DCNN深度特征与交替方向乘子的相关滤波跟踪

针对采用相关滤波的判别式目标跟踪遇到的瓶颈问题：由于目标快速移动引起边界效应，使得相关滤波器在学习与更新过程中可能会引入错误，最终错误的累积将导致跟踪失败。在采集深度学习特征与样本相似性度量的基础上，提出一种引入交替方向乘子方法的改进相关滤波目标跟踪算法，选择DCNN深度特征有效地表征待跟踪目标的初始状态，通过在线分类过程中样本相似性比对与半监督学习，辅助解决相关滤波器在学习过程中存在的自学习问题。所提目标跟踪算法特别适合训练样本为持续获得的、同时存储空间较小的机器学习过程，提高目标在快速运动与部分遮挡等复杂情况下的跟踪成功率，针对VOT2016标准测试视频的实验表明：当目标面临快速运动时，对比CN、SAMF、STC算法，所提DA-CFT跟踪算法将跟踪成功率分别由60.4%～73.4%、67.2%～82.9%、80.9%～88.1%提升至85.6%～91.0%。     

结合句法依存信息的方面级情感分类

引入句法依存信息到原方面术语,提出一种新的方面术语表示方法,利用Glove词向量表示单词以及单词与单词之间的依存关系,构造出包含句法依存信息的依存关系邻接矩阵和依存关系表示矩阵,利用图卷积神经网络和多头注意力机制将句法依存信息融入到方面术语中,使得方面术语表达与上下文结构高度相关。将改进后的方面词术语表示替换到现有模型后,模型泛化能力得到有效提升。对比试验和分析结果表明:该方法具有有效性和泛化性。     

明代盔甲种类及制作工艺研究

本文通过盔甲的制作及材料,区分了非金属材料的皮甲、瓠子作甲、藤甲、绵甲、纸甲以及金属材料组成的齐腰甲、齐腰明甲、缉甲、布甲、铁甲、钢甲、铜甲、锡甲等类型,并对其制作方式等进行分析。结合不同材质盔甲的使用地区、战争方式进行研究,以对明代盔甲有一个整体性的认知。     

Modification of exposure conditions by the magnetic field configuration in helicon antenna-excited helium plasma

Modification of exposure conditions downstream in the diffusion chamber has been performed in helicon antenna-excited helium plasma by adjusting the magnetic field(intensity and geometry).In the inductively coupled mode(H mode), a reduction in ion and heat fluxes is found with increasing magnetic field intensity, which is further explained by the more highly magnetized ions off-axis around the last magnetic field lines(LMFL). However, in helicon wave mode(W mode), the increase in magnetic field intensity can dramatically increase the ion and heat fluxes.Moreover, the effect of LMFL geometry on exposure conditions is investigated. In H mode with contracting LMFL, off-axis peaks of both plasma density and electron temperature profiles shift radially inwards, bringing about a beam with better radial uniformity and higher ion and heat fluxes. In W mode, although higher ion and heat fluxes can be achieved with suppressed plasma cross-field diffusion under converging LMFL, the poor radial uniformity and a small beam diameter will limit the size of samples suitable for plasma irradiation experiments.     

南水北调中线水源区浮游植物群落结构 及其生物多样性

为探究南水北调中线水源区丹江口水库浮游植物群落结构变化及其生物多样性，于 2014—2015 年对丹江 口水库进行为期 1 年的调查和分析。发现浮游植物共 8 门 76 属 101 种，其中，绿藻门、蓝藻门及硅藻门占总种数 90.10%，绿藻门种类数最多（45 种），其次是蓝藻门（26 种）和硅藻门（20 种）。浮游植物群落结构随季节变化有所 不同：夏秋季节生物量明显高于冬春季节；优势种在冬季以颗粒直链藻、颗粒直链藻极狭变种、铜绿微囊藻和小 环藻为主，春季以球衣藻、薄甲藻属及小环藻为主，夏季以简单颤藻、单胞衣藻原变种和球衣藻为主，秋季以类颤 藻鱼腥藻、简单颤藻、球衣藻和小环藻为主。应用 Shannon-Wiener 多样性指数 H′、Pielou 均匀度指数 J′、Margalef 丰富度指数 Dm评价水质结果表明：丹江口水库夏秋季水体营养化程度更高。典型对应分析表明：水温、透明度、 总氮、总磷等是影响浮游植物群落结构的主要环境因子，夏秋季受温度影响蓝藻、绿藻数量逐渐增多，一定程度 上增加了受水区生态变化风险。本研究可为南水北调中线水源区丹江口水库生态环境变化及生物迁移风险提供 数据支持。     

An E-Business Event Stream Mechanism for Improving User Tracing Processes

With the rapid development in business transactions, especially in recent years, it has become necessary to develop different mechanisms to trace business user records in web server log in an efficient way. Online business transactions have increased, especially when the user or customer cannot obtain the required service. For example, with the spread of the epidemic Coronavirus (COVID-19) throughout the world, there is a dire need to rely more on online business processes. In order to improve the efficiency and performance of E-business structure, a web server log must be well utilized to have the ability to trace and record infinite user transactions. This paper proposes an event stream mechanism based on formula patterns to enhance business processes and record all user activities in a structured log file. Each user activity is recorded with a set of tracing parameters that can predict the behavior of the user in business operations. The experimental results are conducted by applying clustering-based classification algorithms on two different datasets; namely, Online Shoppers Purchasing Intention and Instacart Market Basket Analysis. The clustering process is used to group related objects into the same cluster, then the classification process measures the predicted classes of clustered objects. The experimental results record provable accuracy in predicting user preferences on both datasets.     

Heat transfer in dusty fluid with suspended hybrid nanoparticles over a melting surface

The melting effect with the magnetic field performs a significant role in various manufacturing and industrial applications, such as welding, casting, magma-solidification, nuclear engineering, and so forth. The present study focuses on the impact of the melting effect and magnetic field with inhomogeneous heat origination and sink. The formulation of the mathematical model is done by considering fluid with hybrid nanoparticles and dust particles in two different phases. We have considered Fe₂SO₄ and Cu as nanoparticles dispersed in the base fluid water along with suspended dust particles. The set of partial differential equations is reduced by using apt similarity variables and boundary conditions to obtain ordinary differential equations. The numerical solution is approximated using MATLAB-bvp4c adopting the shooting technique. The impact of numerous pertinent physical parameters on the velocity and thermal profiles is plotted and deliberated. Furthermore, the rate of heat flow and friction factor is also tabulated and visualized through the graphs. Streamlines are also drawn to know the behavior of the fluid flow. The rise in values of M_E quickly increases the velocity of the fluid motion but declines the thermal gradient and thickness of its related boundary layer. Also, inclining values of Pr enhance the thermal profile due to the impact of melting.     

2011～2020年华北平原气溶胶光学厚度时空分布特征及潜在源分析

随着经济高速增长和城市化进程不断加快,华北平原区域性空气污染问题愈演愈烈。针对该区域开展长时序气溶胶光学厚度时空分布特征和潜在源分析研究,对华北平原大气污染治理具有重要意义。基于长时序MODIS/Terra C6.1 MOD04_L2气溶胶光学厚度产品,分析华北平原气溶胶光学厚度的时空分布特征,并利用后向轨迹聚类分析讨论华北平原7个重点城市气团输送的季节变化,并以污染较为严重的河北石家庄为例进行潜在源分析和浓度权重分析,探究影响其大气质量的污染物潜在源区。结果表明:2011～2020年华北平原气溶胶光学厚度月均值呈显著的周期性变化,以年为周期,每个周期内峰值一般出现在6月至8月; 气溶胶光学厚度月际年内呈单峰分布,峰值出现在6月(0.75),最小值出现在12月(0.37); 气溶胶光学厚度季均值从大到小依次为夏季(0.67)、春季(0.59)、冬季(0.49)、秋季(0.46); 10年间气溶胶光学厚度呈下降趋势,整体下降幅度达36.84%,其中2011年最高(0.72),2018年最低(0.45); 华北平原7个重点城市春、夏、秋、冬四季主要受短距离气团输送影响较大,长距离气团输送影响较小; 2014～2020年河北石家庄的空气质量优良天数占比相对较小,空气质量状况差,影响其空气质量的污染物多为本地生成,同时也受周边省市近距离输送的影响。