Feasibility Study on the Classification of Persimmon Trees’ Components Based on Hyperspectral LiDAR

Abstract

Intelligent management of trees is essential for precise production management in orchards. Extracting components’ information from individual fruit trees is critical for analyzing and understanding their general growth. This study proposes a method to classify persimmon tree components based on hyperspectral LiDAR data. We extracted nine spectral feature parameters from the colorful point cloud data and performed preliminary classification using random forest, support vector machine, and backpropagation neural network methods. However, the misclassification of edge points with spectral information reduced the accuracy of the classification. To address this, we introduced a reprogramming strategy by fusing spatial constraints with spectral information, which increased the overall classification accuracy by 6.55%. We completed a 3D reconstruction of classification results in spatial coordinates. The proposed method is sensitive to edge points and shows excellent performance for classifying persimmon tree components.

Multispectral Light Detection and Ranging Technology and Applications: A Review

“…Wallace et al [ 43 ] proposed a prototype HSL system, leveraging a super-continuum laser source to have four laser wavelengths in conjunction with Time-Correlated Single Photon Counting (TCSPC) receiver technology, which harnesses the advantages of improved depth resolution and sensitivity of the TCSPC technique. In addition, to enhance the spectral resolution of HSL scanners, innovative Acoustic–Optical Tuneable Filter-based terrestrial HSL systems (AOTF-HSL) have recently been proposed [ 22 , 90 , 91 , 92 , 93 , 94 ]. In such laser scanners, AOTF acts as a spectral bandpass filter on the outgoing laser from the super-continuous laser in the emission unit.…”

“…Abbreviations: P = Point cloud; I = Image; DBM = Deep Boltzmann Machine; NDFI = Normalized Difference Feature Index; PRI = Photochemical Reflectance Index; MNDWI = Modified Normalized Difference Water Index; CHM = Canopy Height Model; LDA = Linear Discriminant Analysis; SVM = Support Vector Machine; RF = Random Forest. Publications Data Sources Application Feature(s) Type Method(s) Hakula et al, 2023 [ 68 ] HeliALS-TW Tree species classification Several geometric, radiometric and return type features P Object-based RF Rana et al, 2023 [ 114 ] Optech Titan Monitoring seedling stands Geometric, intensities P Linear regression Axelsson et al, 2023 [ 138 ] VQ-1560i-DW Tree species classification and stem volumes estimation Several geometric, radiometric and return type features P LDA and k-nearest neighbors models Shao al., 2023 [ 90 ] AOTF-HSL Persimmon tree components classification Intensities, average reflectance, CI red edge, NDVI, NDRE P RF, SVM, and backpropagation neural network methods Xiao et al, 2022 [ 119 ] Optech Titan LULC classification Z, intensity, and SVD-based and deep features P Improved RandLA-Net Tian et al, 2022 [ 110 ] …”