机载激光雷达测深是近年来蓬勃发展的主动式水深测量方法,能够快速精准地获取近岸水深和水下地形,特别是对于浅海、岛礁等船只无法达到的区域具有显著优势。而机载激光雷达系统的测深能力主要受到水体浑浊度的影响。激光测深实验中对实...机载激光雷达测深是近年来蓬勃发展的主动式水深测量方法,能够快速精准地获取近岸水深和水下地形,特别是对于浅海、岛礁等船只无法达到的区域具有显著优势。而机载激光雷达系统的测深能力主要受到水体浑浊度的影响。激光测深实验中对实验区域的水体浑浊度研究将有助于实验方案的设计。以中国海南岛沿岸海域为例,研究了该海域水体浑浊度和机载激光雷达测深系统CZMIL(coastal zone mapping and imaging LiDAR)测深能力之间的关系,建立了运用水体漫衰减系数估算机载激光雷达测深系统测深能力的算法。首先分析并确定了实验区域的漫衰减系数K_d(490)反演算法;其次,运用该区域实测光学数据建立了漫衰减系数K_d(490)和K_d(532)之间的数值关系;接着总结了K_d(532)和CZMIL系统最大测深值之间的关系;最后运用MODIS数据合成了海南岛沿岸海域在CZMIL系统海道测量模式下的测深能力空间分布图,重点分析了海口和陵水附近海域的最大可测水深分布情况。为海南岛沿岸海域开展激光测深实验提供了参考和依据。展开更多
This works presents the first fully validated and predictive capability to model the V_0-V_(100) probabilistic penetration response of a woven fabric using a yarn-level fabric finite element model. The V_0-V_(100) cur...This works presents the first fully validated and predictive capability to model the V_0-V_(100) probabilistic penetration response of a woven fabric using a yarn-level fabric finite element model. The V_0-V_(100) curve describes the probability of complete fabric penetration as a function of projectile impact velocity. The exemplar case considered in this paper comprises of a single-layer, fully-clamped, plain-weave Kevlar fabric impacted at the center by a 17-gr, 0.22 cal FSP or fragment-simulating projectile. Each warp and fill yarn in the fabric is individually modeled using 3 D finite elements and the virtual fabric microstructure is validated in detail against the experimental fabric microstructure. Material and testing sources of statistical variability including yarn strength and modulus, inter-yarn friction, precise projectile impact location, and projectile rotation are mapped into the finite element model. A series of impact simulations at varying projectile impact velocities is executed using LS-DYNA on the fabric models, with each model comprising unique mappings. The impact velocities together with the outcomes(penetration, nonpenetration) are used to generate the numerical V_0-V_(100) curve which is then validated against the experimental V_0-V_(100) curve. The numerical Vi-Vrdata(impact, residual velocities) is also validated against the experimental Vi-Vrdata. For completeness, this paper also reports the experimental characterization data and its statistical analysis used for model input, viz. the Kevlar yarn tensile strengths, moduli, and inter-yarn friction, and the experimental ballistic test data used for model validation.展开更多
Virtual testing of fabric armor provides an efficient and inexpensive means of systematically studying the influence of various architectural and material parameters on the ballistic impact behavior of woven fabrics, ...Virtual testing of fabric armor provides an efficient and inexpensive means of systematically studying the influence of various architectural and material parameters on the ballistic impact behavior of woven fabrics, before actual laboratory prototypes are woven and destructively tested. In this finite element study, the combined effects of individual ply orientations and material properties on the impact performance of multi-layered, non-stitched woven aramid fabrics are studied using 2-and 4-sided clamping configurations. Individual ply orientations of 0°, ±15°, ±30°, and ±45° are considered along with three levels of inter-yarn friction coefficient. Functionally graded fabric targets are also considered wherein the yarn stiffness progressively increases or decreases through the target thickness while keeping the yarn strain energy density constant and with all other material and architectural parameters unchanged for consistency. For each target configuration, one non-penetrating and one penetrating impact velocity is chosen. The impact performance is evaluated by the time taken to arrest the projectile and the backface deformation for the non-penetrating impacts, and by the residual velocity for the penetrating impact tests. All deterministic impact simulations are performed using LS-DYNA. 2-sided clamped targets and lower inter-yarn frictional levels generally resulted in better impact performance.The functionally graded targets generally showed either similar or inferior impact performance than the baseline fabric target configurations for the non-penetrating shots. Some performance improvements were observed for the penetrating shots when the yarn stiffness was progressively decreased through the layers in a direction away from the strike face, with additional performance enhancements achieved by simultaneously reducing the inter-yarn friction.展开更多
伴随着当今更低成本和更高性能的工业相机的趋势,对CMOS图像传感器也提出了更高的要求,需要通过设计系统级芯片(SoC)来实现这一目标.为实现该目标,需通过3D芯片堆栈和背照(Back Side Illuminated,BSI)技术,把多个图像处理任务集成到单...伴随着当今更低成本和更高性能的工业相机的趋势,对CMOS图像传感器也提出了更高的要求,需要通过设计系统级芯片(SoC)来实现这一目标.为实现该目标,需通过3D芯片堆栈和背照(Back Side Illuminated,BSI)技术,把多个图像处理任务集成到单一器件中.在未来将会出现具有精密的机器学习和专有的智能计算芯片结合图像撷取功能的解决方案,创造出紧凑的高速运算视觉系统.可是在实现崭新的大型技术集成之前,必需扫除两个主要发展障碍——芯片的热量管理和功耗.现在,先进的前照(Front Side Illuminated,FSI)CMOS传感器集成了模拟和数字功能,实现了成本和效能兼具的解决方案.能够达成这些目标,关键在于把有利于系统效能及嵌入到图像传感器SoC等各种因素的巧妙分隔.本文将介绍一款面向条形码读码器和其它嵌入式视觉应用的CMOS图像传感器系列产品、它们的应用实例,以及一些未来发展趋势.展开更多
文摘机载激光雷达测深是近年来蓬勃发展的主动式水深测量方法,能够快速精准地获取近岸水深和水下地形,特别是对于浅海、岛礁等船只无法达到的区域具有显著优势。而机载激光雷达系统的测深能力主要受到水体浑浊度的影响。激光测深实验中对实验区域的水体浑浊度研究将有助于实验方案的设计。以中国海南岛沿岸海域为例,研究了该海域水体浑浊度和机载激光雷达测深系统CZMIL(coastal zone mapping and imaging LiDAR)测深能力之间的关系,建立了运用水体漫衰减系数估算机载激光雷达测深系统测深能力的算法。首先分析并确定了实验区域的漫衰减系数K_d(490)反演算法;其次,运用该区域实测光学数据建立了漫衰减系数K_d(490)和K_d(532)之间的数值关系;接着总结了K_d(532)和CZMIL系统最大测深值之间的关系;最后运用MODIS数据合成了海南岛沿岸海域在CZMIL系统海道测量模式下的测深能力空间分布图,重点分析了海口和陵水附近海域的最大可测水深分布情况。为海南岛沿岸海域开展激光测深实验提供了参考和依据。
基金supported by Teledyne Scientific&Imaging(TS&I),Internal Research and Development(IR&D)and approved for public release under TSI-PP-17-08
文摘This works presents the first fully validated and predictive capability to model the V_0-V_(100) probabilistic penetration response of a woven fabric using a yarn-level fabric finite element model. The V_0-V_(100) curve describes the probability of complete fabric penetration as a function of projectile impact velocity. The exemplar case considered in this paper comprises of a single-layer, fully-clamped, plain-weave Kevlar fabric impacted at the center by a 17-gr, 0.22 cal FSP or fragment-simulating projectile. Each warp and fill yarn in the fabric is individually modeled using 3 D finite elements and the virtual fabric microstructure is validated in detail against the experimental fabric microstructure. Material and testing sources of statistical variability including yarn strength and modulus, inter-yarn friction, precise projectile impact location, and projectile rotation are mapped into the finite element model. A series of impact simulations at varying projectile impact velocities is executed using LS-DYNA on the fabric models, with each model comprising unique mappings. The impact velocities together with the outcomes(penetration, nonpenetration) are used to generate the numerical V_0-V_(100) curve which is then validated against the experimental V_0-V_(100) curve. The numerical Vi-Vrdata(impact, residual velocities) is also validated against the experimental Vi-Vrdata. For completeness, this paper also reports the experimental characterization data and its statistical analysis used for model input, viz. the Kevlar yarn tensile strengths, moduli, and inter-yarn friction, and the experimental ballistic test data used for model validation.
基金support from the M.C.Gill Composites Center at the University of Southern California(USC)supported by the USC Center for High-Performance Computing(hpcc.usc.edu)
文摘Virtual testing of fabric armor provides an efficient and inexpensive means of systematically studying the influence of various architectural and material parameters on the ballistic impact behavior of woven fabrics, before actual laboratory prototypes are woven and destructively tested. In this finite element study, the combined effects of individual ply orientations and material properties on the impact performance of multi-layered, non-stitched woven aramid fabrics are studied using 2-and 4-sided clamping configurations. Individual ply orientations of 0°, ±15°, ±30°, and ±45° are considered along with three levels of inter-yarn friction coefficient. Functionally graded fabric targets are also considered wherein the yarn stiffness progressively increases or decreases through the target thickness while keeping the yarn strain energy density constant and with all other material and architectural parameters unchanged for consistency. For each target configuration, one non-penetrating and one penetrating impact velocity is chosen. The impact performance is evaluated by the time taken to arrest the projectile and the backface deformation for the non-penetrating impacts, and by the residual velocity for the penetrating impact tests. All deterministic impact simulations are performed using LS-DYNA. 2-sided clamped targets and lower inter-yarn frictional levels generally resulted in better impact performance.The functionally graded targets generally showed either similar or inferior impact performance than the baseline fabric target configurations for the non-penetrating shots. Some performance improvements were observed for the penetrating shots when the yarn stiffness was progressively decreased through the layers in a direction away from the strike face, with additional performance enhancements achieved by simultaneously reducing the inter-yarn friction.
文摘伴随着当今更低成本和更高性能的工业相机的趋势,对CMOS图像传感器也提出了更高的要求,需要通过设计系统级芯片(SoC)来实现这一目标.为实现该目标,需通过3D芯片堆栈和背照(Back Side Illuminated,BSI)技术,把多个图像处理任务集成到单一器件中.在未来将会出现具有精密的机器学习和专有的智能计算芯片结合图像撷取功能的解决方案,创造出紧凑的高速运算视觉系统.可是在实现崭新的大型技术集成之前,必需扫除两个主要发展障碍——芯片的热量管理和功耗.现在,先进的前照(Front Side Illuminated,FSI)CMOS传感器集成了模拟和数字功能,实现了成本和效能兼具的解决方案.能够达成这些目标,关键在于把有利于系统效能及嵌入到图像传感器SoC等各种因素的巧妙分隔.本文将介绍一款面向条形码读码器和其它嵌入式视觉应用的CMOS图像传感器系列产品、它们的应用实例,以及一些未来发展趋势.
