Window annealing for pixel-labeling problems

The pixel labeling problems in computer vision are often formulated as energy minimization tasks. Algorithms such as graph cuts and belief propagation are prominent; however, they are only applicable for specific energy forms. For general optimization, Markov Chain Monte Carlo (MCMC) based simulated annealing can estimate the minima states very slowly.This paper presents a sampling paradigm for faster optimization. First, in contrast to previous MCMCs, the role of detailed balance constraint is eliminated. The reversible Markov chain jumps are essential for sampling an arbitrary posterior distribution, but they are not essential for optimization tasks. This allows a computationally simple window cluster sample. Second, the proposal states are generated from combined sets of local minima which achieve a substantial increase in speed compared to uniformly labeled cluster proposals. Third, under the coarse-to-fine strategy, the maximum window size variable is incorporated along with the temperature variable during simulated annealing. The proposed window annealing is experimentally shown to be many times faster and capable of finding lower energy compared to the previous Gibbs and Swendsen-Wang cut (SW-cut) sampler. In addition, the proposed method is compared with other deterministic algorithms like graph cut, belief propagation, and spectral method in their own specific energy forms. Window annealing displays competitive performance in all domains.     

An object-based middleware supporting efficient interoperability on a smart home network

This paper proposes a novel object-based home network middleware for supporting the interoperability among home devices and smart grid devices. This middleware provides various types of abstract objects for flexible representation of heterogeneous home devices, which are classified based on their characteristics. It is also flexible enough to allow addition of new protocols and message conversion between different protocols through an abstraction layer, which are not supported by existing home network middlewares so that various protocols can be supported. As a result, it can be utilized to provide the interoperability among a variety of devices from sensors to typical home appliances as well as smart grid devices such as a home electric generator and a battery, which adopt different protocols. In order to demonstrate that home devices and smart grid devices are interoperable through the proposed middleware, we implement applications based on the middleware on a target platform consisting of embedded boards, sensors and laptops for emulation purposes of a home network. According to our emulation, this middleware can provide efficient interoperability among home devices and smart grid devices for future energy efficient home.     

电网电压波动对变频器的危害及其解决措施

结合实际,介绍电压波动对变频器的影响,提出变频器过电压和欠电压两种不同情况的解决措施.     

Bidirectional controlled quantum teleportation and secure direct communication using five-qubit entangled state

A scheme is presented to implement bidirectional controlled quantum teleportation (QT) by using a five-qubit entangled state as a quantum channel, where Alice may transmit an arbitrary single qubit state called qubit A to Bob and at the same time, Bob may also transmit an arbitrary single qubit state called qubit B to Alice via the control of the supervisor Charlie. Based on our channel, we explicitly show how the bidirectional controlled QT protocol works. By using this bidirectional controlled teleportation, espcially, a bidirectional controlled quantum secure direct communication (QSDC) protocol, i.e., the so-called controlled quantum dialogue, is further investigated. Under the situation of insuring the security of the quantum channel, Alice (Bob) encodes a secret message directly on a sequence of qubit states and transmits them to Bob (Alice) supervised by Charlie. Especially, the qubits carrying the secret message do not need to be transmitted in quantum channel. At last, we show this QSDC scheme may be determinate and secure.     

Long-wave interfacial instabilities in a thin electrolyte film undergoing coupled electrokinetic flows: a nonlinear analysis

Spatiotemporal deformations of the free charged surface of a thin electrolyte film undergoing a coupled electrokinetic flow composed of an electroosmotic flow (EOF) on a charged solid substrate and an electrophoretic flow (EPF) at its free surface are explored through linear stability analysis and the long-wave nonlinear simulations. The nonlinear evolution equation for the deforming surface is derived by considering both the Maxwell’s stresses and the hydrodynamic stresses. The electric potential across the film is obtained from the Poisson–Boltzmann equation under the Debye–Hückel approximation. The simulations show that at the charged electrolyte–air interface, the applied electric field generates an EPF similar to that of a large charged particle. The EOF near the solid–electrolyte interface and the EPF at the electrolyte–air interface are in the same (or opposite) directions when the zeta potentials at the two interfaces are of the opposite (or same) signs. The linear and nonlinear analyses of the evolution equation predict the presence of travelling waves, which is strongly modulated by the applied electric field and the magnitude/sign of the interface zeta potentials. The time and length scales of the unstable modes reduce as the sign of zeta potential at the two interfaces is varied from being opposite to same and also with the increasing applied electric field. The increased destabilization is caused by a reverse EPF near the free surface when the interfaces bear the same sign of zeta potentials. Flow reversal by EPF at the free surface occurs at smaller zeta potential of the free surface when the film is thicker because of less influence of the EOF arising at the solid–electrolyte boundary. The amplitude of the surface waves is found to be smaller when the unstable waves travel at a faster speed. The films can undergo pseudo-dewetting when the free surface is almost stationary under the combined influences of EPF and EOF. The free surface instability of the coupled EOF and EPF has some interesting implications in the development of micro/nano fluidic devices involving a free surface.     

Quantitative characterization of the focusing process and dynamic behavior of differently sized microparticles in a spiral microchannel

In this paper, a spiral microchannel was fabricated to systematically investigate particle dynamics. The focusing process or migration behavior of different-sized particles in the outlet region was presented. Specifically, for focused microparticles, quantitative characterization and analysis of how particles migrate towards the equilibrium positions with the increase in flow rate (De = 0.31–3.36) were performed. For unfocused microparticles, the particle migration behavior and the particle-free region’s formation process were characterized over a wide range of flow rates (De = 0.31–4.58), and the emergence of double particle-free regions was observed at De ≥ 3.36. These results provide insights into the design and operation of high-throughput particle/cell filtration and separation. Furthermore, using the location markers pre-fabricated along with the microchannel structures, the focusing or migration dynamics of different-sized particles along the spiral microchannel was systematically explored. The particle migration length effects on focusing degree and particle-free region width were analyzed. These analyses may be valuable for the optimization of microchannel structures. In addition, this device was successfully used to efficiently filter rare particles from a large-volume sample and separate particles of two different sizes according to their focusing states.     

OpenStack Heat与企业应用部署

OpenStack Heat在开源应用的部署上有很大优势,但在企业应用部署方面有不少问题。本文罗列这些问题．并就其中的一部分给出解决方案,对尚无成熟解决方案的问题努力使其明朗化。     

基于FPGA的PXIe总线DMA设计与实现

PXIe总线标准是PCIe总线的工业扩展.通过对基于FPGA的PCIe IP Core的研究,在FPGA上实现了PXIe总线,并提出了PXIe总线DMA的新设计.从DMA作为总线主设备和总线从设备两种功能出发,设计了两个DMA通道,分别完成数据读和包的传输.并且设计了将小块离散数据整合成连续数据块的优化算法,使四通道PXIe能达到500MB/s的读写速度.经过实际信号的测试,验证了DMA模式数据传输的正确性及传输带宽.该系统可以满足高速PXIe总线传输带宽的要求,在工程应用中有显著的参考价值.     

基于LAMP的高性能Web服务器的架构

文章提出了基于LAMP的高性能Web服务器的架构方案,采用了Apache日志、Webalizer日志分析、Cacti流量监控、入侵检测的方法,架构了一个完善的、稳定的、安全的、低廉的高性能Web服务器,满足了中小型企业的要求。     

基于S12微控制器的智能车软件设计

以"飞思卡尔"杯全国大学生智能汽车比赛为背景,制作一个在专门设计的跑道上能自主识别道路行驶的智能车,能在尽可能短时间内跑完全程,不脱离跑道并遵守大赛的一系列规则.以9S12XS128为控制核心主要完成工作如下:①摄像头获得的图像信息的数据采集,并完成图像的滤波去噪,提高采集的图像数据准确性.②采用PID算法对智能车的舵机进行控制,提高了舵机的相应速度,减少了对舵机控制的静态误差.③采用了模糊控制法,对智能车的电机部分进行控制,通过不断调试完成模糊规则库的建立.④编写了无线传输模块的软件程序,可以对智能车行驶过程中的状态进行采集并传输至上位机,方便了智能车的调试.通过实验验证,并参加实际大赛,本设计中的所采用的方法效果较好.