Unmanned Aerial Vehicle (UAV) photogrammetry is an effective method for acquiring terrain information. However, in hillside areas, the terrain is complex, and the altitude varies greatly. The mission flight is planned by using equal altitude; in the actual shooting, the geometry and resolution of the pixel within the same image or between adjacent images will be inconsistent due to the different shooting distances. The number and accuracy of point clouds are affected. We propose a contour mission flight plan method, which involves designing flight plans based on the existing digital elevation model (DEM) and the desired flight altitude. This method for aerial photography is more effective in maintaining a consistent ground shooting distance during image capture. Experiments were conducted using a simulated DEM and the undulating terrain of the Kaohsiung Liugui area in Taiwan to verify the effect of contour mission flight planning in the hillside area. The results show that, due to the significant variation of terrain in the hillside area, the use of a contour mission flight plan for aerial photography can be more consistent with the originally planned altitude but requires more planning and operating time. The minor height difference, higher overlap, and improved accuracy of the results show that contour mission planning can provide a suitable solution for UAVs in hillside areas.

The southwestern foothills in Taiwan contain abundant mudstone, which softens or slakes easily when contacting water. Thereafter, this condition causes high rates of erosion during heavy rain, eventually leading to the formation of badlands. To increase land use value in Taiwan, this study aimed to offer a new solution with the use of compaction techniques and also adding a small amount of cement as soil amendment. We examined the feasibility of this concept by performing a series of tests, including the basic physical property test, compaction test, unconfined compression test, static triaxial CU and UU tests, consolidation test, California bearing ratio (CBR) test, and triaxial permeability test. The specimens were created based on the controlled moisture at the wet side along the compaction curve. Three clods, three preparation methods, and two types of cement content were used for specimen preparation. The test results revealed that clod will affect the results of unconfined compression strength and soil permeability. For mudstone with added cement, its unconfined compression strength increased 7- to 10-fold and its shear strength increased 2- to 3-fold. The results of consolidation and the CBR test showed that improvement is possible by using low amounts of cement. Overall, the present method not only adheres to low-carbon and environmental protection requirements, but also verifies the feasibility of using compacted mudstone as an engineering material.

The creation of building information models requires acquiring real building conditions. The generation of a three-dimensional (3D) model from 3D point clouds involves classification, outline extraction, and boundary regularization for semantic segmentation. The number of 3D point clouds generated using close-range images is smaller and tends to be unevenly distributed, which is not conducive to automated modeling processing. In this paper, we propose an efficient solution for the semantic segmentation of indoor point clouds from close-range images. A 3D deep learning framework that achieves better results is further proposed. A dynamic graph convolutional neural network (DGCNN) 3D deep learning method is used in this study. This method was selected to learn point cloud semantic features. Moreover, more efficient operations can be designed to build a module for extracting point cloud features such that the problem of inadequate beam and column classification can be resolved. First, DGCNN is applied to learn and classify the indoor point cloud into five categories: columns, beams, walls, floors, and ceilings. Then, the proposed semantic segmentation and modeling method is utilized to obtain the geometric parameters of each object to be integrated into building information modeling software. The experimental results show that the overall accuracy rates of the three experimental sections of Area_1 in the Stanford 3D semantic dataset test results are 86.9%, 97.4%, and 92.5%. The segmentation accuracy of corridor 2F in a civil engineering building is 94.2%. In comparing the length with the actual on-site measurement, the root mean square error is found to be ±0.03 m. The proposed method is demonstrated to be capable of automatic semantic segmentation from 3D point clouds with indoor close-range images.

Some researchers in past years have tried to develop a simplified method for analyzing soil liquefaction. However, the correctness of the pore water pressure in the model will affect the results. In addition, the formulas derived are not easy, and the exact parameters of the model are difficult to obtain. This study used a mass-spring-damping system to simulate the repeated strain of liquefaction cyclic triaxial tests. Because the model is simple and the parameters are easy to understand and obtain, it also shows the extensibility of this model. During the parameter study, damping coefficient c and spring coefficient k parameters decreased with the increasing cyclic number. Preliminary results of the research show that this model can further simulate the repeated strain obtained by cyclic triaxial tests without considering the variation of effective stress during cyclic loading. Four samples were used to verify the model’s correctness, and their boring sites were found in Yunlin areas, Taiwan. Simulation results show that the spring-damping system is feasible for simulated cyclic triaxial tests because the simulated results correlate to the testing results in trend. Generally, the first cycle number simulation will be less accurate because the pore water pressure of the specimen changes rapidly when the performance has just started. In contrast, the increase in subsequent cycles may be biased due to cyclic stress variation and soil plasticity during simulation. In the future, pure sand specimens created in the laboratory will be suggested for simulation.

The Taiwan area of mostly rugged mountains and special geological conditions induced catastrophic slope deformation by heavy rainfall and earthquakes. Unstable hillslopes present a significant problem especially to buildings because it affects the living safety. This article mainly discusses regions with buildings in a wide hillslope and proposes a set of hierarchical monitoring procedure. First, persistent scatters interferometric synthetic aperture radar (PS-InSAR) technology observes a wide region of displacement by SAR images, hence detects result of severe displacement regions as the interested areas. Furthermore, multi-rotors UAV (MUAV) captures high-resolution images calculation deformation in the interested areas, which evaluates the serious displacement as a high-risk region. Finally, the self-develop surveillance cameras monitoring system achieves real-time displacement and behavior in the high-risk regions. The study area is located in the campus. From the hierarchical monitoring procedure results, PS-InSAR investigation in the large region is found to be reliable and reasonable, which can provide MUAV flight an effective reference. Both the PS-InSAR and MUAV showed a ground maximum displacement of 30 mm yr-1 (LOS velocity) in the interested areas where ground displacement is higher than the building. The advantages of surveillance cameras include obtaining real-time displacement and providing clear surface erosion behavior. This study experimentally demonstrated and provided a reliable hierarchical monitoring procedure and deformation information of different scales through three remote-sensing techniques.

This study uses a mass-spring-damping system to simulate the repeated strain of liquefaction cyclic triaxial tests. The two results are compared in this paper to understand the feasibility of the mass-spring-damping system developed in this paper and the feasibility of future research on this related topic and development potential. The main factors affecting this mode's repeated strain are the spring coefficient k and the external force Q0. The spring coefficient has an inverse relationship but does not increase in multiples. The external force has a direct proportional relationship but does not increase the result- a non-multiple increase. When the pore water pressure of the liquefied cyclic triaxial test specimen rises, it will cause the specimen to liquefy, decreasing effective stress, shear modulus G, and damping ratio D, causing an increase in strain. The shear modulus and damping ratio are related to the spring coefficients k and c. Both are variables that change with time. This article takes an original thin-tube specimen at point A in the Yunlin area in Taiwan as a testing example. Through Mathematica software, it can be obtained that the mass m=1kg, the spring coefficient k=244e-0.1t kgf/cm, and the damping coefficient c=0.739e-0.257t kgf-s/cm and external force Q0=10.1sin2πt kg; Finally, this study selected the original thin-tube specimens from four locations in the Yunlin area, and simulated the repeated strain amount of the cyclic triaxial test specimens through a spring-damping system. The results show that the spring-damping system is feasible for simulated cyclic triaxial tests because the model is simple and the parameters are easy to understand and obtain, which also shows the extensibility of this model. Preliminary results of the research show that this model can further simulate the repeated strain obtained by cyclic triaxial tests without considering the increased trend in pore water pressure and the decrease in effective stress during cyclic loading.