The imaging plane of inverse synthetic aperture radar （ISAR） is the projection plane of the target. When taking an image using the range-Doppler theory, the imaging plane may have a spatial-variant property, which c...The imaging plane of inverse synthetic aperture radar （ISAR） is the projection plane of the target. When taking an image using the range-Doppler theory, the imaging plane may have a spatial-variant property, which causes the change of scatter＇s projection position and results in migration through resolution cells, In this study, we focus on the spatial-variant property of the imaging plane of a three-axis-stabilized space target. The innovative contributions are as follows. 1） The target motion model in orbit is provided based on a two-body model. 2） The instantaneous imaging plane is determined by the method of vector analysis. 3） Three Euler angles are introduced to describe the spatial-variant property of the imaging plane, and the image quality is analyzed. The simulation results confirm the analysis of the spatial-variant property. The research in this study is significant for the selection of the imaging segment, and provides the evidence for the following data processing and compensation algorithm.展开更多
We investigate a novel spatial geometric phase of hybrid-polarized vector fields consisting of linear, elliptical and circular polarizations by Young＇s two-slit interferometer instead of the widely used Mach-Zehnder ...We investigate a novel spatial geometric phase of hybrid-polarized vector fields consisting of linear, elliptical and circular polarizations by Young＇s two-slit interferometer instead of the widely used Mach-Zehnder interferometer. This spatial geometric phase can be manipulated by engineering the spatial configuration of hybrid polarizations, and is directly related to the topological charge, the local states of polarization and the rotational symmetry of hybrid-polarized vector optical fields. The unique feature of geometric phase has implications in quantum information science as well as other physical systems such as electron vortex beams.展开更多
The development of minimally invasive surgery has transformed the management of gastrointestinal cancer.Notably,three-dimensional visualization systems have increased surgical precision.This editorial discusses a rece...The development of minimally invasive surgery has transformed the management of gastrointestinal cancer.Notably,three-dimensional visualization systems have increased surgical precision.This editorial discusses a recent study by Shen and Zhang,which compared the clinical applications of naked-eye threedimensional laparoscopic systems vs traditional optical systems in radical surgery for gastric and colorectal cancer.Both systems appeared to yield comparable surgical and oncological outcomes in terms of safety parameters,operating times,and quality of lymph node dissection.However,the spectacle-free system’s technical and logistical limitations hindered its effects on the surgical team’s overall competency.This editorial examines the authors’findings within the broader context of the evolution of oncologic laparoscopy,discusses the relevance of the results in light of the current literature,and proposes future research directions focused on multicenter validation,comprehensive ergonomic analysis,and technological advancements aimed at enhancing intraoperative collaboration.As technology continues to evolve,clinical implementation of new methods must be supported by robust scientific evidence and standardized criteria,to ensure tangible improvements in efficiency,safety,and oncologic outcomes.展开更多
农业机械轨迹作业行为模式识别是一项多变量时间序列分类任务,旨在利用轨迹数据的时空特征识别农机的行为模式。针对已有方法未能从频率角度挖掘农机轨迹的全局特性以及识别精度不足的问题,提出了一种面向农机轨迹行为模式识别的频域注...农业机械轨迹作业行为模式识别是一项多变量时间序列分类任务,旨在利用轨迹数据的时空特征识别农机的行为模式。针对已有方法未能从频率角度挖掘农机轨迹的全局特性以及识别精度不足的问题,提出了一种面向农机轨迹行为模式识别的频域注意力和U型残差网络FARNet。该网络包含两个不同网络分支,用于全面挖掘农机轨迹的依赖信息。其中一个分支搭载了基于频域注意力的Transformer(transformer based on frequency attention,FAT)来挖掘农机轨迹在频域空间的全局时序依赖;另一分支部署了基于正交约束的U型残差网络(U-shaped residual network based on orthogonal constraints,URNet),其以ResUnet作为骨干网络提取轨迹特征图在不同感受野的深层语义信息,探索轨迹特征间的局部空间依赖。最后设计了一种特征对齐学习模块(feature alignment learning module,FA)来融合并对齐两个分支的输出特征,全面调节农机轨迹在全局和局部不同范围下的上下文信息,提高算法的识别性能。为验证所提方法的有效性,在真实轨迹数据集上进行了实验,结果表明,所提方法相比现有的SOTA模型在水稻和小麦收割机轨迹数据集上的F1-score提高了13.94和11.47个百分点。展开更多
基金Project supported by the National Natural Science Foundation of China(Grant No.61401024)the Shanghai Aerospace Science and Technology Innovation Foundation,China(Grant No.SAST201240)the Basic Research Foundation of Beijing Institute of Technology(Grant No.20140542001)
文摘The imaging plane of inverse synthetic aperture radar （ISAR） is the projection plane of the target. When taking an image using the range-Doppler theory, the imaging plane may have a spatial-variant property, which causes the change of scatter＇s projection position and results in migration through resolution cells, In this study, we focus on the spatial-variant property of the imaging plane of a three-axis-stabilized space target. The innovative contributions are as follows. 1） The target motion model in orbit is provided based on a two-body model. 2） The instantaneous imaging plane is determined by the method of vector analysis. 3） Three Euler angles are introduced to describe the spatial-variant property of the imaging plane, and the image quality is analyzed. The simulation results confirm the analysis of the spatial-variant property. The research in this study is significant for the selection of the imaging segment, and provides the evidence for the following data processing and compensation algorithm.
基金Supported by the National Natural Science Foundation of China under Grant Nos 11534006,11674184 and 11374166the Natural Science Foundation of Tianjin under Grant No 16JC2DJC31300Collaborative Innovation Center of Extreme Optics
文摘We investigate a novel spatial geometric phase of hybrid-polarized vector fields consisting of linear, elliptical and circular polarizations by Young＇s two-slit interferometer instead of the widely used Mach-Zehnder interferometer. This spatial geometric phase can be manipulated by engineering the spatial configuration of hybrid polarizations, and is directly related to the topological charge, the local states of polarization and the rotational symmetry of hybrid-polarized vector optical fields. The unique feature of geometric phase has implications in quantum information science as well as other physical systems such as electron vortex beams.
文摘The development of minimally invasive surgery has transformed the management of gastrointestinal cancer.Notably,three-dimensional visualization systems have increased surgical precision.This editorial discusses a recent study by Shen and Zhang,which compared the clinical applications of naked-eye threedimensional laparoscopic systems vs traditional optical systems in radical surgery for gastric and colorectal cancer.Both systems appeared to yield comparable surgical and oncological outcomes in terms of safety parameters,operating times,and quality of lymph node dissection.However,the spectacle-free system’s technical and logistical limitations hindered its effects on the surgical team’s overall competency.This editorial examines the authors’findings within the broader context of the evolution of oncologic laparoscopy,discusses the relevance of the results in light of the current literature,and proposes future research directions focused on multicenter validation,comprehensive ergonomic analysis,and technological advancements aimed at enhancing intraoperative collaboration.As technology continues to evolve,clinical implementation of new methods must be supported by robust scientific evidence and standardized criteria,to ensure tangible improvements in efficiency,safety,and oncologic outcomes.
文摘农业机械轨迹作业行为模式识别是一项多变量时间序列分类任务,旨在利用轨迹数据的时空特征识别农机的行为模式。针对已有方法未能从频率角度挖掘农机轨迹的全局特性以及识别精度不足的问题,提出了一种面向农机轨迹行为模式识别的频域注意力和U型残差网络FARNet。该网络包含两个不同网络分支,用于全面挖掘农机轨迹的依赖信息。其中一个分支搭载了基于频域注意力的Transformer(transformer based on frequency attention,FAT)来挖掘农机轨迹在频域空间的全局时序依赖;另一分支部署了基于正交约束的U型残差网络(U-shaped residual network based on orthogonal constraints,URNet),其以ResUnet作为骨干网络提取轨迹特征图在不同感受野的深层语义信息,探索轨迹特征间的局部空间依赖。最后设计了一种特征对齐学习模块(feature alignment learning module,FA)来融合并对齐两个分支的输出特征,全面调节农机轨迹在全局和局部不同范围下的上下文信息,提高算法的识别性能。为验证所提方法的有效性,在真实轨迹数据集上进行了实验,结果表明,所提方法相比现有的SOTA模型在水稻和小麦收割机轨迹数据集上的F1-score提高了13.94和11.47个百分点。
