Now, most existing dynamic RGB-D SLAM methods are based on deep learning or mathematical models. Abundant training sample data is necessary for deep learning, and the selection diversity of semantic ...samples and camera motion modes are closely related to the robust detection of moving targets. Furthermore, the mathematical models are implemented at the feature-level of segmentation, which is likely to cause sub or over-segmentation of dynamic features. To address this problem, different from most feature-level dynamic segmentation based on mathematical models, a non-prior semantic dynamic segmentation based on a particle filter is proposed in this paper, which aims to attain the motion object segmentation. Firstly, GMS and optical flow are used to calculate an inter-frame difference image, which is considered an observation measurement of posterior estimation. Then, a motion equation of a particle filter is established using Gaussian distribution. Finally, our proposed segmentation method is integrated into the front end of visual SLAM and establishes a new dynamic SLAM, PFD-SLAM. Extensive experiments on the public TUM datasets and real dynamic scenes are conducted to verify location accuracy and practical performances of PFD-SLAM. Furthermore, we also compare experimental results with several state-of-the-art dynamic SLAM methods in terms of two evaluation indexes, RPE and ATE. Still, we provide visual comparisons between the camera estimation trajectories and ground truth. The comprehensive verification and testing experiments demonstrate that our PFD-SLAM can achieve better dynamic segmentation results and robust performances.
The dynamic mechanical response of block-in-matrix-soils (bimsoils) is crucial to the long-term stability predication of constructions and building structures made of bimsoils. Although the dynamic ...failure behaviors of bimsoil under cyclic loads have been widely understood, the influence of loading frequencies on the macro-meso structural deterioration governing localized band propagation have not been well understood. Herein, multistage cyclic triaixal tests and post-test X-ray computed tomography (CT) scanning have been performed on bimsoil with various loading frequencies (f = 0.1 Hz, 0.2 Hz, 1.0 Hz, and 2.0 Hz). The cyclic triaxial tests simulated the behavior of bimsoil material under varied velocities of passing vehicle herein. The macro-meso fatigue failure characteristics were analyzed in terms of stress-strain responses, stiffness degradation, and meso-structural changes. Testing results show that the applied loading frequency alters the coupling degree of soil matrix and rock blocks, and damage develops fully within bimsoil under relatively low loading frequency. In addition, the failure mode of bimsoil changes from bulging failure to shear failure with the increases of loading frequency, and the volumetric deformation is the largest for a bimsoil sample subjected to low loading frequency. Moreover, post-test CT scanning reveals the meso-structural degradation within bimsoil, the multiple interface damage and cracking contribute a lot to the strength, volumetric deformation and failure modes. It is suggested the contact, interlocking, occlusion, and separation among the rock blocks promote the volume dilatancy when subjecting to low loading frequency.
•Macro-meso fatigue failure of bimsoil under multi-level cyclic loads was experimentally investigated.•The applied loading frequency alters the coupling degree of soil matrix and rock blocks.•Failure mode of bimsoil changes from bulging failure to shear failure with the increases of loading frequency.•Multiple interface damage and cracking contribute a lot to volumetric deformation and failure evolution.
RGB-D SLAM (Simultaneous Localization and Mapping) generally performs smoothly in a static environment. However, in dynamic scenes, dynamic features often cause wrong data associations, which degrade ...accuracy and robustness. To address this problem, in this paper, a new RGB-D dynamic SLAM method, PLD-SLAM, which is based on point and line features for dynamic scenes, is proposed. First, to avoid under-over segmentation caused by deep learning, PLD-SLAM combines deep learning for semantic information segmentation with the K-Means clustering algorithm considering depth information to detect the underlying dynamic features. Next, two consistency check strategies are utilized to check and filter out the dynamic features more reasonably. Then, to obtain a better practical performance, point and line features are utilized to calculate camera pose in the dynamic SLAM, which is also different from most published dynamic SLAM algorithms based merely on point features. The optimization model with point and line features is constructed and utilized to calculate the camera pose with higher accuracy. Third, enough experiments on the public TUM RGB-D dataset and the real-world scenes are conducted to verify the location accuracy and performance of PLD-SLAM. We compare our experimental results with several state-of-the-art dynamic SLAM methods in terms of average localization errors and the visual difference between the estimation trajectories and the ground-truth trajectories. Through the comprehensive comparisons with these dynamic SLAM schemes, it can be fully demonstrated that PLD-SLAM can achieve comparable or better performances in dynamic scenes. Moreover, the feasibility of camera pose estimation based on both point features and line features has been proven by the corresponding experiments from a comparison with our proposed PLD-SLAM only based on point features.
The natural fractures in rock mass are susceptible to damage evolution when subjecting to repeated freeze-thaw (F-T) weathering in cold regions, which can lead to the instability of rock engineering ...and even occurrence of geological hazards. Knowledge of how natural fracture impacts the overall fracture evolution of freeze-thawed rock is important to predict the stability of rock structure. In this work, we reported uniaxial experimental measurements of the changes in strength, deformation, acoustic emission (AE) pattern, and Felicity effect during increasing amplitude stress-cycling conditions on granite. The results show that the change of fracture aperture is related to the fracture openness and filling characteristics, open-type fracture is sensitive to F-T treatment, and its aperture increases faster than the close-type and fill-type fracture. In addition, strength decreases, and the damping characteristics first decrease and then increase with increasing natural fracture volume. AE activities also present different responses during sample deformation. The proportion of AE signals having low-frequency characteristics increases with increasing natural fracture volume, and the shear sliding along natural fracture results in the surge of AE activities. Moreover, the Felicity effect indicates that the Felicity ratio presents a fluctuation decreasing trend, and the preexisting fractures alter the stress memory characteristics of rock. It is suggested that the changes of the geomechanical and AE pattern are the interactions between the natural fracture and the newly stimulated fracture. The testing results are expected to improve the understanding of the influence of natural fractures on rock fracture evolution and can be helpful to predict the stability of rock structures and rock mass in cold regions.
Celotno besedilo
Dostopno za:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, UILJ, UKNU, UL, UM, UPUK
•Definition of multidimensional semantic features of the building façade.•Multidimensional virtual semantic feature map are constructed based on Multiple View Stereo techniques.•Ensemble-learning are ...used for multi-semantic feature block pre-segmentation.•Color&Texture, orientation and curvature semantics segmentation.•Different virtual semantic feature map and clustering strategies are combined to accurately extract of diverse building façade elements.
Building façade elements are an important foundation for smart cities. As buildings exhibit an array of textures and geometric forms, the process of image acquisition is easily affected, although the robustness of texture in scenes (e.g., dilapidated buildings) is poor, with high point cloud data, and low recognition efficiency; therefore, the accuracy of building element extraction based on a single data source remains limited. In this research, a method for building façade element extraction based on multidimensional virtual semantic feature map ensemble learning and hierarchical clustering is proposed. Point clouds were obtained by multi-view images, and then the multidimensional virtual semantic feature maps, including color, texture, orientation, and curvature semantics, were acquired via reprojection. The multi-semantic feature block pre-segmentation, considering multiple features, was obtained by ensemble learning, and a hierarchical clustering strategy was established for to achieve fine extraction of building façade elements. Experiments were conducted across multiple building types, and the results showed that: 1) The method can use different virtual semantic feature map and clustering strategies to achieve accurate extraction of diverse building façade elements; 2) The method achieved joint learning tasks in both 2D and 3D space; and, 3) The proposed method achieved fine extraction of building elements with pixel accuracy (PA) over 70% in all experiments and mean intersection over union (mIoU) up to 95%, which were better than the image based method. In summary, this method offers a novel, more reliable method for segmenting and extracting building façade elements, which has important theoretical and practical significance.
This work is aimed at investigating the mechanical behaviors of a fault unwelded bimrock subjected to coupled freeze-thaw (FT) and fatigue loads. The influence of FT cycle on rock strength, ...deformation, damage evolution, and failure morphology was detailed investigated. The testing results show that the soil-rock interface is sensitive to freeze-thaw condition. Interface damage and cracking contribute a lot to sample strength, volumetric deformation, and damage evolution. In addition, a sudden increase of stress amplitude results in the increase of damage, afterwards, the coupling degree of bimrock improves with the increase of loading cycles. A coupling freeze-thaw and mechanical damage evolution model was proposed using the damage index defined by ultrasonic velocity and axial strain. It is found that the evolution trend of the model is strongly related to the previous freeze-thaw damage. Moreover, failure mode of bimrock changes from shear failure to bulging failure with increasing FT cycles. The failure process of bimrock is the rearrangement of the fine soil particles and rock blocks, the soil matrix and rock block bear axial cyclic loads by their interactions.
Celotno besedilo
Dostopno za:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, UILJ, UKNU, UL, UM, UPUK
This work presents the experimental results of the fracture behaviors for Tibet hornfels rock obtained from an open pit with different natural fracture ratios under multi-level cyclic shear loads. A ...newly developed rock dynamic shear testing apparatus was employed to perform the increasing-amplitude cyclic loads on the samples; the digital image correlation (DIC) technique was used to monitor the full-field displacement and the principal strain evolution for all tested samples. The results show the shear strength, deformation, full-field displacement, and the final failure pattern are influenced by the inherent natural fractures. The DIC results reveal the full-field displacement evolution and the characteristics of strain concentration. The stimulation of the natural fractures is impacted by the relations between shear direction, natural fracture orientation, and its cement characteristics. High strain concentration and large displacement occur near the natural fractures. For the sample with low natural fracture volume or the sample whose natural fracture orientation is parallel to the shear direction, the morphology of failure surface is the simplest. It is interesting to notice that for the sample whose fracture orientation is almost vertical to the shear direction, the multiple strain concentration zones were identified. Overall, this study confirms the pronounced effect of rock structure on the deformation and fracture evolution of the natural rock mass exposed to cyclic shear loads.
This paper aims to reveal the mechanical properties, energy evolution characteristics, and dynamic rupture process of preflawed granite under impact loading with different rock bridge angles and ...strain rates. A series of dynamic impact experiments were conducted along with the separate Hopkinson press bar (SHPB) testing system to analyze and study the overall rock fracture process. Under the impact load, the peak stress of granite increases with the increase of rock bridge angle and strain rate, but the increase gradually decreases. The peak strain also increases gradually with the increase of rock bridge angle, but there is an upper limit value; the total input strain energy increases with the increase of strain rate and rock bridge angle. It is shown that the higher the strain rate, the higher the unit dissipation energy, and the greater the degree of rock fragmentation. For rock under impact loads, the crack first initiates from the wing end of the prefabricated flaw, the preflaw closes gradually, and finally the crack propagates at the locking section leading to the coalescence of rock bridge. With the increase of strain rate, the fragmentation degree of the specimen increases asymptotically, and the average fragmentation size of the specimen decreases with the increase of strain rate. It is suggested that the stability of large rocked slopes is controlled by the locked section, and understanding the fracture evolution of the rock bridge is the key to slope instability prediction.
In the construction of large-scale water conservancy and hydropower transportation projects, the rock mass structural information is often used to evaluate and analyze various engineering geological ...problems such as high and steep slope stability, dam abutment stability, and natural rock landslide geological disasters. The complex shape and extremely irregular distribution of the structural planes make it challenging to identify and extract automatically. This study proposes a method for extracting structural planes from UAV images based on Geo-AINet ensemble learning. The UAV images of the slope are first used to generate a dense point cloud through a pipeline of SfM and PMVS; then, the multiple geological semantics, including color and texture from the image and local geological occurrence and surface roughness from the dense point cloud, are integrated with Geo-AINet for ensemble learning to obtain a set of semantic blocks; finally, the accurate extraction of structural planes is achieved through a multi-semantic hierarchical clustering strategy. Experimental results show that the structural planes extracted by the proposed method perform better integrity and edge adherence than that extracted by the AINet algorithm. In comparison with the results from the laser point cloud, the geological occurrence differences are less than three degrees, which proves the reliability of the results. This study widens the scope for surveying and mapping using remote sensing in engineering geological applications.
In order to reveal the instability law of open pit mine slope in high-cold and high-altitude area. Firstly, the slope structural plane is scanned by three-dimensional (3D) laser scanning technology, ...and the point cloud data is obtained to realize the intelligent identification of rock mass structural plane. The geometric parameters of rock mass are counted, and the physical structure model is established. Then, we carry out freeze-thaw cyclic tests on granite to obtain the corresponding mechanical parameters. Finally, according to the obtained mechanical parameters, we use RS2 finite element software to calculate the shear strength of structural plane and joint by generalized Hoek-Brown criterion and Barton-Bandis criterion, respectively, establish the geomechanical model, and use the finite element strength reduction method to calculate the safety factor of slope and judge the instability of slope. The results show that the physical and mechanical properties of granite deteriorate with the increase of the freeze-thaw cycle. Under the action of the freeze-thaw cycle, the pore water in the rock mass freezes and forms frost heaving force. The expansion of volume leads to the further development of joint fissures. The strength of rock slope decreases gradually with the increase of freeze-thaw cycle times, and the safety factor of slope decreases continuously. It shows that repeated freeze and thaw alternation makes the stability of mine slope worse and worse. The research results are helpful to prevent the occurrence of slope disasters in advance and are of great significance to effectively and safely manage the stability of slope, the treatment of open pit, and environmental treatment.
Celotno besedilo
Dostopno za:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, UILJ, UKNU, UL, UM, UPUK