Large-field objective lens for multi-wavelength microscopy at mesoscale and submicron resolution

Conventional microscopes designed for submicron resolution in biological research are hindered by a limited field of view,typically around 1 mm.This restriction poses a challenge when attempting to simultaneously analyze various parts of a sample,such as different brain areas.In addition,conventional objective lenses struggle to perform consistently across the required range of wavelengths for brain imaging in vivo.Here we present a novel mesoscopic objective lens with an impressive field of view of 8 mm,a numerical aperture of 0.5,and a working wavelength range from 400 to 1000 nm.We achieved a resolution of 0.74μm in fluorescent beads imaging.The versatility of this lens was further demonstrated through high-quality images of mouse brain and kidney sections in a wide-field imaging system,a confocal laser scanning system,and a two-photon imaging system.This mesoscopic objective lens holds immense promise for advancing multi-wavelength imaging of large fields of view at high resolution.

Human intrusion detection for high-speed railway perimeter under all-weather condition

Purpose – The paper aims to solve the problem of personnel intrusion identification within the limits of highspeed railways. It adopts the fusion method of millimeter wave radar and camera to improve the accuracy ofobject recognition in dark and harsh weather conditions.Design/methodology/approach – This paper adopts the fusion strategy of radar and camera linkage toachieve focus amplification of long-distance targets and solves the problem of low illumination by laser lightfilling of the focus point. In order to improve the recognition effect, this paper adopts the YOLOv8 algorithm formulti-scale target recognition. In addition, for the image distortion caused by bad weather, this paper proposesa linkage and tracking fusion strategy to output the correct alarm results.Findings – Simulated intrusion tests show that the proposed method can effectively detect human intrusionwithin 0–200 m during the day and night in sunny weather and can achieve more than 80% recognitionaccuracy for extreme severe weather conditions.Originality/value – (1) The authors propose a personnel intrusion monitoring scheme based on the fusion ofmillimeter wave radar and camera, achieving all-weather intrusion monitoring;(2) The authors propose a newmulti-level fusion algorithm based on linkage and tracking to achieve intrusion target monitoring underadverse weather conditions;(3) The authors have conducted a large number of innovative simulationexperiments to verify the effectiveness of the method proposed in this article.

A study on aerodynamic noise characteristics of a high-speed maglev train with a speed of 600 km/h

Purpose–This study aims to explore the formation mechanism of aerodynamic noise of a high-speed maglev train and understand the characteristics of dipole and quadrupole sound sources of the maglev train at different speed levels.Design/methodology/approach–Based on large eddy simulation(LES)method and Kirchhoff–Ffowcs Williams and Hawkings(K-FWH)equations,the characteristics of dipole and quadrupole sound sources of maglev trains at different speed levels were simulated and analyzed by constructing reasonable penetrable integral surface.Findings–The spatial disturbance resulting from the separation of the boundary layer in the streamlined area of the tail car is the source of aerodynamic sound of the maglev train.The dipole sources of the train are mainly distributed around the radio terminals of the head and tail cars of the maglev train,the bottom of the arms of the streamlined parts of the head and tail cars and the nose tip area of the streamlined part of the tail car,and the quadrupole sources are mainly distributed in the wake area.When the train runs at three speed levels of 400,500 and 600 km$h1,respectively,the radiated energy of quadrupole source is 62.4%,63.3%and 71.7%,respectively,which exceeds that of dipole sources.Originality/value–This study can help understand the aerodynamic noise characteristics generated by the high-speed maglev train and provide a reference for the optimization design of its aerodynamic shape.

A Study of Precision Factors of Large-scale Object Surface Profile Laser Scanning Measurement

In this paper, we presented a method of using the l as er scanning triangulation for the non-contact 3D surface profile measurement of large-scale object. The characteristic of large-scale object non-contact mea surement is analyzed and the measuring method is proposed. Main factors influenc ing measurement precision such as image distortion and accurate designation of s peckle center are analyzed and methods of solving these problems are proposed. W e designed a combined filter by which the pulse noise and the Gaussian noise of speckle image can be eliminated efficiently. Using the characteristic of intensi ty distribution of laser speckle image we proposed a new approximating method th at could locate the center of laser speckle image at sub-pixel. The auxiliary v ariables are set to linearize the relationship between the image displacement an d the distance, the accurate values of laser triangulation system parameters cou ld be calibrated accurately and the measuring precision is increased remarkabl y. Using the above techniques we designed a measuring system based on laser sc anning triangulation. The results of the experiment show that these methods can raise the measuring precision of large-scale 3D surface profile effectively.

Investigation and analysis on indoor air quality of large commercial office building in summer

To study the indoor air qualities（IAQ）of large commercial office buildings in Hunan province of China and the corresponding improvement methods,the IAQ of a large commercial office building in Changsha in July,2008,is investigated.A questionnaire survey and field tests are used to collect data.According to the data of twelve rooms in this building,objective evaluation and the subjective evaluation of the IAQ are obtained.Almost all of the environmental parameters in these rooms basically meet the standards of the objective evaluation.But the average concentration of carbon dioxide in most rooms cannot reach the value of the cleanliness standards,1 255 mg/m^3.The average acceptability of the IAQ in these rooms is 71%,which is lower than the value of the ASHRAE 55—1992 standards,80%.The proper increase in the wind speed and the indoor fresh air supply can greatly improve the objective evaluation and the subjective evaluation of the IAQ.

Light Environment Analysis and Design Strategies of Large Commercial Buildings-A Case Study in Harbin

This paper presents a comprehensive survey for the situation of light environment in large commercial building through architecture design information,subjective and objective fieldwork and online survey for three typical commercial building forms( centralized form, covered mall shopping center, lifestyle center).Descriptive method is used for statistical analysis which includes the evaluation of light environmental quality,subjective assessment among daylight and artificial illumination,effect of physical characteristics on subjective assessment of light environment. Based on the results of statistical analysis,the improving strategies for light environment are proposed in terms of making use of daylight,artificial lighting design and improvement of the overall light environmental quality. The design strategies which based on the consumer needs is very important can provide help for further impact prediction on the quality of light environment.

For the Greater Good?—A Critical Reflection on Assessing Indirect Economic Effects Caused by Large Transport Projects

Investing in large transport projects affects the (potential) economic development of metropolitan areas. Yet, very little critical research has been performed to understand how to assess these effects. The relationship between infrastructure investments and regional economic development is complex and indirect, and many theoretical and methodological difficulties remain. On the one hand, the assumption that investing in infrastructure is important to sustain economic growth is sometimes doubted. On the other hand, it is argued that investments in infrastructure enhance the accessibility of urban regions and that in the slipstream of such investments, social problems in urban regions can be tackled as well. Despite these contrasting views, there is at least a consensus that transport infrastructure development depends on economic development and vice versa. Yet, in many cases, the method of assessing economic impacts highly affects the results. Therefore, this paper focuses on a critical reflection of methods for estimating economic effects of infrastructure investments. A critical evaluation is made based on Indonesian and Japanese cases. After conducting in-depth desk research on both cases, we found that the broader effects on affected group of people tend to be overlooked due to the problems of time and space dimensions, the chain reaction of effects, and inappropriate data practices. The assessment on the appraisal processes tends to overlook the broader economic implication due to narrow focus and the concept of efficiency of economic theory.

大模型时代:电力视觉技术新起点

随着无人机、巡检机器人和远程监控系统在输电、变电、配电、安监等电力场景中的广泛应用,利用电力视觉技术完成对海量巡检图像的自动处理,能够进一步提升电力系统智能化运维水平,对我国源网荷储一体化进程的快速推进具有至关重要的作用。随着通用视觉大模型的兴起,电力视觉技术正处于从传统深度学习时代向大模型时代跨越的重要节点。该文首先综述了电力视觉技术和通用视觉大模型的最新研究进展,结合视觉大模型在多种公共场景的应用先例,探讨视觉大模型在电力视觉领域将面临的3重能力边界问题。从初步探索通用视觉大模型的潜力,到逐步构建电力视觉大模型的过程,提出4种模型应用范式以突破视觉大模型能力边界。最后分析了视觉大模型对电力视觉研究者的影响,并对大模型浪潮下电力视觉技术的发展方向进行了展望。

改进YOLOv5su模型检测桃树缩叶病

为实现自然环境下桃树缩叶病的检测,该研究提出了一种基于YOLOv5su的桃树缩叶病识别改进模型DLLYOLOv5su。首先,针对桃树缩叶病目标特征变化较大的问题,在骨干网络最后一层C3模块中加入可变形自注意力模块(deformable attention,DA),使模型更加关注目标区域,降低背景对模型的影响,提高模型在复杂背景下的拟合能力。其次在SPPF(fast spatial pyramid pooling)模块中引入LSKA(large separable kernel attention)结构,大核卷积增大了模型的感受野,使模型能够关注更多信息。最后,提出了LAWD(lightweight adaptive weighted downsampling)模块,使用轻量化的下采样结构替换卷积模块,减少计算开销。在桃树缩叶病数据集上进行试验,结果显示,DLL-YOLOv5su模型权重大小为17.6 MB,检测速度为83帧/s。识别准确率P、召回率R和平均精度均值mAP_(50)分别达到了80.7%、73.1%和80.4%,相较于原始YOLOv5su分别提高了4.2、2.4和4.3个百分点。与YOLOv3-tiny、Faster R-CNN、YOLOv7和YOLOv8相比mAP_(50)分别高出了28.5、11.8、2.1和4.1个百分点。改进模型识别精度高,误检、漏检率低,检测速度满足实时检测的要求,可以为桃树缩叶病的实时监测和预警提供参考。

基于CatBoost-MOEAD的大直径泥水盾构施工多目标预测优化

为有效优化盾构施工参数,实现在大直径泥水盾构掘进过程中安全、高效和节能的目标,提出分类助推(CatBoost)和基于分解的多目标进化算法(MOEAD)相结合的混合智能算法;综合考虑盾构施工参数与地质条件,以主要的盾构施工参数为研究对象,选择地表沉降、贯入度和掘进比能为预测和控制目标;优化调控选择的盾构施工参数,并以武汉市轨道交通某号线为例,验证该混合算法的有效性。结果表明:采用CatBoost算法建立的预测模型在大直径泥水盾构上表现出来的预测性能良好,对3个控制目标的拟合精度(R 2)均达到0.9以上;预测模型的重要性排序表明:大直径泥水盾构的总推进力和推进速度对地表沉降、贯入度和掘进比能有显著影响;所提出的CatBoost-MOEAD混合智能算法对3个控制目标的优化效果明显,地表沉降、贯入度和掘进比能分别达到12.35%、7.47%和10.70%的优化幅度,并给出相应盾构施工参数的控制范围。

超越大语言模型:物导向哲学路径下的意义生成

人类创造是人们通过构建关系之网以确证自身在世界之中的位置,塑造意义世界的基础,此种构建需要某种“超越”的外部之物作为支撑。目前,大语言模型已经在智能生成方面表现出巨大的创造潜力,甚至在部分原创性任务中可以取代人类创造,这或将对人类的意义世界造成潜在性威胁。究其症结在于,大语言模型的底层生成逻辑是建立在相关主义基础之上的,而这种基础预设在激进化演绎的同时不可避免地暴露出其深层局限。一方面,大语言模型以“相关主义循环”的方式将自身封闭于字面主义的符号表征与现有的固定范式之内,使得“超越”之维不再可能;另一方面,大语言模型的智能生成功能对于现实架构的渗透性影响又转而超越了相关主义的可理解范畴。从长远角度来看,这种深层矛盾必将冲击一直以来人类从单一主体视域出发建构而成的意义世界。人类何以突破相关主义的内在生成逻辑,维护自身意义世界,这是智能技术时代人类必须直面的问题。立足“物导向哲学”,实现对大语言模型生成逻辑的“超越”,其意义在于这种“超越”的双重性,既能更包容地将大语言模型类智能机器及其功能纳入“物”的可阐释范畴,又能以去中心化的方式突破相关主义束缚,借助“联合实在”的美学活动,通向超越性的外部,为人类创造添注活力,为人类意义世界提供支撑,以此开辟出“超越”大语言模型的意义生成之路。

一种采用混合策略的大规模多目标进化算法

现实中存在大量的大规模多目标优化问题,这些问题所固有的目标函数间冲突性、巨大的搜索空间以及决策变量可能存在的交互等特征对传统的多目标进化算法构成了巨大的挑战.研究者根据此类问题的特点基于不同的视角提出了多种大规模多目标进化算法,但它们在解题的质量和效率方面尚存较大的提升空间.基于此,提出一种采用混合策略的大规模多目标进化算法LSMOEA/HS.该算法提出的一种黄金分层分组方法将大规模决策变量分成收敛性组和多样性组,然后对收敛性变量组执行基于变量组的相关性检测操作,将收敛性变量组划分成若干更小规模的子组,最后算法采用不同的优化策略分别优化收敛性变量组和多样性变量组以获得最终的解题结果.为验证LSMOEA/HS的有效性,将其与五种新近提出的高效的大规模多目标进化算法一同在决策变量维度为200、500、1000、2000和5000的2-目标和3-目标的LSMOP系列测试实例上进行IGD和HV性能测试,实验结果表明LSMOEA/HS具有显著较优的收敛性和多样性.由此表明,LSMOEA/HS是一种颇具前景的大规模多目标进化算法.

1981~2020年华南地区区域性极端降水事件研究

基于华南地区176个国家级自动气象站资料以及1981~2020年ECMWF ERA5再分析资料,采用区域性极端事件的客观识别方法(OITREE)、合成分析等方法,本文研究了华南地区区域性极端降水事件的时空分布特征,并分析了事件偏多年及偏少年的大尺度环流特征。主要结论如下:区域性极端降水事件的频次在年际尺度上的周期变化较为明显,并具有较明显的月变化特征,高发时段为5~6月;在极端强度及影响范围上,华南地区大部分区域性极端降水事件强度约130 mm d^(-1),较少事件强度超出320 mm d^(-1),且区域性极端降水事件的影响范围呈显著上升趋势(约310 km^(2)a^(-1));在事件的综合强度上,综合指数Z呈现显著的上升趋势[0.05(10 a)^(-1)],表明事件强度呈现显著增加的趋势;在大湾区及广东北部,区域性极端降水事件的累计降水及其对总降水的贡献呈显著上升趋势,而在广西南部地区,两者呈下降趋势;在事件偏多年,华南地区存在显著的西南风水汽输送及整层水汽通量强辐合的特征,而在事件偏少年,华南地区具有整层水汽通量辐合偏弱的特征;一般降水日,850 hPa上华南地区位于弱偏东南风区,区域性极端降水事件降水日,华南地区位于气旋性环流的东南部,受到明显的西南风风速大值带影响。

基于人工蜂群算法的大规模武器目标分配研究

针对大规模武器目标分配问题,本文提出一种改进的多目标武器目标分配模型,该模型将武器平台泛化为武器,并将武器平均飞行时间作为第2个优化目标。为有效解决这类问题,本文还提出了改进的自适应离散多目标人工蜂群算法。该算法基于人工蜂群算法和非支配排序策略,引入了自适应算子操作数、重用蜜源探索信息的变异概率策略,并通过蜜源之间、蜜源与外部解集之间的交互以提高算法的收敛性,通过算子的随机选择保持种群多样性。最后通过不同规模武器目标分配的对比实验,证明了所提自适应算子操作数与重用蜜源探索次数的变异概率策略的有效性,并与MOABC、MOPSO、NSGA-II算法在反向世代距离、超体积、时间3个方面进行比较,本文算法能够在保证时效性的前提下得到质量更好的Pareto解集。

面向无人机视角下小目标检测的YOLOv8s改进模型

从无人机视角进行目标检测,面临图像目标小、分布密集、类别不均衡等难点,且由于无人机的硬件条件限制了模型的规模,导致模型的准确率偏低。提出一种融合多种注意力机制的YOLOv8s改进模型,在骨干网络中引入感受野注意力卷积和CBAM(concentration-based attention module)注意力机制改进卷积模块,解决注意力权重参数在感受野特征中共享问题的同时,在通道和空间维度加上注意力权重,增强特征提取能力;通过引入大型可分离卷积注意力思想,改造空间金字塔池化层,增加不同层级特征间的信息交融;优化颈部结构,增加具有丰富小目标语义信息的特征层;使用inner-IoU损失函数的思想改进MPDIoU(minimum point distance based IoU)函数,以innerMPDIoU代替原损失函数,提升对困难样本的学习能力。实验结果表明,改进后的YOLOv8s模型在VisDrone数据集上mAP、P、R分别提升了16.1%、9.3%、14.9%,性能超过YOLOv8m,可以有效应用于无人机平台上的目标检测任务。

基于改进多目标粒子群的大型设备群检测策略优化方法

在大型履带起重机群(大型设备群)检验检测中,存在检测工期紧、检测质量和人员安全要求高、经济效益最大化等多目标约束下的最优检测策略制定问题,为此,提出了一种基于改进多目标粒子群的大型设备群检测策略优化方法(MOPSO)。首先,利用变异算子对传统的多目标粒子群算法优化方法进行了改进;然后,根据大型设备群检测项目的实际需求构建了检测工期、检测成本、检测质量与安全的多目标优化模型,并确定了各子目标的约束条件;最后,将该优化算法和构建的模型应用于大型履带起重机群的检测项目中,对该方法的有效性进行了验证。研究结果表明:与传统检测策略相比,在保证质量和安全的前提下,利用该方法得到最优的检测策略,其检测周期仅需3/4,检测单位成本节省了14%,受检单位成本节约了32.8%,极大地提升了检测效率,降低了企业和检验单位人力、经济和时间成本。因此,该方法具有良好的实用性和推广应用价值。

基于RSM-NCGA算法的大型邮轮钢骨架复合材料夹芯板结构优化研究

采用一种钢骨架复合材料夹芯板结构来降低大型邮轮上层建筑重量.运用3D-Hashin失效准则以及损伤材料刚度矩阵退化方法预测结构的渐进损伤,基于ABAQUS二次开发将该结构参数化建模.采用DOE试验设计中的优化拉丁超立方设计采样,依据采样点建立四次多项式响应面代理模型.以结构质量和极限载荷为优化目标函数,基于NCGA遗传算法在响应面代理模型上开展钢骨架复合材料夹芯板结构多目标优化.结果表明:优化结果的质量比初始方案降低12.1%,极限承载能力提高17.4%;通过DOE试验设计与遗传算法混合优化策略可以应用于钢骨架复合材料夹芯板结构设计中的不同问题.

基于深度学习的小目标检测技术研究进展(特邀)

小目标检测在自动驾驶、安防等领域具有重要的应用价值。然而,由于小目标自身视觉特征不明显、复杂背景干扰以及信噪比低等因素,使得小目标检测一直以来都是一个极具挑战性的难题。笔者系统回顾了当前基于深度学习方法的小目标检测技术,对现有算法进行了系统地归类、分析和比较:界定了小目标检测的概念,总结了小目标检测所面临的主要挑战;着重讨论了几种主要的网络优化策略,如利用数据增强技术提高模型的泛化能力,通过超分辨率技术改善小目标可视性,采用多尺度信息融合技术提升检测精度,以及基于上下文信息学习和大核卷积策略改进特征表达能力、无锚框检测机制、DETR技术和针对特定应用场景的多模态小目标检测等方法并详细分析了其优缺点;全面介绍了现有小目标数据集,并在常用公共数据集上对目前经典的小目标检测算法进行了测试和性能评估;对小目标检测领域未来的研究方向进行了展望,旨在推动小目标检测技术的进一步发展和应用拓展。

多策略大规模多目标优化算法

在解决大规模多目标优化问题(LSMOP)时,随着决策变量维数的增加会使得MOEA/D算法在决策空间扩展性差且容易收敛于局部最优.针对这一问题,提出了一种大规模多策略MOEA/D算法(MSMOEA/D). MSMOEA/D算法在优化过程中引入了一种基于自动编码器的混合初始化策略,以改善初始种群的覆盖程度,从而促进全局搜索.然后,提出一种基于聚合函数值的邻域调整策略,通过调整邻域大小,能够在搜索过程中更精确地控制搜索范围,避免因邻域过大或过小而导致的搜索效率低下.此外,在优化过程中采用了基于非支配排序的变异选择策略.不同的子问题根据位于非支配排序第1层的个体数量选择变异策略,避免种群陷入局部最优,提高算法的整体性能.最后,使用LSMOP和DTLZ测试问题对MSMOEA/D算法和其他已有算法进行了评估.实验结果证实了MSMOEA/D算法解决大规模多目标优化问题的有效性.

基于自变量简约的大规模稀疏多目标优化

现有的大多数进化算法在求解大规模优化问题时性能会随决策变量维数的增长而下降。通常,多目标优化的Pareto有效解集是自变量空间的一个低维流形,该流形的维度远小于自变量空间的维度。鉴于此,提出一种基于自变量简约的多目标进化算法求解大规模稀疏多目标优化问题。该算法通过引入局部保持投影降维,保留原始自变量空间中的局部近邻关系,并设计一个归档集,将寻找到的非劣解存入其中进行训练,以提高投影的准确性。将该算法与四种流行的多目标进化算法在一系列测试问题和实际应用问题上进行了比较。实验结果表明,所提算法在解决稀疏多目标问题上具有较好的效果。因此,通过自变量简约能降低问题的求解难度,提高算法的搜索效率,在解决大规模稀疏多目标问题方面具有显著的优势。