V. Le,et al.  Don't Decay the Learning Rate, Increase the Batch Size , 2017, ICLR.

[20]  Tsuyoshi Inoue,et al.  Lidar-based individual tree species classification using convolutional neural network , 2017, Optical Metrology.

[21]  Leonidas J. Guibas,et al.  PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space , 2017, NIPS.

[22]  Jixian Zhang,et al.  Advances in fusion of optical imagery and LiDAR point cloud applied to photogrammetry and remote sensing , 2017 .

[23]  B. Chandra,et al.  Deep learning with adaptive learning rate using laplacian score , 2016, Expert Syst. Appl..

[24]  Nicholas C. Coops,et al.  Tree species classification in subtropical forests using small-footprint full-waveform LiDAR data , 2016, Int. J. Appl. Earth Obs. Geoinformation.

[25]  Wuming Zhang,et al.  An Easy-to-Use Airborne LiDAR Data Filtering Method Based on Cloth Simulation , 2016, Remote. Sens..

[26]  H. Spiecker,et al.  Non Destructive Method for Biomass Prediction Combining TLS Derived Tree Volume and Wood Density , 2015 .

[27]  Michael J. Falkowski,et al.  A review of methods for mapping and prediction of inventory attributes for operational forest management , 2014 .

[28]  T. Ohta,et al.  Growth and physiological responses of larch trees to climate changes deduced from tree-ring widths and δ13C at two forest sites in eastern Siberia , 2014 .

[29]  Gregory P. Asner,et al.  Observing Changing Ecological Diversity in the Anthropocene , 2013 .

[30]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[31]  L. Bruzzone,et al.  Tree species classification in the Southern Alps based on the fusion of very high geometrical resolution multispectral/hyperspectral images and LiDAR data , 2012 .

[32]  Barbara Koch,et al.  Investigating multiple data sources for tree species classification in temperate forest and use for single tree delineation , 2012, Int. J. Appl. Earth Obs. Geoinformation.

[33]  O. Zyryanova,et al.  White Birch Trees as Resource Species of Russia : Their Distribution, Ecophysiological Features, Multiple Utilizations , 2010 .

[34]  Roberta E. Martin,et al.  Airborne spectranomics: mapping canopy chemical and taxonomic diversity in tropical forests , 2009 .

[35]  P. Axelsson DEM Generation from Laser Scanner Data Using Adaptive TIN Models , 2000 .