Identification of tree species based on the fusion of UAV hyperspectral image and LiDAR data in a coniferous and broad-leaved mixed forest in Northeast China

Abstract

Rapid and accurate identification of tree species via remote sensing technology has become one of the important means for forest inventory. This paper is to develop an accurate tree species identification framework that integrates unmanned airborne vehicle (UAV)-based hyperspectral image and light detection and ranging (LiDAR) data under the complex condition of natural coniferous and broad-leaved mixed forests. First, the UAV-based hyperspectral image and LiDAR data were obtained from a natural coniferous and broad-leaved mixed forest in the Maoer Mountain area of Northeast China. The preprocessed LiDAR data was segmented using a distance-based point cloud clustering algorithm to obtain the point cloud of individual trees; the hyperspectral image was segmented using the projection outlines of individual tree point clouds to obtain the hyperspectral data of individual trees. Then, different hyperspectral and LiDAR features were extracted, respectively, and the importance of the features was analyzed by a random forest (RF) algorithm in order to select appropriate features for the single-source and multi-source data. Finally, tree species identification in the study area were conducted by using a support vector machine (SVM) algorithm together with hyperspectral features, LiDAR features and fused features, respectively. Results showed that the total accuracy for individual tree segmentation was 84.62%, and the fused features achieved the best accuracy for identification of the tree species (total accuracy = 89.20%), followed by the hyperspectral features (total accuracy = 86.08%) and LiDAR features (total accuracy = 76.42%). The optimal features for tree species identification based on fusion of the hyperspectral and LiDAR data included the vegetation indices that were sensitive to the chlorophyll, anthocyanin and carotene contents in the leaves, the partial components of the transformed independent component analysis (ICA), minimum noise fraction (MNF) and principal component analysis (PCA), and the intensity features of the LiDAR echo, respectively. It was concluded that the framework developed in this study was effective in tree species identification under the complex conditions of natural coniferous and broad-leaved mixed forest and the fusion of UAV-based hyperspectral image and LiDAR data can achieve enhanced accuracy compared the single-source UAV-based remote sensing data.

The prediction model of nitrogen nutrition in cotton canopy leaves based on hyperspectral visible‐near infrared band feature fusion

“…Some common feature types are: Spectral features refer to the numerical characteristics extracted from spectral data, including wavelength, reflectivity, and absorptivity. For example, for remote sensing image data, different spectral features can be extracted from different bands, such as NDVI, DDVI, DSI, and other indexes, to characterize the land cover type and environmental change [ 24–27 ] . Spatial features refer to the feature extracted from spatial information, including pixel position, difference between adjacent pixels and texture.…”

“…For example, for remote sensing image data, different spectral features can be extracted from different bands, such as NDVI, DDVI, DSI, and other indexes, to characterize the land cover type and environmental change. [24][25][26][27] Spatial features refer to the feature extracted from spatial information, including pixel position, difference between adjacent pixels and texture. For example, in remote sensing image data, the location coordinates of pixels, texture information and edge information of surrounding pixels can be extracted.…”

Classification of Trees in Hyperspectral Canopy Data Using Machine Learning: Comparative Analysis of Forest Structure Complexity

“…Methods fall into one of the following groups according to the data type they use: (i) spectral data only (Onishi et al, 2022), (ii) structural data only, i.e. point clouds obtained via photogrammetry or LiDAR scanning and the metrics derived from them (Tinkham and Swayze, 2021), and (iii) combining both types of data (Zhong et al, 2022). Nevertheless, gathering structural data from LiDAR scanners remains highly expensive for local applications.…”

“…On the other hand, deep learning methods have shown very good results in hyperspectral image classification (Paoletti et al, 2019). Deep learning techniques, particularly Convolutional Neural Networks (CNNs), have proven to be effective in mapping forests and trees using hyperspectral images (Zhong et al, 2022). By harnessing the capabilities of artificial intelligence (AI) and deep learning, researchers have developed robust models for accurate vegetation mapping.…”

Investigating the potential for detecting Oak Decline using Unmanned Aerial Vehicle (UAV) Remote Sensing

“…In later publications on UAV-based tree species classifications, a combination of these methods is widely adopted. More recent research employs data fusion from multiple UAV sensors to further enhance classification accuracy and improve segmentation for an object-based approach (Moura et al, 2021;Qin et al, 2022;Schiefer et al, 2020;Zhong et al, 2022). The authors of these publications recognise that the high information content are not separable on the same sensor, there is spectral contamination due to broad and overlapping bands (see also Pauly, 2016) as well as the inability to correct for changing light conditions (as downwelling irradiance is not captured).…”