Liner services network design and fleet deployment with empty container repositioning

Transportation demand of shipping container fluctuates due to the seasonality of international trade, thus, every 3–6 months, the liner company has to alter its current liner shipping service network, redeploy ships and design cargo routes with the objective of minimizing the total cost. To solve the problem, the paper presents a mixed integer linear program model. The proposed model incorporates several relevant constraints, such as weekly frequency, the transshipment of cargo between two or more service routes, and transport time. Extensive numerical experiments based on realistic date of Asia–Europe–Oceania shipping operations show that the proposed model can solve real-case problems efficiently by CPLEX. The results demonstrate that the model can reduce ship’s capacity consumption and raise ships’ capacity utilization.     

Hierarchical Attention for Part-Aware Face Detection

International Journal of Computer Vision - Expressive representations for characterizing face appearances are essential for accurate face detection. Due to different poses, scales, illumination,...     

Mechanical properties of graphene films enhanced by homo-telechelic functionalized polymer fillers via π–π stacking interactions

Most researches on graphene/polymer composites are focusing on improving the mechanical and electrical properties of polymers at low graphene content instead of paying attention to constructing graphene’s macroscopic structures. In current study the homo-telechelic functionalized polyethylene glycols (FPEGs) were tailored with π-orbital-rich groups (namely phenyl, pyrene and di-pyrene) via esterification reactions, which enhanced the interaction between polyethylene glycol (PEG) molecules and chemical reduced graphene oxide (RGO) sheets. The π–π stacking interactions between graphene sheets and π-orbital-rich groups endowed the composite films with enhanced tensile strength and tunable electrical conductivity. The formation of graphene network structure mediated by the FPEGs fillers via π–π stacking non-covalent interactions should account for the experimental results. The experimental investigations were also complemented with theoretical calculation using a density functional theory. Atomic force microscope (AFM), scanning electron microscope (SEM), X-ray diffraction (XRD), nuclear magnetic resonance (NMR), thermal gravimetric analysis (TGA), UV–vis and fluorescence spectroscopy were used to monitor the step-wise preparation of graphene composite films.     

视频侦查中人像智能分析应用及算法优化

人像智能分析指的是对视频或录像中的人像进行结构化和可视化分析,对目标人物进行性别、年龄、发型等特征的智能识别,这项技术在视频侦查中有极高的应用价值。人像识别早期的算法是通过人工提取特征,通过学习低级视觉特征来针对不同属性进行分类学习,这种基于传统方法的模型表现常常不尽如人意。在计算机视觉领域,通过海量图像数据学习的神经网络比传统方法有更丰富的信息量和特征可以被提取。文章尝试通过深度学习技术训练神经网络模型对行人进行检测和识别,对于衣着不同的行人进行智能识别,具有更好的鲁棒性,提升了视频人像识别的准确率,拓展了人工智能技术在身份识别领域的应用。     

膨胀剂对约束条件下UHPC早期抗裂性能影响的试验研究

超高性能混凝土(UHPC)在钢桥面铺装结构中由于铺装材料受到钢板、栓钉等的约束,加上材料本身水胶比小,更易收缩,早期容易承受拉应力导致开裂。通过在拌和过程中添加膨胀剂可以起到减小收缩的作用,但该方法对于约束条件下的UHPC铺装的早期抗裂性能提升是否有促进作用,仍需通过更多试验进行合理评估和深入研究。通过圆环法对3组不同配方的UHPC在均匀约束条件下的早期抗裂性能进行定量测试,并将试验数据进行回归分析和计算,得到了各组的平均应力发展速率,进而评估了各组的开裂风险等级。试验结果表明,通过在UHPC中加入适量膨胀剂,可使其最终收缩应力减小,并可降低其收缩应力发展速率,进一步降低其在约束条件下的开裂风险。     

实验室的质量监督(上)

质量监督是实验室保持人员能力,进行自我完善,日常渐进的重要手段,是持续改进的重要组成部分,也是实验室管理工作的难点,本文给出质量监督的目的、对象、方法、记录等.     

循环流液相等电聚焦方法及应用研究

利用循环流液相等电聚焦技术，对Ｈｂ、Ｃｙｔ．Ｃ和ＢＳＡ混合物进行分离，对ＲＮａｓｅＡ粗制品进行精制。该方法上样量大，活性回收好。对聚焦过程中扩散、对流和电渗的影响进行了讨论。     

电子签章安全漏洞分析研究

文章介绍了电子签章产品的一般结构,提出了电子签章产品的安全漏洞为无法保护文档框架和格式、无法保护文档所有数据和产品自身安全漏洞,分析了漏洞存在原因,并从产品的标准制定、功能完善、自身安全性加固三方面提出了改进建议。     

物联网网络层中基于IPv6的信息安全产品发展趋势研究

文章通过对物联网中的IPv6相关原理、技术等方面的研究和分析,对信息安全产品在新环境下的技术以及应用发展趋势进行分析阐述。为确定信息安全产品标准编制、研发应用等方面的发展方向提供参考,对物联网建设过程中的信息安全产业发展具有积极的建设意义。     

本体在Web服务发现中的应用

现在人们主要通过UDDI来查找Web服务,但其仅仅基于关键字,这决定了它不能根据所需功能的语义来更好地发现Web服务。本文先试图用DAML-S语言来描述Web服务的功能本体,然后提出了一个Web服务匹配框架-DAML-S/UDDI匹配器框架（DUMF）。虽然这一框架本身存在一些不完善的地方,但随着进一步改进,预计将在一定程度上解决Web服务发现的问题。