结合自集成和对抗学习的域自适应城市场景语义分割

针对城市场景标签获取的高额成本问题,文中提出结合自集成和对抗学习的域自适应城市场景语义分割方法.对于源域和目标域的较大域间差异问题,采用风格转换的方法将源域数据集合成具有目标域风格的新数据集,作为新的源域数据集,从而有效减少源域与目标域的域间差异.对于目标域的域内差异问题,引入自集成方法,构造教师网络,利用教师网络在目...     

冗余最小化的IPv6拓扑发现方法

随着网络技术的高速发展,网络管理的重要性越来越突出,正确的网络拓扑是进行网络管理的基础.IPv6是公认的下一代互联网协议,其庞大的地址空间和独特的特征为拓扑发现带来了新的挑战.目前,基于ICMP的拓扑发现分为分布式和集中式两种,其主动探测的特征不可避免地产生探测冗余.分布式拓扑发现方法布署困难并且成本高.更重要的是在冗余减少上存在由探测点间冗余引起的诸多限制,因此它不能以网络友好的方式发现拓扑.由于IPv6路由器对源路由的支持,集中式的拓扑发现方法能够发现交叉链路以保证覆盖率.测量了IPv6环境下单个探测源产生的冗余,提出了冗余最小化的集中式拓扑发现方法.在引入减少冗余的后退算法基础上提出了实际网络环境下的改进算法,说明了集中式拓扑发现在IPv6环境下的可行性.实验结果表明对靠近探测点的节点减少了高达两个数量级的冗余,并能够保证令人满意的覆盖率.     

二进制翻译中间接调用动态链接库恢复技术

针对使用C后端的静态二进制翻译框架,提出了间接调用动态链接库函数恢复方法。该方法使用伪地址来代替需要动态解析的动态链接库函数的地址,在后端构建了伪地址与目标机上动态链接库函数地址的对应表,在后端C代码生成过程中插入根据伪地址获得动态链接库函数地址的查表代码,使用查表得到的库函数地址恢复对于动态链接库函数的间接调用。     

基于单源分段编码的无线对等感知网数据收集模型

针对大规模无线对等感知网络在触发型数据场景下的部分数据收集问题，提出单源分段编码模型。利用单个源节点的多个分段之间编码而成的数据来记录源数据，提高数据可靠性的同时使其适应大规模网络下的部分数据收集。通过使用游走包进行编码操作，避免源节点过多的能量消耗。同时提出针对节点存储空间提出动态划分编码单元策略，利用邻居间的信息交换动态的调整源数据切分的编码单元大小，实现节点存储空间与收集效率的动态调整。并针对灾难场景下的数据收集提出了危险感知编码冗余量动态调整策略，通过对邻居状态的感知动态调整发送的随机游走包的个数，自适应地提升编码冗余量，提高数据恢复率。     

基于嵌入式技术的传感器网络无尺度路由算法

提出了一种基于无尺度特征的传感器网络路由算法，该算法可以实现自组创建并动态调整网络的无尺度网络拓扑结构，并在该拓扑结构的上层构成虚拟骨干网，支撑整个网络的数据传输。为提高数据的冗余可靠性，采用了多路径路由的设计原则。同时结合了信息融合的概念，各骨干节点把收到的冗余互补数据遵循“先融合后转发”的原则，可有效地获得被感知对象的确知信息，大大降低了网络的传播代价，提高了网络的生命周期。     

基于中性突变的朴素基因表达式编程

分子进化中性学说认为生物的进化主要是由中性突变决定的.基因表达式编程(GEP)是一种将基因型和表现型分离的新的进化模型,其突出表现在基因组存在不被表达的中性区.基于朴素基因表达式编程(NGEP)模型研究了NGEP中性区在进化中的作用.主要工作包括:1)进一步完善了基于完全树编码方案的NGEP模型的概念;2)分析了传统GEP和NGEP的基因中性区域特点,指出NGEP存在更自由灵活的中性区域;3)通过控制基因长度和基因数量,调控中性区的大小和数量,研究了NGEP和传统GEP的中性区域在进化中的特殊作用,验证了NGEP的有效性;4)实验表明,在存在相同适度的中性区域条件下,NGEP比传统GEP进化更有效,且NGEP的成功率随中性区域的增加不会发生剧烈变化.     

高速列车数据驱动无模型自适应容错控制

针对高速列车运行控制中的牵引/制动力约束和执行器故障问题,提出一种基于偏格式动态线性化的无模型自适应容错控制(PFDL-MFAFTC)算法.首先,利用无模型自适应控制框架下的伪梯度概念,将难以精确获取参数(列车质量、阻力以及执行器故障等)的高速列车动力学模型转化为偏格式动态线性化数据模型;其次,利用径向基函数神经网络(RBFNN)处理执行器故障引起的非线性;然后,通过压缩映射方法对算法进行严格的收敛性证明,保证算法的收敛性;最后,通过高速列车仿真验证PFDL-MFAFTC算法的有效性和容错能力.     

电涡传感器在水泵橡胶轴承间隙在线测量中的应用

电涡流传感器被用于对水泵转轴与橡胶轴承的间隙进行在线测量,以监测轴承磨损状况。首先通过电磁计算工具设计一个直径为14mm的双层平面电涡流电感。接着对制作在印刷线路板上的电感分别在空气和水中两种条件下的电阻抗特性进行测量。然后制作相应的电感检测电路,得到一个探头和检测电路一体的电涡流传感器。传感器采用RS485接口输出,不受通讯电缆长度影响。传感器经密封防水处理后尺寸为76×25×12mm。传感器经过实验室测试和水泵工况测试。实验室测试进行静态测试和动态测试,获得了传感器在水中进行间隙测量的特性,测量精度达到0.02mm。水泵工况测试在一台轴流泵上进行,试验结果表明电涡流传感器可以对水泵橡胶轴承间隙进行在线测量。     

低开销的无线网络编码机会路由协议设计

针对基于无线Mesh网络编码的机会路由协议的节点转发冗余问题,从线性空间的角度对该类协议的信息传输过程进行建模与分析,提出一种基于动态冗余控制的无线Mesh网络编码机会路由协议。该协议利用零空间确认技术估计各节点的更新信息总量,根据更新总量动态调整节点转发冗余,通过引入转发优先级调度机制,避免不同的转发节点重复转发来自于相同线性空间的编码包。仿真实验结果显示,与经典的MORE协议相比,该协议能提高30％-100％的网络吞吐量,同时降低20％-45％的归一化开销。     

Adaptive Zooming in Web Cartography

Beyond any doubt much of the current web mapping and web GIS applications lack cartographic quality. Thereasons aren't only the technical limitations related to Internet delivery, but also the neglect of one of the maincartographic principles of digital mapping, namely adaptive zooming. Adaptive zooming describes the adjustmentof a map, its contents and the symbolization to target scale in consequence of a zooming operation. The approachdescribed in this paper proposes the combination of two commonly known concepts: on the one hand levelsof detail (LoD) for those object classes, that require high computational cost for the automated generalizationprocess (e.g. buildings, road network); on the other hand an on‐the‐fly generalization for those object classeswhich can be generalized by less complex methods and algorithms (e.g. rivers, lakes). Realizing such interactiveand dynamic concept for web mapping requires the use of vector based visualization tools. The data format bestmeeting the criteria is the W3C standard Scalable Vector Graphics (SVG). Thus, it has been used to implementthe presented ideas in a prototype application for topographic web mapping based on the landscape modelVECTOR25 of the Swiss Federal Office of Topography.     

Fault tolerant control design using adaptive control allocation based on the pseudo inverse along the null space

Fault‐tolerant control systems are vital in many industrial systems. Actuator redundancy is employed in advanced control strategies to increase system maneuverability, flexibility, safety, and fault tolerability. Management of control signals among redundant actuators is the task of control allocation algorithms. Simplicity, accuracy and low computational cost are key issues in control allocation implementations. In this paper, an adaptive control allocation method based on the pseudo inverse along the null space of the control matrix (PAN) is introduced in order to adaptively tolerate actuator faults. The proposed method solves the control allocation problem with an exact solution and optimized l_∞ norm of the control signal. This method also handles input limitations and is computationally efficient. Simulation results are used to show the effectiveness of the proposed method. Copyright © 2016 John Wiley & Sons, Ltd.     

Metamorphosis of arbitrary triangular meshes

Three-dimensional metamorphosis (or morphing) establishes a smooth transition from a source object to a target object. The primary issue in 3D metamorphosis is to establish surface correspondence between the source and target objects, by which each point on the surface of the source object maps to a point on the surface of the target object. Having established this correspondence, we can generate a smooth transition by interpolating corresponding points from the source to the target positions. We handle 3D geometric metamorphosis between two objects represented as triangular meshes. To improve the quality of 3D morphing between two triangular meshes, we particularly consider the following two issues: 1) metamorphosis of arbitrary meshes; 2) metamorphosis with user control. We can address the first issue using our recently proposed method based on harmonic mapping (T. Kanai et al., 1998). In that earlier work, we developed each of the two meshes (topologically equivalent to a disk and having geometrically complicated shapes), into a 2D unit circle by harmonic mapping. Combining those two embeddings produces surface correspondence between the two meshes. However, this method doesn't consider the second issue: how to let the user control surface correspondence. The article develops an effective method for 3D morphing between two arbitrary meshes of the same topology. We extend our previously proposed method to achieve user control of surface correspondence     

Improving locality in binary representation via redundancy.

Binary representation suffers from the problem of positional dependence, where the amplitude of phenotype variation is dependent on the position of the altered genotype bits. However, this is contrary to conventional variation operations that treat each genotype bit equally. Positional dependence can be attributed to the poor locality, which results in neighboring genotypes having low correlation in the phenotype space, reducing the effectiveness of systematic local search and evolutionary search based on small mutation steps. For this purpose, this paper will propose an alternative genotype-phenotype mapping for binary representation that introduces redundancy into the mapping and removes the exponential orderings between the alleles, hence improving the locality between the genotype and phenotype search space. Empirical study conducted based on distribution, locality, and mutation innovation revealed key algorithmic characteristics of the proposed code, and its practicality is validated by comparative studies based on different benchmark optimization problems. Possible approaches to resolve the overrepresentation problem due to redundancy will be suggested, exhibiting its flexibility and variability in implementation.     

基于多分辨率自蒸馏网络的小样本图像分类

因图像数据具有大量的空间冗余信息,传统的多分辨率网络在处理图像数据时会产生较高的计算成本。而自蒸馏学习方法能够在精度与计算成本之间实现动态平衡,使模型在不增加网络深度和宽度的基础上,有效地提高模型精度。提出一种多分辨率自蒸馏网络（MRSDN）,用于解决小样本学习中输入样本的空间冗余问题。从原始网络中分出一个浅层子网络以识别图像的低分辨率表示,并且保持该原始网络识别高分辨率图像特征的能力。同时,在多分辨率网络中引入改进的全局注意力机制,以减少信息损失且放大全局交互表示。利用自蒸馏学习方法将网络中更深层的知识压缩到浅层子网络中,以提升浅层子网络的泛化能力。在此基础上,将低分辨率网络中的粗粒度特征融合到高分辨率网络中,从而提高模型提取图像特征的能力。实验结果表明,在Mini-ImageNet数据集上MRSDN网络对5-way 1-shot与5-way 5-shot任务的准确率分别为56.34%和74.35%,在Tiered-ImageNet数据集上对5-way 1-shot与5-way 5-shot任务的准确率分别为59.56%和78.96%,能有效缓解高分辨率图像输入时的空间冗余问题,提高小样本图像分类的准确率。     

Correspondence consensus of two sets of correspondences through optimisation functions

We present a consensus method which, given the two correspondences between sets of elements generated by separate entities, enounces a final correspondence consensus considering the existence of outliers. Our method is based on an optimisation technique that minimises the cost of the correspondence while forcing (to the most) to be the mean correspondence of the two original correspondences. The method decides the mapping of the elements that the original correspondences disagree and returns the same element mapping when both correspondences agree. We first show the validity of the method through an experiment in ideal conditions based on palmprint identification, and subsequently present two practical experiments based on image retrieval.     

一种鲁棒性强的无线传感器网络位置隐私保护方案

传统无线传感器网络（WSNs）位置隐私保护方案难以解决安全性与网络能耗之间的均衡,为了提高网络隐私信息的安全性,提出一种鲁棒性强的无线传感器网络位置隐私保护方案.首先通过增加伪源节点和伪汇聚节点防止攻击者获得关键节点的位置信息;然后采用伪汇聚节点分组、概率丢弃冗余数据包降低网络资源消耗;最后在Matlab 2012平台下进行仿真对比实验.结果表明：该方案可以提高网络攻击事件检测率,降低网络时延,有效地保护源节点和汇聚节点的位置隐私.     

Topological integration of BIM and geospatial water utility networks across the building envelope

Utility networks comprise a fundamental part of our complex urban systems and the integration of digital representations of these networks across multiple spatial scales can be used to help address priority challenges. Deteriorating water utility infrastructure and low routing redundancy result in network fragility and thus supply outages when assets fail. Water distribution network configurations can be optimised for higher resilience but digital representations of the networks used for simulations and analyses are not integrated with the finer scale networks inside buildings. This integration is hindered by differences in conceptualisation and semantics employed by the relevant data standards. We suggest that the geospatial and geometric data contained in Building Information Modelling (BIM) and water distribution network (WDN) models can be used for their integration; and that this supports the use cases of optimising dynamic network partitioning, reducing the risk of underground utility strikes and planning for future network configurations with higher topological redundancy. In this study, we develop and demonstrate the application of a weight-based spatial algorithm for inferring water network connections between urban-scale WDNs and BIM models, showing that spatial data can be used in the absence of complete or consistent semantic representations. We suggest that the method has potential for transferability to infrastructure for other utility resources (such as waste water, electricity and gas) and make recommendations such as standardising the representation of connection points between disjoint utility network models and extending the normal practical spatial remit of BIM MEP modelling to encompass the space between buildings and WDNs.     

一种基于传输容量控制的DTN动态分段编码路由算法*

针对时延容忍网络（DTN）中基于动态分段编码的路由算法DSNC存在的数据冗余编码和节点投递成功率低的问题,提出了一种基于传输容量控制的DTN动态分段编码路由算法NC-TCC。在源节点处,采用AONT矩阵对分段中的数据包进行编码,以减少数据重复编码的次数;在中间节点处,利用构造的节点运动模型计算当前连接的传输容量,并在满足传输容量的前提下按照优先级从高到低的顺序选择合适数量的数据包进行编码,以提高节点投递成功率。仿真结果表明,提出的算法能够有效地减少数据冗余,提高传输效率。     

Document image decoding by heuristic search

This correspondence describes an approach to reducing the computational cost of document image decoding by viewing it as a heuristic search problem. The kernel of the approach is a modified dynamic programming (DP) algorithm, called the iterated complete path (ICP) algorithm, that is intended for use with separable source models. A set of heuristic functions are presented for decoding formatted text with ICP. Speedups of 3-25 over DP have been observed when decoding text columns and telephone yellow pages using ICP and the proposed heuristics