Search

Aiming at the issue of complex 3D shape analysis and recognition, this paper presented a novel 3D graph convolu- tion classification method. It established a joint graph convolution learning mechanism of local geometry and global structure to provide both geometrical features and global context features, which effectively improved the robustness and stability of 3D data learning. Firstly, it constructed the local graph in spatial domain by farthest point sampling and K-nearest neighbor method, and designed a dynamic spectral graph convolution operator to extract local geometric features effectively. Meanwhile, it con- structed the global feature graph based on random sampling in the feature domain, and obtained the global structure context by spectral graph convolution. Furthermore, it established a weighted graph convolution network with an attention mechanism to achieve adaptive feature fusion. Finally, under the optimization of objective function, it improved the performance of feature learning effectively. Experimental results show that the proposed joint network learning mechanism, which combined local geo- metric features with global structure features, enhances the representation ability and discrimination of deep features, and ob- tains better recognition and classification performance compared with advanced methods. This method can be used for large- scale point clouds recognition, 3D shape reconstruction and data compression. It has important research significance and broad application prospects in robot, product digital analysis, intelligent navigation, virtual reality and other fields. [ABSTRACT FROM AUTHOR]

When working in complex scenarios, traditional simultaneous localization and mapping（SLAM） systems cannot distinguish whether the visible objects are moving. Moving objects in the images may lead to wrong feature association, resulting in the inaccuracy of positioning and the deviation of mapping. To improve the robustness and reliability of the SLAM system in dynamic scenarios, an enhanced visual SLAM system with dynamic object perception is proposed in this paper. Firstly, the object detector is used to initially detect the dynamic objects in each image, and then the multi-view geometry method is further used to extract the dynamic regions that cannot be determined by the object detection. The robustness of the system is improved by eliminating feature points belonging to dynamic objects. The proposed method is tested in public datasets TUM and KITTI. The results show that the localization accuracy of the proposed method in dynamic scenes has been significantly improved, especially in high dynamic sequences. Compared with the original algorithm, the accuracy is improved by more than 92%. Compared with other SLAM systems in dynamic scenarios, the proposed method not only maintains the accuracy advantage, but also improves the stability of running results and time efficiency. [ABSTRACT FROM AUTHOR]

Aiming at the problem of low accuracy and poor robustness of Visual Simultaneous Localization and Mapping (VSLAM） in dynamic environment, This paper proposed an adaptive Dynamic SLAM（AD-SLAM）. The scene dynamic classification front-end based on epipolar geometry constraint perceives the dynamic changes of the working environment in real time. The dynamic feature culling in low dynamic environment based on geometric constraints and moving probability and the dynamic target processing in high dynamic environment based on semantic segmentation eliminates moving objects. Experimental verification on the computer vision dataset TUM shows that the proposed method can improve the accuracy and robustness of localization in the mapping process of mobile robots while ensuring the real-time performance of the algorithm. [ABSTRACT FROM AUTHOR]

Existing Non-negative Matrix Factorization（NMF） studies mostly consider a single view to decompose data, ignoring the comprehensiveness of data information. Additionally, NMF limit their access to the intrinsic geometry of the data. This paper proposes a novel matrix factorization method, called Structure Regularized Multi-View Nonnegative Matrix Factorization（SRMNMF）. Specifically, this paper first perform discriminative learning of the global structure of the data through principal component analysis. Then, it used manifold learning to capture the local structure of the data. Finally, it learnt representations by exploiting the diversity and difference of multi-view data. The model improves the overall performance of the algorithm clustering and mines the structural information of the data more effectively. The objective function of SRMNMF can be easily optimized using an efficient alternate iterative algorithm. Comparing with existing representative methods on 6 datasets, the proposed SRMNMF achieves a maximum improvement of 4.4%, 6.1% and 4.05% in accuracy, NMI and Purity, respectively. [ABSTRACT FROM AUTHOR]

In onshore oilfields, limited by surface obstacles, conventional seismic data acquisition based on regular sampling theory is becoming more and more difficult to achieve. At the same time, in order to solve more and more complex geological problems, spatial sampling is more and more dense, so seismic exploration costs have risen sharply. In order to adapt to complex surface conditions and save exploration costs, this paper studies the design method of seismic survey geometry based on compressive sensing theory, and the regularization of random seismic data is completed by using a low-rank constraint algorithm in high, dimensional space. Finally, the result of numerical simulation shows that under the same sampling density, the method can obtain better imaging results than regular sampling, and explore a new approach for high, efficiency and high, density seismic exploration in the eastern old oil area. [ABSTRACT FROM AUTHOR]

The current highway abnormal event detection efficiency is low, the leakage rate is high, real-time poor, and so on, this paper proposed highway abnormal parking detection method based on space-time relationship. Firstly, it used the traffic flow frequency to quantify the split image and removed the smaller connectivity domain to extract the road surface information, then used the perspective geometry of the pinhole camera to normalize the same scale. It used YOLOv4 network to obtain the exact position of the vehicle target, and established the space-time information matrix. To detect abnormal region and the detection results can output by the update of the space-time matrix. The experimental results show that the accuracy of this method is 95% in long-distance scenes and 93% in crowded scenes. [ABSTRACT FROM AUTHOR]

Aiming at the problem that the visual SLAM algorithm in dynamic scenes is susceptible to motion feature points, resulting in low accuracy and poor robustness of pose estimation, this paper proposed a RGB-D visual SLAM algorithm based on dynamic region culling. Firstly, identify the feature points belonging to the moving objects with the help of semantic information, and detected whether the feature points remain stationary at this point used multi-view geometry with the help of the depth information of the camera. Use feature points extracted from static objects and static feature points derived from movable objects to fine-tune the camera pose estimation to achieve accurate and robust operation of the system in dynamic scenes. Finally, it used the dynamic indoor scenes in the TUM dataset for experimental verification. Experiments show that the proposed algorithm can effectively improve the accuracy of the camera′s pose estimation in the indoor dynamic environment, realize the map update in the dynamic environment, and improve the accuracy of the environment construction while enhancing the robustness of the system. [ABSTRACT FROM AUTHOR]

Aiming at the deficiency of physical geometry and traffic organization canalization description in traffic network data model at present, this paper presented a road network data model oriented to refined intersections. The model described logical topology and rule of markings of intersections from physical, logical and geometric levels. It also could support the calculation requirements of traffic analysis. At the same time, based on physical topology and logical topology, this paper presented a modeling method of intersection geometry topology by using linear reference technology based on elements. The results show that the proposed method can effectively realize the automatic generation of precise intersection models. [ABSTRACT FROM AUTHOR]

When the ground-based pseudolite positioning system is independent networking, distribution of base station is a multi-objective optimization problem to improve the system signal coverage and optimize the geometric layout of the base stations, thereby reduces the number of base stations and ensures positioning accuracy. This paper proposed the pseudolite base station distribution method based on the non-dominated sorting genetic algorithm ( NSGA- II ) with elite strategy. The method used the visual domain analysis technology to calculate the system signal coverage and evaluate the geometric layout of the base stations by a weighted geometric precision factor of multi-matrix multiplication. Finally, the NSGA- II algorithm searched the non-dominated optimal solution set of two optimization targets. The simulation of a mountainous area in Zhangjiajie shows only 28 base stations can ensure the system signal coverage reach 90%. At the same time, base station geometry is also good, the performance is better than the traditional genetic algorithm, it has certain guiding significance in practical engineering applications. [ABSTRACT FROM AUTHOR]

According to the design requirements related to high endurance factor, short takeoff and landing, deforming resistance of MALE UAV, the investigation of the design methodology on multi-constraints two-element airfoil is conducted based upon the earlier studies on the one-element airfoil. In this paper, the geometry of the two-element airfoil is generated with the method of controllable conic curves accompanied with the control points, and the parameters of control curves, slot and spin axis position are optimized using Genetic Algorithms. The results indicate that* compared with the ellipse cutting method, the present method using more control parameters is more flexible to the earlier airfoil with more reasonable pressure distributions. Moreover, the endurance factor, take-off lift and lift-to-drag ratio of the designed slot airfoil are improved substantially. In addition, flaps with wide angle deflection play the role of deceleration during landing. The designed slot airfoil satisfies multiple demands and the present method is suitable for the MALE UAV design. [ABSTRACT FROM AUTHOR]

Through the study on drawing behavior and habit of designers, the paper gave a sketch understanding method which combined stroke semantic feature and geometric feature. The paper's purpose was to build a hand-drawn interface system for concept design personnel in the creative development research. According to conceptual sketching process, the paper extracted strokes types, and analyzed the semantic features. By the concept definition of "bending threshold" and "pressure strength threshold", it used threshold detection to determine the strokes types, which could be used into rough stages. The results made different stages, different types of stroke function effectively classified and auxiliary recognized. The designers could choose the way between drafting interface and understanding interface by man-machine interactive function in the sketching process. The test result shows that this kind of method not only can get a hand-drawn effect, but also shall not prevent the designer's creative thinking process and can meet the personalized drawing habit. [ABSTRACT FROM AUTHOR]

To the question that classic phase-only correlation algorithm is not appropriate in the field of high resolution dense matching because of amount of calculation, this paper introduced a new dense matching algorithm. Based on dense matching strategy and relying on the phase-only correlation faction under epipolar geometry constraint, it transformed two dimension faction fitting into one dimension faction fitting and greatly shortened the running time of the algorithm with same matching accuracy to classic phase-only correlation algorithm. Experiments show that the algorithm achieves higher accuracy and spends less time to matching algorithm with a similar matching strategy based on gray, so the algorithm has very high practicability. [ABSTRACT FROM AUTHOR]

This paper describes a series of compression tests conducted on lipped channels with stiffened-web fabricated from mild steel plate of thickness 1.5 mm with nominal strength Grade Q345. A total of 18 channel specimens including two cross-section dimensions with three different column lengths were tested to failure with both ends of the column fixed, in order to determine the structural behavior and ultimate strength of cold-formed thin-walled steel columns experiencing post-buckling mode interaction between local and distortional buckling, due to the closeness of the critical buckling stresses. Test results consist of the column geometries, material properties, initial geometrical imperfections, non-linear equilibrium paths, deformed configuration, collapse modes and ultimate strength values. The results indicate that the column specimens undergoing interaction between local buckling and distortional displayed a significant post-buckling strength reserve. Moreover, the adverse interaction of local and distortional buckling reduced significantly the ultimate strength of the column specimen. Most of all, the post-local buckling interaction with distortional buckling (L + D) and the post-distortional buckling interaction with local buckling (D + L) have different effects on structural behavior, collapse mechanism, and failure load. Besides, the tested specimens were simulated by means of finite element program ABAQUS. Results from the analysis agree well with the experimental data. [ABSTRACT FROM AUTHOR]

In this paper, based on the fixed hypersonic inlet with rectangular section, a new concept of variable geometry hypersonic inlet that the tip can turn and move back off to increase the mass capture ratio is presented, and operating in a large Mach number range from 4 to 6.5. The flow in the hypersonic inlet was analyzed with numerical analyzer, and the inlet performance comparisons were made between the variable geometry hypersonic inlet and fixed inlet at different Mach number of inflow. The results indicated that the performance of the variable geometry inlet as much as fixed one under design Mach number, and the performance of the variable geometry inlet was better than fixed inlet under non design point. All of that indicate that the variable geometry hypersonic inlet was satisfying. [ABSTRACT FROM AUTHOR]

Traditional buffer operators can not realize fine tuning of effect intensity, which leads to problems that the effectiveness of buffer action may be too strong or too weak. Considering this situation, several kinds of new buffer operators with variable weights are constructed in this paper, and the inner link among these operators is proposed. At the same time, the corresponding relationship between the strengthening buffer with variable weights and the weakening buffer with variable weights is revealed. The relationship between the regulation degree of buffer operator and their variable weights is defined. The results show that, the regulation degree and variable weights change as the same (opposite) direction, and have nothing to do with 7-The arithmetic buffer operators with variable weights and the geometry buffer operators with variable weights are all special cases of these constructed buffer operators with variable weights. [ABSTRACT FROM AUTHOR]

The mesh morphing technology is applied to the retrofit design of a baseline car model in this paper. Specifically, according to the morphing scheme worked out, the geometric dimension of FE model for its body-in-white is changed to build the FE model for new car model efficiently, with its performance predicted by finite element analysis. The results indicate that with mesh morphing technology, the dependency of design on detailed CAE data and the time-consuming preparation activities for simulation can be avoided, with the lead time for car model development shortened and the development cost reduced, truly embodying the vision of 'CAE leads design'. [ABSTRACT FROM AUTHOR]

By considering the factors such as arch section types, initial geometric crookedness, residual stresses, riseto- span ratio and support conditions, this paper presented the out-of-plane stability design curves for circular arches under uniform compression. It can be not only applied in the practical design of arch under uniform compression directly, but also involved in the design of arch under the combination of uniform compression and bending. It indicates that load with constant direction which causes lower out-of-plane load-carrying capacity of arches is more adverse than one with direction pointing to a constant center. Compared with initial geometric crookedness and residual stresses, rise-to-span ratio and support conditions much more affect the out-of-plane load-carrying capacity. However, stability design curves for arches with different rise-to-span ratio and support conditions are unified by using the normalized arch slenderness. Finally, out-of-plane stability design method of arches with hot-rolled pipe section, welded box section and welded I section is proposed by employing stability design curves for columns of type b, c, d in Chinese code (GB 50017-2003) respectively. [ABSTRACT FROM AUTHOR]

To research the compressive behavior of circular section X-joints with concrete filled brace, six joint specimens with different geometric parameters were tested under axial compression in brace. The test procedure, joint failure modes, jack load versus global displacement curves, axial compression in brace versus deformation of chord tube and strain distribution curves were presented in the paper. The effects off Β,γ and τ on compressive capacity and elastic compressive stiffness of circular section X-joints with concrete filled braces were also studied. Results of these tests show that the joints all show good ductility. The ultimate deformation of joints is all larger than multiply diameter of chord by 0. 03. The ultimate deformation criterion can be adopted to evaluate the compressive capacity of the joints. As values off Β and τ increase, value of γ decreases, accordingly capacity and elastic compressive stiffness of joints increase. Spots of brace still keep elastic while spots between crown and saddle of chord yield first. The concrete of brace not significantly improves the capacity of joints, even reduce the capacity of joints in certain geometric parameters. Equation of code for design of steel structure can' t calculate the compressive capacity of circular section X-joints with concrete filled braces. [ABSTRACT FROM AUTHOR]

To research the compressive capacity of concrete filled chord circular section X-joints under axial compression, six joint specimens with different geometry parameters were tested under axial compression in brace. The test procedure, joint failure phenomenon, axial compression in brace-global displacement curves, axial compression in brace-deformation of chord tube and strain distribution curves were presented in the paper. The effects of Β, γ and τ on compressive capacity and stiffness of the concrete flled chord X-joints were also studied. Results of these tests show that as the values off Β and γ increase, the compressive stiffness of joints increase accordingly. The value of τ has little effect on stiffness. The crown spot of brace first yields and its strain strength will stay at the maximum till failure. Spot of chord still keeps elastic, concrete of chord obviously improves the capacity of joints. The calculated capacity values of joints based on the full yielding of brace section are unsafe. There is a good agreement between the experimental and the finite element analysis results in terms of failure modes, capacity and stress contour plot. [ABSTRACT FROM AUTHOR]

This paper gained an efficient blind algorithm by selecting characteristics, building and fitting statistics distribution, improving and optimizing classification algorithm, that was triangular mesh geometric features steganographic algorithm. It used classic steganography algorithm to build database and train classifier. This method not only achieves classification correctly (lower embedding rate, lower capacity),but also formes a theoretical framework of steganalysis based on 3D model. [ABSTRACT FROM AUTHOR]