Using UAV LiDAR Intensity Frequency and Hyperspectral Features to Improve the Accuracy of Urban Tree Species Classification

The accurate classification of tree species in urban forests is crucial for effective management of urban forest resources and ecological environment evaluation. To achieve this goal, unmanned aerial vehicle (UAV) hyperspectral systems and LiDAR systems are useful technologies for monitoring urban forest resources. Accurately classifying tree species in urban forests remains a major challenge due to the limitations of existing methods. While hyperspectral imaging is capable of capturing detailed spectral information, it struggles with the issue of 'same species different spectrum'. On the other hand, the intensity frequency curve also has the problem of 'unsimilar intensity frequency curves for the same species' and 'similar intensity frequency curves for different species'. Therefore, the aim of this study is to overcome these challenges by a new method. The method combines UAV LiDAR intensity frequency features (IF) and hyperspectral features (HF) to accurately classify tree species, abbreviated as HICM. The method uses the complementarity of the two techniques to effectively solve the "same species different spectra" and "same species different frequencies" problems in hyperspectral analysis and intensity frequency analysis, respectively, while ensuring data quality and processing efficiency, significantly improving the accuracy of urban forest tree species classification. Firstly, high-density LiDAR data (230 points/m²) and hyperspectral data are obtained. Then, individual tree crowns are segmented using the deep learning Mask R-CNN algorithm, and IF and HF features are extracted. Finally, a random forest (RF) model is used to classify the 16 main tree species in the study area. The results of the study demonstrate that HICM can accurately classify 16 common urban tree species with an overall accuracy (OA) of 92.0% and kappa of 91.2%. Compared with using only IF or HF for classification, HICM improves OA by 16.9% and 12%, and kappa by 21.6% and 10.1%, respectively. This shows that HF and IF have complementarity, and the combination of these two features effectively addresses the problems of "same species different frequency" and "same species different spectrum".