Accurately estimating the mean trace length (mtl) of discontinuities is crucial for three-dimensional (3D) network simulations of random discontinuities. Currently, the mtl is mainly estimated by laying out measurement lines or sampling windows, based on the intersection probability of traces with these lines or windows. However, these methods project actual or interpreted 3D traces onto a 2D plane, which leads to the loss of part of the trace length. This study aims to address this limitation and proposes a 3D mean trace length estimation model (3D-MTLEM) by extending the 2D rectangular sampling window to a 3D cuboid sampling block, enabling the full utilization of the 3D trace information. Firstly, the proposed 3D-MTLEM is applied to three 3D trace simulated datasets. The results indicate that compared to traditional 2D window estimation methods (such as Laslett, Huang-Huang (HH), Generalized Huang-Huang (GHH), Kulatilake, Multi-scanline, and Circular window methods (Circle)), the 3D-MTLEM provides more accurate and stable estimates of mtl. Furthermore, the 3D-MTLEM conducts an analysis of the influence of four key factors: trace length (tl), trace direction (td), trace length distribution function (tldf), and block size (bs). Subsequently, the 3D-MTLEM is applicable to two case studies of artificial quarry slope, further confirming its applicability in practical engineering.
Extracting road information from high-resolution remote sensing images (HRI) can provide crucial geographic information for many applications. With the improvement of remote sensing image resolution, the image data contain more abundant feature information. However, this phenomenon also enhances the spatial heterogeneity between different types of roads, making it difficult to accurately discern the road and non-road regions using only spectral characteristics. To remedy the above issues, a novel residual attention and local context-aware network (RALC-Net) is proposed for extracting a complete and continuous road network from HRI. RALC-Net utilizes a dual-encoder structure to improve the feature extraction capability of the network, whose two different branches take different feature information as input data. Specifically, we construct the residual attention module using the residual connection that can integrate spatial context information and the attention mechanism, highlighting local semantics to extract local feature information of roads. The residual attention module combines the characteristics of both the residual connection and the attention mechanism to retain complete road edge information, highlight essential semantics, and enhance the generalization capability of the network model. In addition, the multi-scale dilated convolution module is used to extract multi-scale spatial receptive fields to improve the model’s performance further. We perform experiments to verify the performance of each component of RALC-Net through the ablation study. By combining low-level features with high-level semantics, we extract road information and make comparisons with other state-of-the-art models. The experimental results show that the proposed RALC-Net has excellent feature representation ability and robust generalizability, and can extract complete road information from a complex environment.
A rock slope can be characterized by tens of persistent discontinuities. A slope can be massive. The slip surface of the slope is usually easier to expand along with the discontinuities because the shear strength of the discontinuities is substantially lower than that of the rock blocks. Based on this idea, this paper takes a jointed rock slope in Hengqin Island, Zhuhai as an example, and establishes a three-dimensional (3D) model of the studied slope by digital close-range photogrammetry to rapidly interpret 222 fracture parameters. Meanwhile, a new Floyd algorithm for finding the shortest path is developed to realize the critical slip surface identification of the studied slope. Within the 3D fracture network model created using the Monte Carlo method, a sequence of cross-sections is placed. These cross-sections containing fractures are used to search for the shortest paths between the designated shear entrances and exits. For anyone combination of entry point and exit point, the shortest paths corresponding to different cross-sections are different and cluttered. For the sake of safety and convenience, these shortest paths are simplified as a circular arc that is regarded as a potential slip surface. The fracture frequency is used to determine the probability of sliding along a prospective critical slip surface. The potential slip surface through the entrance point (0, 80) and exit point (120, 0) is identified as the final critical slip surface of the slope due to the maximum fracture frequency.
Due to earthquakes and large-scale exploitation of oil, gas, groundwater, and coal energy, large-scope surface deformation has occurred in Songyuan City, Jilin Province, China, and it is posing a serious threat to sustainable development, including urban development, energy utilization, environmental protection, and construction to improve saline–alkali land. In this study, we selected the Chagan Lake region, which is located in Songyuan City, as our research area. Using temporarily coherent point synthetic aperture radar interferometry (TCPInSAR), we obtained a time series of land surface deformation and the deformation rate in this area from 20 ALOS PALSAR images from 2006 to 2010. The results showed that the deformation rate in the Chagan Lake region ranged from −46.7 mm/year to 41.7 mm/year during the monitoring period. In three typical land cover areas of the Chagan Lake region, the subsidence in the wetland area was larger than that in the saline–alkali area, while the highway experienced a small uplift. In addition, surface deformation in lakeside areas with or without dykes was different; however, as this was mainly affected by soil freeze–thaw cycles and changes in groundwater level, the deformation showed a negative correlation with temperature and precipitation. By monitoring and analyzing surface deformation, we can provide a data reference and scientific basis for sustainable ecological and economic development in the Chagan Lake region and adjacent areas.
We propose an improved ant colony algorithm for avoiding obstacles in complex static environments that addresses the problems of a single evaluation factor and low path quality of the traditional ant colony algorithm in path planning. The improvements are: 1) a fuzzy planner is constructed according to the comprehensive evaluation method of fuzzy mathematics and the analytic hierarchy process to comprehensively evaluate and determine the impact of environmental factors, 2) the probability selection formula of the ant colony algorithm is optimized, 3) the pheromone update formula is optimized, and 4) the corner system mechanism is introduced as a post-processing method of path optimization to further smooth the path. Results from simulation experiments of the traditional ant colony algorithm were analysed and compared with those of the improved ant colony algorithm, showing that the latter has a stronger path planning ability and higher algorithm efficiency, resulting in a smoother path with a lower negative impact by environmental factors. Thus, the proposed algorithm is expected to provide a computational basis for effective multi-factor path planning in realistic environments, thereby saving human and material resources.
Abstract The exploitation of underground fluid is an important factor leading to land subsidence. The effects of mining depth, frequency, and mode on land subsidence are also different. The objective of this study was to develop a multisource method—including optical remote sensing interpretation, Interferometric Synthetic Aperture Radar (InSAR) technology, and unmanned aerial vehicle (UAV)—to reveal the long-term temporal and spatial evolution law of subsidence characteristics driven by groundwater and oil extraction, as well as to reveal the formation mechanism and seasonal response law of land subsidence under the action of different driving factors. In this paper, we select the western region of Jilin Province located in Songnen Plain as the study area. The subsidence funnels in the study area are distributed in a porphyritic manner, and the distribution of the subsidence funnels has a certain correlation with the distribution of the pumping wells. In farmland areas, the subsidence is mainly caused by pumping groundwater. The annual land subsidence rate in the study area is -3.14 mm/a, and the maximum deformation rate in the study area is -22.05 mm/a. The subsidence is affected by the season, shown by the fact that it rises in the dry season and decreases in the rainy season. The subsidence in the west of Songnen Plain is caused by oil pumping and groundwater pumping, and groundwater pumping is dominant. The exploitation of underground fluid transfers the pressure borne by water or oil to the soil skeleton so as to increase and consolidate the effective stress of the soil layer and lead to land subsidence. The continuous observation of the surface in the western area of Songnen Plain is helpful to guide the safe production of agriculture and industry and ensure the smooth development of local industry and agriculture.
We present both a theoretical and a methodological framework that addresses a critical challenge in applying deep learning to physical systems: the reconciliation of non-linear expressiveness with SO(3)-equivariance in predictions of SO(3)-equivariant quantities, such as the electronic-structure Hamiltonian. Inspired by covariant theory in physics, we address this problem by exploring the mathematical relationships between SO(3)-invariant and SO(3)-equivariant quantities and their representations. We first construct theoretical SO(3)-invariant quantities derived from the SO(3)-equivariant regression targets, and use these invariant quantities as supervisory labels to guide the learning of high-quality SO(3)-invariant features. Given that SO(3)-invariance is preserved under non-linear operations, the encoding process for invariant features can extensively utilize non-linear mappings, thereby fully capturing the non-linear patterns inherent in physical systems. Building on this foundation, we propose a gradient-based mechanism to induce SO(3)-equivariant encodings of various degrees from the learned SO(3)-invariant features. This mechanism can incorporate non-linear expressive capabilities into SO(3)-equivariant representations, while theoretically preserving their equivariant properties as we prove. Our approach offers a promising general solution to the critical dilemma between equivariance and non-linear expressiveness in deep learning methodologies. We apply our theory and method to the electronic-structure Hamiltonian prediction tasks, demonstrating state-of-the-art performance across six benchmark databases.
In recent decades, high-resolution (HR) remote sensing images have shown considerable potential for providing detailed information for change detection. The traditional change detection methods based on HR remote sensing images mostly only detect a single land type or only the change range, and cannot simultaneously detect the change of all object types and pixel-level range changes in the area. To overcome this difficulty, we propose a new coarse-to-fine deep learning-based land-use change detection method. We independently created a new scene classification dataset called NS-55, and innovatively considered the adaptation relationship between the convolutional neural network (CNN) and the scene complexity by selecting the CNN that best fit the scene complexity. The CNN trained by NS-55 was used to detect the category of the scene, define the final category of the scene according to the majority voting method, and obtain the changed scene by comparison to obtain the so-called coarse change result. Then, we created a multi-scale threshold (MST) method, which is a new method for obtaining high-quality training samples. We used the high-quality samples selected by MST to train the deep belief network to obtain the pixel-level range change detection results. By mapping coarse scene changes to range changes, we could obtain fine multi-type land-use change detection results. Experiments were conducted on the Multi-temporal Scene Wuhan dataset and aerial images of a particular area of Dapeng New District, Shenzhen, where promising results were achieved by the proposed method. This demonstrates that the proposed method is practical, easy-to-implement, and the NS-55 dataset is physically justified. The proposed method has the potential to be applied in the large scale land use fine change detection problem and qualitative and quantitative research on land use/cover change based on HR remote sensing data.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)