对于复杂天气场景图像模糊、低对比度和颜色失真所导致的深度信息预测不准的问题,以往的研究均以标准场景的深度图作为先验信息来对该类场景进行深度估计。然而,这一方式存在先验信息精度较低等问题。对此,提出一个基于多尺度注意力机...对于复杂天气场景图像模糊、低对比度和颜色失真所导致的深度信息预测不准的问题,以往的研究均以标准场景的深度图作为先验信息来对该类场景进行深度估计。然而,这一方式存在先验信息精度较低等问题。对此,提出一个基于多尺度注意力机制的单目深度估计模型TalentDepth,以实现对复杂天气场景的预测。首先,在编码器中融合多尺度注意力机制,在减少计算成本的同时,保留每个通道的信息,提高特征提取的效率和能力。其次,针对图像深度不清晰的问题,基于几何一致性,提出深度区域细化(Depth Region Refinement,DSR)模块,过滤不准确的像素点,以提高深度信息的可靠性。最后,输入图像翻译模型所生成的复杂样本,并计算相应原始图像上的标准损失来指导模型的自监督训练。在NuScence,KITTI和KITTI-C这3个数据集上,相比于基线模型,所提模型对误差和精度均有优化。展开更多
自监督单目深度估计受到了国内外研究人员的广泛关注。现有基于深度学习的自监督单目深度估计方法主要采用编码器-解码器结构。然而,这些方法在编码过程中对输入图像进行下采样操作,导致部分图像信息,尤其是图像的边界信息丢失,进而影...自监督单目深度估计受到了国内外研究人员的广泛关注。现有基于深度学习的自监督单目深度估计方法主要采用编码器-解码器结构。然而,这些方法在编码过程中对输入图像进行下采样操作,导致部分图像信息,尤其是图像的边界信息丢失,进而影响深度图的精度。针对上述问题,提出一种基于拉普拉斯金字塔的自监督单目深度估计方法(Self-supervised Monocular Depth Estimation Based on the Laplace Pyramid,LpDepth)。此方法的核心思想是:首先,使用拉普拉斯残差图丰富编码特征,以弥补在下采样过程中丢失的特征信息;其次,在下采样过程中使用最大池化层突显和放大特征信息,使编码器在特征提取过程中更容易地提取到训练模型所需要的特征信息;最后,使用残差模块解决过拟合问题,提高解码器对特征的利用效率。在KITTI和Make3D等数据集上对所提方法进行了测试,同时将其与现有经典方法进行了比较。实验结果证明了所提方法的有效性。展开更多
Coal seam water injection in tunnels is an effective technical measure for preventing coal mine rock bursts.This study used the improved split Hopkinson pressure bar(SHPB)to apply three equal static stresses to water-...Coal seam water injection in tunnels is an effective technical measure for preventing coal mine rock bursts.This study used the improved split Hopkinson pressure bar(SHPB)to apply three equal static stresses to water-saturated coal to simulate the initial stress environment of coal at different depths.Then,dynamic mechanical experiments were conducted on the saturated coal at different depths to investigate the effects of water saturation and depth on the coal samples’dynamic mechanical properties.Under uniaxial compression and without lateral compression,the strength of coal samples decreased to varying degrees in the saturated state;under different depth conditions,the dynamic strength of coal in the saturated state decreased compared with that in the natural state.However,compared with that at 0 m,the reduction in the strength of coal under the saturated condition at 200,400,600,and 800 m was significantly reduced.The findings of this study provide a basic theoretical foundation for the prevention and control of dynamic coal mine disasters.展开更多
During upward horizontal stratified backfill mining,stable backfill is essential for cap and sill pillar recovery.Currently,the primary method for calculating the required strength of backfill is the generalized three...During upward horizontal stratified backfill mining,stable backfill is essential for cap and sill pillar recovery.Currently,the primary method for calculating the required strength of backfill is the generalized three-dimensional(3 D)vertical stress model,which ignores the effect of mine depth,failing to obtain the vertical stress at different positions along stope length.Therefore,this paper develops and validates an improved 3 D model solution through numerical simulation in Rhino-FLAC^(3D),and examines the stress state and stability of backfill under different conditions.The results show that the improved model can accurately calculate the vertical stress at different mine depths and positions along stope length.The error rates between the results of the improved model and numerical simulation are below 4%,indicating high reliability and applicability.The maximum vertical stress(σ_(zz,max))in backfill is positively correlated with the degree of rock-backfill closure,which is enhanced by mine depth and elastic modulus of backfill,while weakened by stope width and inclination,backfill friction angle,and elastic modulus of rock mass.Theσ_(zz,max)reaches its peak when the stope length is 150 m,whileσ_(zz,max)is insensitive to changes in rock-backfill interface parameters.In all cases,the backfill stability can be improved by reducingσ_(zz,max).The results provide theoretical guidance for the backfill strength design and the safe and efficient recovery of ore pillars in deep mining.展开更多
The non-scaling effect on the penetration depth of rigid projectiles is an important issue that must be considered when extending the results of scaled experiments to prototype scenes.In this study,the evolution of th...The non-scaling effect on the penetration depth of rigid projectiles is an important issue that must be considered when extending the results of scaled experiments to prototype scenes.In this study,the evolution of the stress and strain of the target under penetration was analyzed.Expressions for the penetration resistance and penetration depth were obtained based on the conservation equation and continuity condition of the target.The penetration coefficients that characterize the nose shape,target resistance,and non-scaling effect were defined.Simplified calculation methods for the coefficients within the range of rigid projectile penetration were developed.Two methods for estimating the target parameters are proposed.The results show that the non-scaling effect is related to the failure process of the target and depends on the ratio of cavity radius to comminuted region radius.The nose shape coefficient can be approximated as a linear function of the length-to-diameter ratio of the nose.The noseshape coefficient of a flat-nosed projectile is 0.57.The caliber coefficient is related to the projectile diameter and reflects the non-scaling effect,which increases with the projectile diameter.A practical formula for calculating the penetration depth of rigid projectiles considering the non-scaling effect is also proposed.This formula is in good agreement with penetration experiments on rock and concrete.展开更多
The model test result of earth force in the side of anti-slide pile of anchor bars was introduced.There are three groups of the tests.The loads were on the back side of the slope in two groups.The other one was loaded...The model test result of earth force in the side of anti-slide pile of anchor bars was introduced.There are three groups of the tests.The loads were on the back side of the slope in two groups.The other one was loaded just behind the pile by the jack.In order to get the force of the soil,some earth-pressure boxes were used to get the earth pressure on the side of the piles.The part of the max pressure and the earth pressure was mainly focused under the slip line展开更多
In this paper, a new one-dimensional phenomenological model is developed for the assessment of the ballistic performance of Adobe. Adobe is a masonry largely spread in areas of the world involved in military operation...In this paper, a new one-dimensional phenomenological model is developed for the assessment of the ballistic performance of Adobe. Adobe is a masonry largely spread in areas of the world involved in military operations. Addressing fundamental ballistic parameters such as residual velocity or penetration depth for this building technology is necessary. The model follows the hypotheses for the ballistic response of concrete targets to high velocity impacts, provided with a dominant contribution of shear friction typical of soils. The hypotheses at the basis of the model are consistent with all experimental evidence collected by authors on Adobe. Adobe brick and mortar belong to the material class of concrete,whereas the overall mechanical parameters are determined by the internal soil mixture, including the percentage of fibre reinforcement. Despite its relative simplicity, the model is capable of well predicting ballistic test results currently available in literature for Adobe, including the data of an experimental campaign recently performed by the authors on real Adobe walls in the field.展开更多
This paper presents an integrated investigation of the ground stability of a deep gateroad with a 1 km burial depth based on a field test, case studies, and numerical modelling. In situ stress measurements and mechani...This paper presents an integrated investigation of the ground stability of a deep gateroad with a 1 km burial depth based on a field test, case studies, and numerical modelling. In situ stress measurements and mechanical properties tests were first conducted in the test site. Then, the deformation behavior, stress and yield zone distributions, as well as the bolts load of the gateroad, were simulated using FLAC3D software. The model results demonstrated that the soft rock properties and high in situ stress were the main factors for the deep gateroad instability, and the shear slip failure induced by the high stress was the primary failure model for the deep rock mass. In addition, the unsuitable support patterns, especially the relatively short bolts/cables with low pre-tensions, the lack of high-strengthen secondary supports and the unsupported floor strata, also contributed to the gateroad instability. Subsequently, a new combined supporting strategy, incorporating longer bolts/cables, yielding ring supports, and grouting measures, was proposed for the deep gateroad, and its validity was verified via field monitoring. All these could be a reference for understanding the failure mechanism of the gateroad with 1 km burial depth.展开更多
Aluminium matrix surface composites are gaining alluring role especially in aerospace, defence, and marine industries. Friction stir processing(FSP) is a promising novel solid state technique for surface composites fa...Aluminium matrix surface composites are gaining alluring role especially in aerospace, defence, and marine industries. Friction stir processing(FSP) is a promising novel solid state technique for surface composites fabrication. In this study, AA6061/SiC surface composites were fabricated and the effect of tool plunge depth on pattern of reinforcement particles dispersion in metal matrix was investigated. Six varying tool plunge depths were chosen at constant levels of shoulder diameter and tool tilt angle to observe the exclusive effect of plunge variation. Process parameters chosen for the experimentation are speed of rotation, travel speed and tool tilt angle which were taken as 1400 rpm, 40 mm/min, and 2.5 °respectively. Macro and the microstructural study were performed using stereo zoom and optical microscope respectively. Results reflected that lower plunge depth levels lead to insufficient heat generation and cavity formation towards the stir zone center. On the other hand, higher levels of plunge depth result in ejection of reinforcement particles and even sticking of material to tool shoulder. Thus, an optimal plunge depth is needed in developing defect free surface composites.展开更多
Real-time hand gesture recognition technology significantly improves the user's experience for virtual reality/augmented reality(VR/AR) applications, which relies on the identification of the orientation of the ha...Real-time hand gesture recognition technology significantly improves the user's experience for virtual reality/augmented reality(VR/AR) applications, which relies on the identification of the orientation of the hand in captured images or videos. A new three-stage pipeline approach for fast and accurate hand segmentation for the hand from a single depth image is proposed. Firstly, a depth frame is segmented into several regions by histogrambased threshold selection algorithm and by tracing the exterior boundaries of objects after thresholding. Secondly, each segmentation proposal is evaluated by a three-layers shallow convolutional neural network(CNN) to determine whether or not the boundary is associated with the hand. Finally, all hand components are merged as the hand segmentation result. Compared with algorithms based on random decision forest(RDF), the experimental results demonstrate that the approach achieves better performance with high-accuracy(88.34% mean intersection over union, mIoU) and a shorter processing time(≤8 ms).展开更多
文摘对于复杂天气场景图像模糊、低对比度和颜色失真所导致的深度信息预测不准的问题,以往的研究均以标准场景的深度图作为先验信息来对该类场景进行深度估计。然而,这一方式存在先验信息精度较低等问题。对此,提出一个基于多尺度注意力机制的单目深度估计模型TalentDepth,以实现对复杂天气场景的预测。首先,在编码器中融合多尺度注意力机制,在减少计算成本的同时,保留每个通道的信息,提高特征提取的效率和能力。其次,针对图像深度不清晰的问题,基于几何一致性,提出深度区域细化(Depth Region Refinement,DSR)模块,过滤不准确的像素点,以提高深度信息的可靠性。最后,输入图像翻译模型所生成的复杂样本,并计算相应原始图像上的标准损失来指导模型的自监督训练。在NuScence,KITTI和KITTI-C这3个数据集上,相比于基线模型,所提模型对误差和精度均有优化。
文摘自监督单目深度估计受到了国内外研究人员的广泛关注。现有基于深度学习的自监督单目深度估计方法主要采用编码器-解码器结构。然而,这些方法在编码过程中对输入图像进行下采样操作,导致部分图像信息,尤其是图像的边界信息丢失,进而影响深度图的精度。针对上述问题,提出一种基于拉普拉斯金字塔的自监督单目深度估计方法(Self-supervised Monocular Depth Estimation Based on the Laplace Pyramid,LpDepth)。此方法的核心思想是:首先,使用拉普拉斯残差图丰富编码特征,以弥补在下采样过程中丢失的特征信息;其次,在下采样过程中使用最大池化层突显和放大特征信息,使编码器在特征提取过程中更容易地提取到训练模型所需要的特征信息;最后,使用残差模块解决过拟合问题,提高解码器对特征的利用效率。在KITTI和Make3D等数据集上对所提方法进行了测试,同时将其与现有经典方法进行了比较。实验结果证明了所提方法的有效性。
基金Projects(52225403,52074112)supported by the National Natural Science Foundation of ChinaProject(2022CFD009)supported by the Hubei Natural Science Foundation Innovation and Development Joint Fund Key Project,China+2 种基金Project(SDGZK2423)supported by the State Key Laboratory of Intelligent Construction and Healthy Operation and Maintenance of Deep Underground Engineering,ChinaProject(HJZKYBKT2024111)supported by the Xiangyang Federation of Social Sciences“Hanjiang Think Tank”Project,ChinaProject supported by the Hubei Superior and Distinctive Discipline Group of“New Energy Vehicle and Smart Transportation”,China。
文摘Coal seam water injection in tunnels is an effective technical measure for preventing coal mine rock bursts.This study used the improved split Hopkinson pressure bar(SHPB)to apply three equal static stresses to water-saturated coal to simulate the initial stress environment of coal at different depths.Then,dynamic mechanical experiments were conducted on the saturated coal at different depths to investigate the effects of water saturation and depth on the coal samples’dynamic mechanical properties.Under uniaxial compression and without lateral compression,the strength of coal samples decreased to varying degrees in the saturated state;under different depth conditions,the dynamic strength of coal in the saturated state decreased compared with that in the natural state.However,compared with that at 0 m,the reduction in the strength of coal under the saturated condition at 200,400,600,and 800 m was significantly reduced.The findings of this study provide a basic theoretical foundation for the prevention and control of dynamic coal mine disasters.
基金Project(2024ZD1003704)supported by the Deep Earth Probe and Mineral Resources Exploration-National Science and Technology Major Project,ChinaProjects(51834001,52130404)supported by the National Natural Science Foundation of China。
文摘During upward horizontal stratified backfill mining,stable backfill is essential for cap and sill pillar recovery.Currently,the primary method for calculating the required strength of backfill is the generalized three-dimensional(3 D)vertical stress model,which ignores the effect of mine depth,failing to obtain the vertical stress at different positions along stope length.Therefore,this paper develops and validates an improved 3 D model solution through numerical simulation in Rhino-FLAC^(3D),and examines the stress state and stability of backfill under different conditions.The results show that the improved model can accurately calculate the vertical stress at different mine depths and positions along stope length.The error rates between the results of the improved model and numerical simulation are below 4%,indicating high reliability and applicability.The maximum vertical stress(σ_(zz,max))in backfill is positively correlated with the degree of rock-backfill closure,which is enhanced by mine depth and elastic modulus of backfill,while weakened by stope width and inclination,backfill friction angle,and elastic modulus of rock mass.Theσ_(zz,max)reaches its peak when the stope length is 150 m,whileσ_(zz,max)is insensitive to changes in rock-backfill interface parameters.In all cases,the backfill stability can be improved by reducingσ_(zz,max).The results provide theoretical guidance for the backfill strength design and the safe and efficient recovery of ore pillars in deep mining.
基金the National Natural Science Foundation of China(Grant Nos.52422808,52378401)to provide funds for this research。
文摘The non-scaling effect on the penetration depth of rigid projectiles is an important issue that must be considered when extending the results of scaled experiments to prototype scenes.In this study,the evolution of the stress and strain of the target under penetration was analyzed.Expressions for the penetration resistance and penetration depth were obtained based on the conservation equation and continuity condition of the target.The penetration coefficients that characterize the nose shape,target resistance,and non-scaling effect were defined.Simplified calculation methods for the coefficients within the range of rigid projectile penetration were developed.Two methods for estimating the target parameters are proposed.The results show that the non-scaling effect is related to the failure process of the target and depends on the ratio of cavity radius to comminuted region radius.The nose shape coefficient can be approximated as a linear function of the length-to-diameter ratio of the nose.The noseshape coefficient of a flat-nosed projectile is 0.57.The caliber coefficient is related to the projectile diameter and reflects the non-scaling effect,which increases with the projectile diameter.A practical formula for calculating the penetration depth of rigid projectiles considering the non-scaling effect is also proposed.This formula is in good agreement with penetration experiments on rock and concrete.
文摘The model test result of earth force in the side of anti-slide pile of anchor bars was introduced.There are three groups of the tests.The loads were on the back side of the slope in two groups.The other one was loaded just behind the pile by the jack.In order to get the force of the soil,some earth-pressure boxes were used to get the earth pressure on the side of the piles.The part of the max pressure and the earth pressure was mainly focused under the slip line
文摘In this paper, a new one-dimensional phenomenological model is developed for the assessment of the ballistic performance of Adobe. Adobe is a masonry largely spread in areas of the world involved in military operations. Addressing fundamental ballistic parameters such as residual velocity or penetration depth for this building technology is necessary. The model follows the hypotheses for the ballistic response of concrete targets to high velocity impacts, provided with a dominant contribution of shear friction typical of soils. The hypotheses at the basis of the model are consistent with all experimental evidence collected by authors on Adobe. Adobe brick and mortar belong to the material class of concrete,whereas the overall mechanical parameters are determined by the internal soil mixture, including the percentage of fibre reinforcement. Despite its relative simplicity, the model is capable of well predicting ballistic test results currently available in literature for Adobe, including the data of an experimental campaign recently performed by the authors on real Adobe walls in the field.
基金Project（2017RCJJ011） supported by the Scientific Research Foundation of Shaaadong University of Science and Technology for Recruited Talents, China Projects（01CK03203, 02CK02302） supported by the Shaaadong Provincial First-Class Discipline Fundamental, China Proj ect（ZR2018QEE001） supported by the Natural Science Foundation of Shandong Province, China
文摘This paper presents an integrated investigation of the ground stability of a deep gateroad with a 1 km burial depth based on a field test, case studies, and numerical modelling. In situ stress measurements and mechanical properties tests were first conducted in the test site. Then, the deformation behavior, stress and yield zone distributions, as well as the bolts load of the gateroad, were simulated using FLAC3D software. The model results demonstrated that the soft rock properties and high in situ stress were the main factors for the deep gateroad instability, and the shear slip failure induced by the high stress was the primary failure model for the deep rock mass. In addition, the unsuitable support patterns, especially the relatively short bolts/cables with low pre-tensions, the lack of high-strengthen secondary supports and the unsupported floor strata, also contributed to the gateroad instability. Subsequently, a new combined supporting strategy, incorporating longer bolts/cables, yielding ring supports, and grouting measures, was proposed for the deep gateroad, and its validity was verified via field monitoring. All these could be a reference for understanding the failure mechanism of the gateroad with 1 km burial depth.
基金the University Grants Commission (UGC) for its financial assistance (vide sanction order No. F.3-40/2012 (SAP-II) under its SAP (DRS-I) sanctioned to the Department of Mechanical Engineering for the project entitled Friction Stir Welding, Ultrasonic Machining
文摘Aluminium matrix surface composites are gaining alluring role especially in aerospace, defence, and marine industries. Friction stir processing(FSP) is a promising novel solid state technique for surface composites fabrication. In this study, AA6061/SiC surface composites were fabricated and the effect of tool plunge depth on pattern of reinforcement particles dispersion in metal matrix was investigated. Six varying tool plunge depths were chosen at constant levels of shoulder diameter and tool tilt angle to observe the exclusive effect of plunge variation. Process parameters chosen for the experimentation are speed of rotation, travel speed and tool tilt angle which were taken as 1400 rpm, 40 mm/min, and 2.5 °respectively. Macro and the microstructural study were performed using stereo zoom and optical microscope respectively. Results reflected that lower plunge depth levels lead to insufficient heat generation and cavity formation towards the stir zone center. On the other hand, higher levels of plunge depth result in ejection of reinforcement particles and even sticking of material to tool shoulder. Thus, an optimal plunge depth is needed in developing defect free surface composites.
文摘Real-time hand gesture recognition technology significantly improves the user's experience for virtual reality/augmented reality(VR/AR) applications, which relies on the identification of the orientation of the hand in captured images or videos. A new three-stage pipeline approach for fast and accurate hand segmentation for the hand from a single depth image is proposed. Firstly, a depth frame is segmented into several regions by histogrambased threshold selection algorithm and by tracing the exterior boundaries of objects after thresholding. Secondly, each segmentation proposal is evaluated by a three-layers shallow convolutional neural network(CNN) to determine whether or not the boundary is associated with the hand. Finally, all hand components are merged as the hand segmentation result. Compared with algorithms based on random decision forest(RDF), the experimental results demonstrate that the approach achieves better performance with high-accuracy(88.34% mean intersection over union, mIoU) and a shorter processing time(≤8 ms).
