Ultra-long organic phosphorescent materials, also known as organic long afterglow materials, have the advantages of rich excited state properties, low price and easy to synthesize, good flexibility, ultra-long luminescence lifetime. After the fluorescence decay of the background, the collection of the remaining light can effectively eliminate the background light interference. Based on this, a lateral flow immunochromatographic detection platform based on ultra-long organic phosphorescent nanoprobes (UOP-LFIAs) was constructed in this paper. Firstly, long afterglow crystals were prepared and synthesized, and then the long afterglow solution was obtained by one-step coating polyvinylpyrrolidone and carboxymethyl cellulose on the surface. Next, it was coupled with antibodies to obtain ultra-long organic afterglow phosphorescent nanoprobes. A portable UOP-LFIAs test strip detection analyzer was then designed and produced again. Finally, pepsinogen (PG) I and II were detected using a portable long afterglow immunochromatographic detection device. The results showed that the detection limits were 0.36 ng/mL and 0.57 ng/mL for PG I and PG II, and the detection ranges were 0.36-1200 ng/mL and 0.57-500 ng/mL, respectively. The average recoveries were 101.30 % and 99.30 %, respectively. The UOP-LFIAs assay platform established in this study is not only simple, rapid, cost-effective and macro-preparation, but also can effectively reduce the interference of background light signal and improve the phosphorescence efficiency, thus improving the detection sensitivity. In addition, the platform realizes the combined detection of PG I and PG II, and the results can be obtained quickly in only 5 min. It provides a powerful disease detection platform for bedside testing.
Aiming at the problems of time-consuming and low diagnosis accuracy during the unsupervised training process of traditional DBN, an analog circuit fault diagnosis method based on Improved Multi-Objective Dragonfly Optimized Deep Belief Network (IMODA-ADBN) is proposed. The method employs an improved MODA algorithm instead of the BP algorithm, which improves the classification accuracy of the network and ameliorates the problem of being prone to falling into local optima. The algorithm is tested on three multi-objective mathematical benchmark problems and compared with three well-known meta-heuristic optimization algorithms such as MODA, MOPSO and NSGA-II, and the results demonstrate the stability of the IMODA-ADBN network model. Finally, IMODA-ADBN is applied to the diagnostic experiments of a two-stage quad op-amp dual second-order low-pass filter, and the results show that the method improves the classification accuracy and diagnostic rate while guaranteeing the convergence speed, and is able to effectively realize the classification and localization of difficult faults.
The deterioration of dual-function performance and increased complexity of communication receivers challenge the efficiency of dual-functional radar-communication (DFRC) system based on linear frequency modulation (LFM). This paper designs a delayed jump mapping LFM (DJM-LFM) waveform-based DFRC system. The frequency delay jumps are generated at the specific points in the LFM cycle to modulate data. After carrier mixing and undersampling, the communication receiver realizes dechirp in the digital domain, to reduce the performance requirements of the coherent signal generation and sampling. In the radar processing link, we propose the breakpoint area deletion and concatenation (BADC) scheme, making the radar performance of DFRC consistent with that of unmodulated frequency modulated continuous wave radar (FMCW). The simulation results show that, compared with the scheme without compensation, the BADC scheme reduces the sidelobe amplitude of the range domain profile by about 38 dB. DJM-LFM reduces the symbol error rate of 2 dB compared with existing DFRC schemes based on the LFM waveform.
Existing deep learning (DL)-based synthetic aperture radar (SAR) ship instance segmentation models mostly extract feature subsets at the single level of feature pyramid network (FPN), and also ignore context information of the region of interest (ROI), which both hinder accuracy improvements. Thus, a full-level context squeeze-and-excitation ROI extractor (FL-CSE-ROIE) is proposed to handle these problems. FL-CSE-ROIE has three novelties: 1) full-level, i.e., extract feature subsets at each level of FPN to retain multi-scale features; 2) context, i.e., add multi-context surroundings of different scopes to ROIs to ease background interferences; and 3) squeeze-and-excitation (SE), i.e., balance contributions of different scope contexts to highlight valuable features and suppress useless ones. FL-CSE-ROIE is applied to the fashionable hybrid task cascade (HTC) model. Results on two open SAR ship detection dataset (SSDD) and high-resolution SAR images dataset (HRSID) confirm its effectiveness. Moreover, another two improvements to HTC are also proposed to enhance accuracy further: 1) the raw deconv is replaced with a content-aware reassembly of features block (CARAFEB) to enable larger receptive fields and 2) the raw $1\times1$ conv for the mask information interaction is replaced with a global feature self-attention block (GFSAB) to enhance interaction benefits. Finally, FL-CSE-ROIE surpasses the other nine advanced models, better than the suboptimal model by 2.4%/2.3% detection average precision (AP) and 3.0%/2.5% segmentation AP on SSDD/HRSID.
We consider two-stage hybrid protocols that combine quantum resources and classical resources to generate classical correlations shared by two separated players. Our motivation is twofold. First, in the near future, the scale of quantum information processing is quite limited, and when quantum resource available is not sufficient for certain tasks, a possible way to strengthen the capability of quantum schemes is introducing extra classical resources. We analyze the mathematical structures of these hybrid protocols, and characterize the relation between the amount of quantum resources and classical resources needed. Second, a fundamental open problem in communication complexity theory is to describe the advantage of sharing prior quantum entanglement over sharing prior randomness, which is still widely open. It turns out that our quantum and classical hybrid protocols provide new insight into this important problem.
宽带隙红外光谱响应由于其在硅基光电探测器中的潜在应用而受到了广泛关注.利用离子注入和飞秒脉冲激光制备了一系列掺杂硅基光电二极管,并研究了硫掺杂硅基材料及器件后的宽带隙红外光谱响应特性.结果发现,PN型光电二极管在近红外和中红外光谱区域内表现出几个典型的光响应特征峰值.这几个特征峰对应于不同的子带隙光响应特征的起始能量,与硅带隙内掺杂硫的活性能级一致.这种光谱响应拓宽技术为制造低成本宽带隙硅光电探测器提供了有力的参考方案.
包覆药通常被嵌入固体火箭或导弹发动机的动力系统中,其外观质量直接影响该类动力系统的性能表现.针对包覆药外观存在的形状、尺寸和表面缺陷,提出了一种基于动态先验特征的包覆药多类型外观缺陷深度检测框架,包括:1)将基于深度分类器的形状缺陷检测和基于深度分割网络的尺寸缺陷检测模型集成,去除不同任务间的冗余特征,同时将深度分割网络当前迭代形成的过程特征作为动态先验特征,作用于深度分类器参数下一次迭代更新,加快模型收敛速度;2)将深度分割网络产生的过程特征映射至基于卷积自编码器的表面缺陷检测模型中,指导检测模型快速聚焦于包覆药,抑制任务无关特征重复提取.实验结果表明,该方法在模型功耗、检测效率及检测准确率等方面具有较好的表现.
在输电线路的缺陷检测中,鸟巢以及塑料、碎布等挂空悬浮物多为小目标.其所占像素少,容易被背景干扰,检测精度有待提高.设计了一种全新的两阶段目标检测算法,用于改善对输电线路中鸟巢以及挂空悬浮物的检测效果.为了提高小目标检测的性能,在特征提取模块中融入注意力机制,以学习更为丰富的上下文信息.此外,在检测模块中,设计了基于更为柔和非极大值抑制算法的后处理方法,以减少小目标的丢失.与常用的两阶段目标检测算法相比,该方法在两个类别的平均准确率上分别提高了约4.7%和5.9%,有着更高的实际应用价值.
提出了一种基于结构平衡理论和高阶互信息的符号网络表示算法SNSH,通过反转符号网络中的正负关系生成负图,来挖掘符号网络中隐含的高阶互信息.该方法旨在通过加强的社会平衡理论来模拟符号网络的局部隐含特征,并通过节点局部嵌入、网络全局结构和节点特征属性三者之间的高阶互信息,得到更全面的符合符号网络特性的节点嵌入.
随着电网建设的发展,配电站房及其设备规模持续增长,对边缘物联代理的实用化提出了更多挑战.在Kubernetes常规容器化架构的基础上,提出轻量化设计思路减少边缘物联代理装置对硬件资源的占用和消耗,实现云-边-端业务协同、数据协同和智能协同,同时在边缘侧开展了业务应用APP以及AI视觉学习模型算法库的规划设计,较好地满足了配电网运检业务的需求.
为减少毫米波波束训练的时间和功耗开销,提出了一种基于通信场景的波束特征选择和预测算法.首先,根据功率损耗概率最小化准则选择最优特征波束,并利用最优波束概率生成特征波束集(波束索引的子集).其次,为了获得通信场景的最优波束概率,采用基于局部学习的特征选择聚类算法(LLC-fs).最后,由于场景化特征波束集与最优波束之间为隐式、非线性映射关系,利用了DNN模型逼近该映射,进而使用离线训练模型实现从特征波束集到最优波束的预测.仿真结果表明,使用离线训练场景化DNN模型即可在线预测最优毫米波波束.预测性能可以逼近穷举波束搜索算法,并有效减小波束搜索的开销.
各类光/电孔径的雷达隐身实现及配套用的低散射载体设计已是当前工程领域非常关注的问题,但长期以来一直欠缺系统性的、通用性的方法.该文首先梳理了相关工作的背景和发展动态,接着分析了低散射载体设计中所用到的基本电磁散射原理以及高精度高效率的仿真方法.面向一般性的技术需求,不仅深入研究了总体的设计逻辑以及详细的分区设计过程,结合软件还给出了具体的操作步骤.通过结合典型应用需求的案例设计及结果分析,展示了所述低散射载体设计方法的科学性和有效性,同时也探讨了后续的研究空间.
相比各专业机构发布的高校评价排名,对高校的招生、科研与社会协作影响更为直接的往往是大众心目中的高校排名.采用问卷调查方式,以在校大学生为代表性社会群体,挖掘该群体对高校的大众评价排名与实际高校排名之间的差异,分析高校所在城市的经济指标与人口规模等因素同这一差异的关联,并通过拟构高校的大众评价排名,挖掘排名改变量同城市的经济、人口指标的关系,定量刻画了各项城市指标对高校的大众评价排名的影响.研究发现,驻于城市化水平高、第三产业发达的城市的高校,其大众评价排名往往会大大靠前于其实际排名,展示出地区经济和城市人口等因素对高校的大众评价的深刻影响.
针对数字仪表图像噪声大、图像特征信息不足导致图像识别准确率低的问题,提出了一种基于卷积递归神经网络结合投影阈值分割和数字序列校正的高噪数字仪表图像识别方法.首先,用投影阈值分割二值化算法对图像进行预处理:使用垂直投影法将图像划分为不同区域,根据不同区域的噪声强度自适应设定二值化阈值,对图像进行二值化处理,降低噪声;其次,根据图像之间数字规律变化特点,利用数字序列校正算法将单个数字识别转换为数字序列识别,通过对比不同数字序列的识别概率得出识别结果,解决单张图像特征信息不足导致识别准确率低等问题.实验结果表明,在高噪声数据集上,相较于卷积递归神经网络模型,提出的高噪声数字仪表识别模型在准确率方面提高了约61.95%,达到93.58%.
基于配体-受体(L-R)互作的细胞间通信是细胞相互协同完成复杂生命活动的重要方式.随着单细胞测序技术的快速发展,在单细胞水平上系统地解析细胞间通信网络及功能迅速成为细胞生物学研究的热点.生物信息学家已开发了大量细胞间信号通信预测的方法及平台,为细胞间信号通信研究提供了重要技术支撑.该文简要阐述了细胞间通信的基本生物学过程;并系统比较了目前较具代表性的细胞间通信预测相关数据库、算法以及评测分析研究;最后系统总结了细胞间通信预测方法的发展趋势,并展望了其未来的研究方向.
可逆缩放网络(IRN)的潜变量采用高斯分布嵌入图像高频信息,因其独立随机性无法充分保存图像高频信息,嵌入效果一般,影响重建性能.通过改进可逆缩放网络来提高嵌入高频信息的能力并进一步降低模型的复杂度.首先,特征提取模块采用密集连接结构和通道注意力机制来获取足够的特征信息,同时减少模块参数量;其次,网络的潜变量采用小波域高频子带插值设计,改善高频信息嵌入能力.实验结果显示该算法相比IRN,在Set5、Set14、BSD100和Urban100这4个基准测试集上的PSNR和SSIM分别平均提升了 0.380 dB和 0.014,参数量减少约 1.64×106,计算量减少约 0.43×109,运行时间减少3 ms.表明该算法的重建性能优良,模型复杂度低,具有实用价值.
利用药物重定位和生物信息学的方法预测治疗胃癌的候选小分子药物及其潜在靶点,为胃癌的治疗提供新的思路.首先在胃癌数据集GSE54129、GSE26899和GSE65801中鉴定了 203个差异表达基因,并利用关联图谱(connectivity map,CMap)分析筛选得到 10种候选的小分子化合物.然后,通过对差异表达基因组成的蛋白质-蛋白质相互作用网络进行拓扑性质分析,鉴定TIMP1、PDGFRB、COL1A1等为枢纽基因.进一步,分子对接的结果显示,新鉴定的小分子药物左炔诺孕酮(levonorgestrel)与TIMP1具有较强的结合能力,结合能为-9.24 kcal/mol;已报道的胃癌药物西地尼布(cediranib)与靶点PDGFRB的结合能为-6.28 kcal/mol.另外,潜在靶点基因TIMP1和PDGFRB的表达模式、诊断和预后价值在胃癌数据集中得到了验证.
为解决现有协议普遍存在的离线字典攻击、缺少匿名性、无前向安全等安全缺陷,基于最新安全模型,将KSSTI攻击和注册合法用户攻击加入安全模型评价标准中,形成增强安全模型,提出了一种面向无线传感器网络的多因素安全增强认证协议,实现了用户通过网关与传感器节点两端的安全会话密钥协商.BAN逻辑和启发式分析结果表明该协议实现了双向认证,满足匿名性、前向安全、抵抗内部攻击、抵抗KSSTI攻击等重要安全属性.相比于已有协议,该文协议的安全等级更高且计算量与通信量适中,适用于安全等级要求高且传感器节点计算资源受限的应用场景.
