Compared with ground-based light detection and ranging (LiDAR) data, the differential distribution of the quantity and quality of point cloud data from airborne LiDAR poses difficulties for tree species classification. To verify the feasibility of using the PointNet++ algorithm for point cloud tree species classification with airborne LiDAR data, we selected 11 tree species from the Minjiang River Estuary Wetland Park in Fuzhou City and Sanjiangkou Ecological Park. Training and testing sets were constructed through pre-processing and segmentation, and direct and enhanced down-sampling methods were used for tree species classification. Experiments were conducted to adjust the hyperparameters of the proposed algorithm. The optimal hyperparameter settings used the multi-scale sampling and grouping (MSG) method, down-sampling of the point cloud to 2048 points after enhancement, and a batch size of 16, which resulted in 91.82% classification accuracy. PointNet++ could be used for tree species classification using airborne LiDAR data with an insignificant impact on point cloud quality. Considering the differential distribution of the point cloud quantity, enhanced down-sampling yields improved the classification results compared to direct down-sampling. The MSG classification method outperformed the simplified sampling and grouping classification method, and the number of epochs and batch size did not impact the results.

In this study, the K-means algorithm based on particle swarm optimization (K-PSO) and game theory are introduced to establish the quality evaluation model of a rock mass. Five evaluation factors were considered, i.e., uniaxial saturated compressive strength of rock, discontinuity spacing, acoustic velocity, rock quality designation (RQD), and integrity coefficient. The rock mass of an elevation adit at the abutment of Maji hydropower station was taken as a case study. The subjective weight of the evaluation factor was determined by the weighted least squares method, and the objective weight of the evaluation factor was determined by the entropy method. The combined weights of each influencing factor were determined by game theory to be 0.142, 0.179, 0.035, 0.116, and 0.108. The rock mass quality evaluation in the study area was analyzed by K-PSO algorithm. The results indicate that the K-PSO clustering results are almost the same as the evaluation results of the traditional basic quality (BQ) classification method and the widely used extension evaluation method and are consistent with the preliminary judgment of the expert field. The results are consistent with the field observation law. It is considered that the K-PSO clustering theory can reflect the engineering geological characteristics of the rock mass of the hydropower project in the rock mass quality evaluation.

To address the disorderliness issue of point cloud data when directly used for tree species classification, this study transformed point cloud data into projected images for classification. Building upon this foundation, the influence of incorporating multiple distinct projection perspectives, integrating depth information, and utilising various classification models on the classification of tree point cloud projected images was investigated. Nine tree species in Sanjiangkou Ecological Park, Fuzhou City, were selected as samples. In the single-direction projection classification, the X-direction projection exhibited the highest average accuracy of 80.56%. In the dual-direction projection classification, the XY-direction projection exhibited the highest accuracy of 84.76%, which increased to 87.14% after adding depth information. Four classification models (convolutional neural network, CNN; visual geometry group, VGG; ResNet; and densely connected convolutional networks, DenseNet) were used to classify the datasets, with average accuracies of 73.53%, 85.83%, 87%, and 86.79%, respectively. Utilising datasets with depth and multidirectional information can enhance the accuracy and robustness of image classification. Among the models, the CNN served as a baseline model, VGG accuracy was 12.3% higher than that of CNN, DenseNet had a smaller gap between the average accuracy and the optimal result, and ResNet performed the best in classification tasks.

Remote sensing provides an easy, inexpensive, and rapid method for detecting forest stocks. However, the saturation of data from different satellite sensors leads to low accuracy in estimations of the growing stock volume in natural forests with high densities. Thus, this study added actual data to improve the accuracy. The Daiyun Mountain Reserve was the study area. Landsat 8 operational land imager data were combined with remote sensing data and actual measurements. Multiple linear regression (MLR) and machine learning methods were used to construct a model for estimating the growing stock volume. The decision tree model showed the best fit. By adding the measured data to the model, the saturation could effectively be overcome to a certain extent, and the fitting effect of all the models can be improved. Among the estimation models using only remote sensing data, the normalized difference vegetation index showed the strongest correlation with the model, followed by the annual rainfall and slope. The decision tree model was inverted to produce a map of the accumulation distribution. From the map, the storage volume in the west was lower than that in the east and was primarily confined to the middle-altitude area, consistent with field survey results.

This study introduces the Hierarchical Pixel-Wavele ngth Fusion Network (HPWF-Net), a novel hyperspectral reconst ruction framework designed to address the challenges of tree spe cies classification and spectral reconstruction in complex forest ecosystems. The proposed method employs a hierarchical pixel-wise and target-wavelength-wise reconstruction strategy, progressivel y recovering hyperspectral images by dynamically selecting and r efining target wavelengths at each layer. HPWF-Net utilizes an e nhanced U-Net architecture with multi-level loss supervision and convolutional consistency constraints, enabling accurate reconstr uction by supervising intermediate encoder outputs, convolutiona l regressions, and final decoder predictions. This design preserves consistency with the original multispectral input while enhancin g spectral fidelity and robustness. The model is trained on hypers pectral data from the NEON dataset and applied to Sentinel-2 m ultispectral imagery of the Daiyun Mountain National Nature Re serve. Experimental results demonstrate that the reconstructed h yperspectral data improve tree species classification accuracy by 5.9% compared to the original Sentinel-2 multispectral data. This work contributes to hyperspectral reconstruction and fine-grain ed ecological monitoring, offering a practical solution for remote sensing applications in forest ecosystems. Furthermore, we offer a Python-based package that enable to replicate our results at https://github.com/DarkFlameMaster02/Hierarchical-Pixel-Wave length-Fusion-Network-for-Progressive-Hyperspectral-Image-Re construction.