Terrestrial laser scanning (TLS) is an important technique for tree stem detection. In this paper, a point-based method for stem detection is proposed using single-scan TLS data. One of the main concerns is the point density, which decreases rapidly with the increasing distance to the scanner position. In the proposed method, the search radius is generated adaptively, based on the relationship between the distance and point density, to make sure that the neighborhood maintains a similar scale to the corresponding point density. The belonging of each point is recognized with cuckoo search-based support vector machine, and the points labeled as stem are then clustered and filtered for further verification. The threshold for the small cluster filtering is also adaptive to deal with the problem of the cluster point number decreasing as a function of distance. The stem position is calculated with the lowest cylinder from the cluster segmentation and modeling for the stem mapping. Experiments were carried out on two plots with radii of more than 130 m. The overall detection rate was 76.1%, and 75% of the stems outside 80 m were detected with the adaptive radius, despite the point density being less than 5 cm.

Extreme loadings, a hostile environment and dangerous operation lead to the unsafe state of bridges under construction, especially large-span bridges. Global Navigation Satellite Systems (GNSS) tend to be the best choice for real-time deformation monitoring due to the significant advantage of automation, continuation, all-weather operation and high precision. Unfortunately, the traditional geodetic GNSS instrument with its high price and large volume is limited in its applications. Hence, we design and develop low-cost GNSS equipment by simplifying the monitoring module. The performance of the proposed solution is evaluated through an experimental dynamic scenario, proving its ability to track abrupt deformation down to 3–5 mm. We take Chongqing Guojiatuo Suspension Bridge in China as a case study. We build a real-time low-cost GNSS monitoring cloud platform. The low-cost bridge GNSS monitoring stations are located at the top of the south and north towers, midspan upstream and downstream respectively and the reference station is located in the stable zone 400 m away from the bridge management buildings. We conducted a detailed experimental assessment of low-cost GNSS on 5 April and a real-time deformation detection experiment of the towers and main cables during the dynamic cable saddle pushing process on 26 February 2022. In the static experiment, the standard deviation of the residual using the multi-GNSS solution is 2 mm in the horizontal direction and 5 mm in the vertical direction. The multi-GNSS solution significantly outperforms the BDS/GPS single system. The dynamic experiment shows that, compared with the movement measured by the robotic total station, the horizontal error of the south tower and north tower measured by low-cost GNSS is below 0.005 m and 0.008 m respectively. This study highlights the potential of low-cost GNSS solutions for Structural Health Monitoring (SHM) applications.

According to historical information, more than 300 metal smelting enterprises have been in the southwest of Xiongan for 300 years; however, these polluting enterprises have been gradually closed with the increased intensity of environmental protection. In the paper, 264 soil samples were collected and analyzed in the range of 400 nm–2500 nm by the spectra vista corporation (SVC), and the spectral noise was smoothed by the Savitzky–Golay filter. In order to enhance the spectral differences and curve shapes, mathematical transformations, such as the standard normal variate (SNV), first-order differential (FD), second-order differential (SD), multiple scattering correction (MSC), and continuum removal (CR), were performed on the data, and the correlation between spectral transformation and contents of REEs was analyzed. Moreover, three machine learning models—partial least-squares (PLS), random forest (RF), back propagation neural network (BPNN)—were used to predict the contents of REEs. Experimental results prove that REEs are combined with spectral active substances, such as organic compounds, clay minerals, and iron oxide, and it is possible to determine the contents of REEs using the reflection spectrum. The R² between the predicted values and measured contents reached 0.986 by using BPNN after FD transformation. More importantly, the predicted values basically agree with the actual situation for CASI/SASI airborne hyperspectral images, and this is an effective technique to obtain the contents of REEs in soil at the study area.

Insulators are key components to ensure the normal operation of power facilities in transmission corridors. Existing insulator identification methods mainly use image data and lack the acquisition of three-dimensional information. This paper proposes an efficient insulator extraction method based on UAV (unmanned aerial vehicle) LiDAR (light detection and ranging) point cloud, using five histogram features: horizontal density (HD), horizontal void (HV), horizontal width (HW), vertical width (VW) and vertical void (VV). Firstly, a voxel-based method is employed to roughly extract power lines and pylons from the original point cloud. Secondly, the VV histogram is used to categorize the pylons into suspension and tension types, and the HD histogram is used to locate the tower crossarm and further refine the roughly extracted powerlines. Then, for the suspension tower, insulators are segmented based on the HV histogram and HD difference histogram. For the tension tower, the HW histogram is used to recognize the jumper conductor (JC) and transmission conductor (TC) from the power line. The HW histogram and VW histogram are used to extract the tension insulator in the TC and suspension insulator in the JC, respectively. Finally, considering the problem of setting a suitable grid width when constructing the feature histogram, an adaptive method of multi-scale histograms is proposed to refine the extraction result. Two 220 kV long transmission lines are used for the validation, and the overall object-based accuracy for suspension and tension towers are 100% and 97.3%, respectively. Compared with the point feature-based method, the mean F1 score of the proposed method improved by 0.3, and the runtime for each tower is within 2 s.