基于图像内容及底层特征的敏感图像检测方法

内容安全是信息安全领域中的一个新兴且重要的研究方向.敏感图像分析与监控是该方向的一个重要研究分支,近年来愈来愈受到关注.然而,由于其研究起步较晚且图像本身所具有的复杂多样性,相对来说,目前已有的研究成果还比较少且性能上也有待于进一步提升,为此,本文提出了一种新的敏感图像分析检测方法.该方法从基于图像内容分析和基于图像底层特征分析两个方面进行综合分析检测,在基于内容分析方面,通过分析图像是否含有大量的皮肤区域来初步检测敏感图像;在基于图像底层特征分析方面,通过颜色直方图矢量和颜色聚合矢量提取图像的颜色及空间信息,并在此基础上利用SVM进行分类,从而实现敏感图像的初步检测;最后,对上述两方面的初步检测结果进行综合判决,实现敏感图像的最终检测.仿真实验结果出示了该新方法的有效性.     

Design of Distributed Beamforming for Dual‐Hop Multiple‐Access Relay Networks

This paper studies a dual‐hop multiple‐access relay network where two independent source nodes transmit information to a common destination node with the aid of multiple single‐antenna amplify‐and‐forward relays. Each relay node is subject to an individual power constraint. We focus on the design of distributed beamforming schemes for the relays to support the transmission rate requirements of the two sources. To this end, we first characterize the achievable rate region for this network via solving a sequence of corner point optimization problems proposed in this paper. We also develop several low‐complexity suboptimal schemes in closed form. Two inner bounds of the achievable rate region are theoretically shown to be approximately optimal in two special scenarios. Finally, numerical results demonstrate the effectiveness of our proposed approaches.     

Decision Support for DCs Location and Distribute Range Planning: Case Study in Asia-Pacific Region

A major concern is to locate manufacturing plants and/or distribution centers to serve the needs of consumers widely and rapidly with reasonably distributive cost and flexible delivery time. The purpose of this research is to help enhance a decision support for three distribution center locations with different ranges of area in order that the distributors are able to serve appropriately the retailers in twelve large cities or capitals of the nine countries in Asia-Pacific region. The researcher applies the theory of supply chain and logistics management systems with Excel, Visual Basic and Genetic Algorithms programs to find the research results. The research results revealed that the interfaces of Excel, Visual Basic, and Genetic Algorithms programs helped decision support in distribution range selection for 3 distribution center locations. Each of which had different ranges of area for distribution with a rapidly, flexibly distributed time. However, variables resulting in a decision-making should be adjusted under certain circumstances for more reliable, specific needs.     

Simulating and modeling the breakdown voltage in a semi-insulating GaAs P+N junction diode

This work aims to determine the characteristic PN junction diode, subject to a reverse polarization, while I （breakdown voltage） of the inverse current in a GaAs specifying the parameters that influence the breakdown voltage of the diode. In this work, we simulated the behavior of the ionization phenomenon by impact breakdown by avalanche of the PN junctions, subject to an inverse polarization. We will take into account both the trapping model in a stationary regime in the P＋N structure using like material of basis the Ⅲ-Ⅴ compounds and mainly the GaAs semi-insulating in which the deep centers have in important densities. We are talking about the model of trapping in the space charge region （SCR） and that is the trap density donor and acceptor states. The carrier crossing the space charge region （SCR） of W thickness creates N electron-hole pairs： for every created pair, the electron and the hole are swept quickly by the electric field, each in an opposite direction, which comes back, according to an already accepted reasoning, to the crossing of the space charge region （SCR） by an electron or a hole. So the even N pair created by the initial particle provoke N2 ionizations and so forth. The study of the physical and electrical behaviour of semiconductors is based on the influence of the presence of deep centers on the characteristic I（V） current-tension, which requires the calculation of the electrostatic potential, the electric field, the integral of ionization, the density of the states traps, the diffusion current of minority in the regions （1） and （3）, the current thermal generation in the region （2）, the leakage current in the surface, and the breakdown voltage.     

基于特征区域分割的立体视频视觉舒适度评价方法

为准确反映人眼在观看立体视频过程中的视觉舒适程度,提出了一种基于特征区域分割的立体视频视觉舒适度评价方法。首先自适应地分割提取出立体视频序列中的人眼的感兴趣区域,以及相邻帧间的运动区域,然后结合深度感知理论和空间合并技术,对视觉舒适度这一特征量进行建模,最后实现对立体视频视觉舒适度的评定。实验结果表明,使用本文的方法对立体视频视觉舒适度进行评价,可以很好的模拟人眼直接观测的视觉感受,对于视觉舒适度的研究具有重要的参考价值。     

Necessary and Sufficient Stability Criterion and New Forbidden Region for Load Impedance Specification

A necessary and sufficient impedance stability criterion for the DC Distributed power system （DPS） is proposed in small-signal sense, which has no conservativeness when compared with other stability criteria. This paper discusses the stability margin and gives it a definition for the round forbidden region. The reasons why the stability margin varies at different interfaces are also discussed in this paper. With the impedance stability criterion and stability margins considered, the sector forbidden region, which is a new forbidden region and can be easily calculated in the Bode diagram, is proposed to solve the calculation problem of the round forbidden region in engineering application. This paper also provides examples and simulations to verify the necessary and sufficient impedance stability criterion and the proposed sector forbidden region for the load impedance specification.     

基于区域划分和余弦定理的DV-Hop算法改进

针对传统距离向量-跳段定位算法(Distance Vector-Hop,DV-HOP)存在的节点密度分布不均引起误差比较大的问题,根据DV-Hop算法特性,采用余弦定理,证明节点密度越大所得定位坐标越准确,使用余弦定理减少在节点密度过大时因空间角度导致节点路径曲折过多而引起的误差,同时提出一种区域划分的改进算法,减少在节点密度较小区域的误差;利用接收信号强度指示(Received Signal Strength Indicator,RSSI)对跳数进行量化分级处理,改善实际距离与计算出的距离相差过大的情况;最后采用最小二乘法估计未知节点坐标。仿真表明,所提出的算法与传统的DV-Hop算法相比,误差有一定的减少。     

MOSFET栅下碰撞电离与击穿研究

研究了场板终端技术对改善 MOSFET栅下电场分布和碰撞电离率的作用 ,结果表明 ,MOSFET在高压应用时 ,漏极靠近表面的 PN结处电场最强 ,决定器件的击穿特性。通过对实验研究与计算机模拟结果的分析 ,表明在不同的栅压下 ,此处场板长度的大小对栅下电场强度有直接的影响 ,合理地控制场板长度能有效地提高器件的击穿电压。     

基于NCC的图像匹配快速算法

在图像匹配过程中,针对传统归一化积相关（NCC）算法计算量大的问题,提出一种对NCC进行改进的图像匹配快速算法。该算法首先使用差分求和定理改造NCC相似度量函数,以降低匹配计算量。然后提出模板区域分割,设定阈值,进一步去除大量不必要的计算,优化匹配搜索过程,实现了快速匹配。实验结果证明,与传统的匹配算法相比,在保证精度的前提下,计算复杂度大大降低。     

双机协同定位误差分析的研究

测向交叉定位技术是协同定位常用方法,针对双机协同测向交叉定位算法及其定位精度进行理论研究和仿真分析。首先给出了交叉定位造成的模糊区的计算方法,并进行仿真;而后重点对测向定位精度进行分析,给出定位算法和误差计算公式,并针对不同载机间距和间距误差、测角误差下的目标GDOP的分布进行了仿真分析。仿真结果表明,定位误差随着间距误差、测角误差的增大而增大,随着双机间距的增大先减小而后增大。