最优控制点选取的遥感图像亚像素配准算法

分析了已有图像配准算法应用遥感图像配准方面的面临的问题,针对提高不同模态遥感图像配准精度问题,提出了一种人工辅助多模态图像配准算法。该算法首先由人工对待配准图像(测试图像)和参考图像输入控制点,利用高斯差分算子确定测试图像极值点;其次利用投影变换和最小线性平方差算法计算双边平均配准误差;最后,根据配准误差自动对控制点进行亚像素调整,取得亚像素级控制点匹配,实现遥感图像精确配准。实验结果表明,该算法具备更高的配准精度。      

基于曲线拟合的干涉合成孔径雷达(InSAR)复图像自动配准算法

由于乘性噪声和相干斑的存在,干涉合成孔径雷达(InSAR) 复图像对的配准算法比传统的图像配准算法更困难。该文提出了一种基于B-样条曲线拟合和匹配的InSAR复图像对的自动配准算法,能够有效地配准InSAR复图像对。在利用最小二乘法建立配准参数后,应用两步法完成了亚像元级配准。实验结果表明,该算法具有较高的精度和稳健性。     

基于PCA与聚类匹配的光谱图像二次配准

针对光谱图像配准方法效果较差的问题,提出一种基于PCA算法和聚类算法相结合新的特征点匹配算法并将二次配准方法引入到配准过程中。该方法首先提取SURF特征点并描述特征点,接着利用本文提出的特征点匹配算法对特征点进行匹配以完成初次配准,最后以初次配准结果作为初始值,采用改进的Powell算法利用图像灰度信息进行精配准。实验结果表明,该方法配准精度高,同时可以有效地缩短了光谱图像配准的时间。     

基于IPSO和综合信息的医学图像配准新方法

针对医学图像配准中采用互信息作为配准相似度函数存在配准精度不高和收敛速度慢等问题,根据图像灰度和空间结构信息,构造了一种新的基于互信息和改进型形态学梯度算子的信息配准测度函数,采用一种适用于医学图像自动配准的改进型粒子群优化(IPSO)算法,给出了一种新的基于IPSO的医学图像配准算法。实验结果表明,该配准算法稳定性好、收敛速度快,在多模态医学图像自动配准中是可行的。     

一种新的SAR图像自动配准算法

由于图像间光谱特性的差异，对图像配准的精度产生严重的影响，因此该文提出了一种基于改进灰度熵的SAR图像自动配准算法。文中首先对SAR图像进行配准前的预处理，提取图像掩模，然后计算图像行和列的灰度熵，将其作为图像的特征信息完成图像间的配准，提出了配准优化算法的完整步骤，并利用多种方法对配准结果进行评价。实验证明该方法可以有效克服图像间亮度不均的影响，充分利用了像素的有用信息，具有很高的配准精度和配准效率，是一种正确可行的图像配准算法。     

时序特征图像配准算法的抗噪性分析

提出了一种基于时序特征变化的图像配准算法,通过时序灰度特征的分析和提取,实现了可见光与红外非固形动态目标图像的有效配准。重点分析了该算法的抗噪能力,当噪声对配准精度产生明显影响时,设计了一种预处理滤波方法,通过减小噪声项在时序特征提取中的影响,从而降低了算法的误配率。利用蜡烛火焰作为探测目标,通过实验表明,在图像信噪比为10 dB的情况下该算法的抗噪性较好,能够实现非固形目标的可见光与红外图像配准。当目标图像信噪比等于6 dB时,该算法就无法完成图像配准,此时采用预处理滤波方法,算法的误配率得以降低,能成功实现非固形目标的可见光与红外图像配准。实验中计算得到预处理方法的关键参数m取0.25时效果最好。     

融合显性和隐性度量的多模图像配准算法

为解决基于隐性度量的图像配准算法初始配准误差大容易引起失配的问题,本文提出了一种融合显性度量和隐性度量的多模图像分层配准算法。首先采用金字塔分解算法得到低分辨率待配准图像。然后在每一层金字塔图像中,先采用互信息作为显性度量,利用粒子群算法获得初始的配准参数;以此作为初始值,采用基于隐性度量的配准算法,利用融合粒子群和鲍威尔搜索法的优化算法获取更准确的配准参数。基于低分辨率图像计算得到配准参数后,先对高分辨率待配准图像进行变换,然后利用提出的上述算法进行参数优化,根据每层得到的配准参数计算最终的配准参数。最后,采用可见光与红外图像、多波段SAR图像进行了配准实验。实验结果表明,提出的算法适用于多模图像配准,能够减小配准误差,具有一定的适用性。      

一种针对多通道GMTI的SAR复图像精确配准方法

提出了一种针对多通道GMTI的SAR复图像精确配准算法。首先,系统分析了配准误差对干涉相位的影响,推导出了对应配准误差的干涉相位Cramer-Rao界,并利用Monte Carlo仿真数据对配准误差的影响进行了量化。在此基础上,通过对待配准图像进行二维精确插值处理,生成模板图像库和相应的相关系数库,以适合多通道GMTI的复图像的相位相关为准则,找出模板图像库和相关系数库中与参考图像具有最大相位相关值的图像作为配准图像,从而实现多通道SAR复图像之间的精确配准。对基于所提算法的实测三通道SAR复图像进行慢动目标检测实验,结果表明所提算法能够提供多通道GMTI所需要的配准精度,证明了该算法的有效性和实用性。      

基于联合平动补偿的InISAR 成像方法研究

图像配准是干涉逆合成孔径雷达(InISAR)成像的关键步骤。针对传统InISAR 三维成像方法在图像配准步骤计算量过大、限制条件过多等问题,提出了一种基于联合平动补偿的InISAR 三维成像算法。该算法通过距离向和方位向的联合平动补偿,在得到二维逆合成孔径雷达(ISAR)图像的同时,完成了不同ISAR 图像的精确配准。与传统的基于相关图像配准的三维成像方法相比,该算法简单易行,计算效率更高。仿真结果验证了不同信噪比条件下该算法的有效性和稳健性。     

基于MRF模型的多模态图像配准技术研究

在医学图像处理领域中,医学图像配准技术极其重要,其价值体现在临床医学中对图像处理技术的应用。在解决多模态图像配准的相关问题时,基于互信息方法的应用最广泛,但在某些特定的应用中该方法受到的约束仍然较多。针对这一情况,提出一种新的医学图像配准算法,模态变换的引入作为此研究算法的基础,之后新的马尔可夫能量函数则根据两幅通过模态变换后的图像矩阵以及原配准图像得以构建。同时,为了优化能量函数引入了一种改进的梯度下降算法,从而得到配准结果。最后,运用不同的医学图像进行配准实验来验证该算法,通过实验证明该配准算法具有良好的有效性及抗噪性能。     

X-ray metrology for high-k atomic layer deposited HfxZr1−xO2 films

Hafnium-based dielectrics are the most promising material for SiO₂ replacement in future nodes of CMOS technology. While devices that utilize HfO₂ gate dielectrics suffer from lower carrier mobility and degraded reliability, our group has recently reported improved device characteristics with a modified Hf_xZr_1−xO₂ [R.I. Hegde, D.H. Triyoso, P.J. Tobin, S. Kalpat, M.E. Ramon, H.-H. Tseng, J.K. Schaeffer, E. Luckowski, W.J. Taylor, C.C. Capasso, D.C. Gilmer, M. Moosa, A. Haggag, M. Raymond, D. Roan, J. Nguyen, L.B. La, E. Hebert, R. Cotton, X.-D. Wang, S. Zollner, R. Gregory, D. Werho, R.S. Rai, L. Fonseca, M. Stoker, C. Tracy, B.W. Chan, Y.H. Chiu, B.E. White, Jr., in: Technical Digest - International Electron Devices Meet, vol. 39, 2005, D.H. Triyoso, R.I. Hegde, J.K. Schaeffer, D. Roan, P.J. Tobin, S.B. Samavedam, B.E. White, Jr., R. Gregory, X.-D. Wang, Appl. Phys. Lett. 88 (2006) 222901]. These results have lead to evaluation of X-ray reflectivity (XRR) for monitoring high-k film thickness and control of Zr addition to HfO₂ using measured film density. In addition, a combination of XRR and spectroscopic ellipsometry (SE) is shown to be a fast and non-intrusive method to monitor thickness of interfacial layer between high-k and the Si substrate.     

Students’ thoughts about the importance and costs of their mobile devices’ features and services

In recent years, handheld devices have become one of the fastest growing communication gadgets. Mobile technology is becoming widespread and research in this area is urgently needed. Using a survey instrument, the thoughts of male and female students regarding the importance and costs of mobile devices were investigated. It was found that students tend to consider the following features important: battery life, mp3 player, video camera, photo camera, storage memory, Bluetooth, design and elegance, clock, calendar, organizer and reminder. Also, they are eager to spend an amount of money so as their mobile device to support them. On average, both genders would pay extra money for such features. However, the majority of females think less of the price than males do. On the contrary, most of the respondents do not consider the following important: touch screen, voice commands, chat, teleconference, encryption and cryptography, common use of files, printing. Therefore, they would not spend any money for these features. Interested decision makers would try to increase their interest on such features. Moreover, all respondents appear to own a mobile phone while most of them do not have Internet connection at home. In general, some gender differences are found in the importance and costs of the mobile devices, but they are not statistically significant.     

Job schedulers for Big data processing in Hadoop environment: testing real-life schedulers using benchmark programs

At present, big data is very popular, because it has proved to be much successful in many fields such as social media, E-commerce transactions, etc. Big data describes the tools and technologies needed to capture, manage, store, distribute, and analyze petabyte or larger-sized datasets having different structures with high speed. Big data can be structured, unstructured, or semi structured. Hadoop is an open source framework that is used to process large amounts of data in an inexpensive and efficient way, and job scheduling is a key factor for achieving high performance in big data processing. This paper gives an overview of big data and highlights the problems and challenges in big data. It then highlights Hadoop Distributed File System (HDFS), Hadoop MapReduce, and various parameters that affect the performance of job scheduling algorithms in big data such as Job Tracker, Task Tracker, Name Node, Data Node, etc. The primary purpose of this paper is to present a comparative study of job scheduling algorithms along with their experimental results in Hadoop environment. In addition, this paper describes the advantages, disadvantages, features, and drawbacks of various Hadoop job schedulers such as FIFO, Fair, capacity, Deadline Constraints, Delay, LATE, Resource Aware, etc, and provides a comparative study among these schedulers.     

Progress in ZnO materials and devices

ZnO is a wide-band-gap semiconductor material that is now being developed for many applications, including ultraviolet (UV) light-emitting diodes, UV photodetectors, transparent thin-film transistors, and gas sensors. It can be grown as boules, as thin films, or as nanostructures of many types and shapes. However, as with any useful semiconductor material, its electrical and optical properties are controlled by impurities and defects. Here, we consider various important donor-type impurities, such as H, Al, Ga, and In, and acceptor-type impurities, such as N, P, As, and Sb. We also examine the effects of a few common point defects, including Zn interstitials, Zn vacancies, O vacancies, and complexes of each. The main experimental techniques of interest here include temperature-dependent Hall-effect and low-temperature photoluminescence measurements, because they alone can provide donor and acceptor concentrations and donor energies. The important topic of p-type ZnO is also considered in some detail.     

智能语音操控在广电端到端的系统设计及应用

在数字化中国的大背景下,在公司个人宽带与互动电视在线交费用户均突破300万户的情况下,为解决我们的用户总找不到自己喜欢看的节目,记不住频道名称和喜欢的节目,提升公司竞争力。通过对前端的点播系统、AAA系统、EPG系统、PORTAL系统、ISMP系统、媒资系统、机顶盒终端的功能进行剥离、整合和详细设计,最后完成终端用户语音操控电视直播、点播、时移、回看、页面操控、系统设置,并支持普通话和四川话方言的目的。达到提升用户体验,增加公司竞争力,促进业务的发展。系统试运行后,用户体验效果得到很大提升,完全达到预期设计目的。     

Thyristor Input-Protection Device Suitable for CMOS RF IC’s

Based on simulation results and accompanying analysis, we suggest a thyristor-type ESD protection device structure suitable for implementation in standard CMOS processes to reduce the parasitic capacitances added to the input nodes, which is very important in CMOS RF ICs. We compare DC breakdown characteristics of the suggested device to those of a conventional NMOS protection device to show the benefits of using the suggested device for ESD protection. The characteristic improvements are demonstrated and the corresponding mechanisms are explained based on simulations. Structure dependencies are also examined to define the optimal structure. AC simulation results are introduced to estimate the magnitude of reduction in the added parasitic capacitance when using the suggested device for ESD protection. The analysis shows a possibility of reducing the added parasitic capacitance down to about 1/45 of that resulting with a conventional NMOS protection transistor, while maintaining robustness against ESD.Jin-Young Choi was born in Seoul, Korea in 1956. He received the B.S. degree in electrical engineering from the Seoul National University, Korea, in 1979, and the M.S. and Ph.D. degrees in electrical engineering from the University of Florida, USA, in 1986 and 1991, respectively. In 1991, he joined Samsung Electronics Memory Division, Korea, where he was engaged in high-speed SRAM development. In 1992, he moved to the Hongik University, Jochiwon, Korea, where he is now an associate professor. His recent research interests include the high-frequency modeling of CMOS devices, CMOS RF circuit design, and analysis & design for ESD protection.Woo Suk Yang was born in Seoul, Korea in 1957. He received the B.S. degree in electrical engineering from the Seoul National University, Korea, in 1979, and the Ph.D. degree in electrical and computer engineering from the North Calorina State University, USA, in 1990. His doctorial research was in the area of signal processing. In 1990, he joined LG Electronics Co. Korea. In 1991, he moved to the Hongik University, Jochiwon, Korea, where he is now a professor. His recent research interests include the high-frequency modeling and various topics in signal processing area.Dongmin Kim was born in Korea in 1956. He received the B.S. and M.S. degrees in electrical engineering from the Seoul National University, Korea, in 1979 and 1984, respectively and the Ph.D. degree in ECE from the University of Michigan, USA, in 1996. Now, he is an assistant professor of the Hongik University, Jochiwon, Korea. His recent research interests include circuit design and analysis.Youngju Kim was born in Seoul, Korea in 1957. He received the B.S. and M.S. degrees in electrical engineering from the Seoul National University, Korea in 1980 and 1985, respectively and the Ph.D. degree in electrical engineering from the Polytechnic University of New York, USA, in 1995, respectively. In 1996, he joined the Hongik University, Jochiwon, Korea, where he is now an assistance professor. His recent research interests include the RF circuit design and LIN wireless systems.     

谈对电子元器件企业设计开发过程中“六性”的审核

通过对电子元器件企业＂可靠性、维修性、保障性、测试性、安全性和环境适应性＂的审核,帮助企业识别＂可靠性、维修性、保障性、测试性、安全性和环境适应性＂,认识＂可靠性、维修性、保障性、测试性、安全性和环境适应性＂在产品设计开发中的重要性,实现产品的＂可靠性、维修性、保障性、测试性、安全性和环境适应性＂指标。     

A Progress Report on Metal–Sulfur Batteries

Nonaqueous conversion‐reaction sulfur chemistry has been attracting increasing attention over the past decade for the development of next‐generation lithium‐based batteries. Li–S batteries are currently approaching a nexus stage from lab‐scale experiments to possible pragmatic applications. Inspired by the success of Li–S chemistry, other metal–sulfur batteries with a variety of metallic anodes, such as sodium, potassium, magnesium, calcium, and aluminum, have also started to attract attention. In comparison to lithium, Na, Mg, Al, K, and Ca are naturally more abundant and affordable. The Na‐S, Mg‐S, Al‐S, K‐S, and Ca‐S battery systems provide a great potential for improving the volumetric energy density of sulfur‐based batteries. The multivalent metal‐sulfur systems, Mg‐S, Al‐S, and Ca‐S, offer better safety features as well. However, the research and development on Na‐S, Mg‐S, Al‐S, K‐S, and Ca‐S batteries is far behind the Li–S system due to many critical challenges. In this progress report, the fundamental principles of various metal–sulfur chemistries are first presented and compared. Then, the historical progress, recent advances, and key challenges of the Li–S, Na‐S, Mg‐S, Al‐S, K‐S, and Ca‐S systems are summarized and discussed. Finally, future efforts and directions for both the fundamental and practical research are prospected.     

Data Integration for Medical Information Management

We present a new method for data integration and security by mixing medical waveforms and images with encrypted patient identifiers and unencrypted ancillary information, such as acquisition parameters, diagnostic comments and notes in textual, pictorial, and voice forms. We vary the sampling rate according to the instantaneous frequency of the signal. Redundant samples (or pixels) are eliminated and replaced by associative data which are labeled using a status string encoded based on the Huffman and run-length techniques. This method achieves both data compression and integration simultaneously, allows synchronized presentation of information from different sources by using multimedia technology, and provides data security features. Mingui Sun received a B.S. degree from the Shenyang Chemical Engineering Institute, China, in 1982, and M.S. and Ph. D. degrees in Electrical Engineering from the University of Pittsburgh in 1986 and 1989, respectively. He was a Graduate Student Researcher from 1985 to 1989 working on signal and image processing projects. Currently, he is a Associate Professor and an Associate Director of the Center for Clinical Neurophysiology in the Department of Neurosurgery at the University of Pittsburgh, and a Director of Research at Computational Diagnostics, Inc. His current research and development interests include advanced biomedical electronic devices, biomedical signal and image processing, sensors and transducers, biomedical instruments, artificial neural networks, wavelet transforms, time-frequency analysis, and the inverse problem of neurophysiological signals. He has over 160 publications in these areas. Qiang Liu received his B.S. and M.S. degrees in electrical engineering from Xidian University, Xian, China, in 1996 and 1999 respectively. He is currently a Ph.D. student at the University of Pittsburgh, Pittsburgh, USA. His further research interests include biomedical signal processing, medical imaging, and image/video segmentation, coding and transmission. Robert J. Sclabassi received the B.S.E. degree from Loyola University, Los Angeles, the M.S.E.E., Engineer in Electrical Engineering, and Ph.D. degrees in electrical engineering from the University of Southern California, and the M.D. degree from the University of Pittsburgh. He was employed in the Advanced Systems Laboratory at TRW, Los Angeles, and was a postdoctoral fellow at the Brain Research Institute at the University of California, Los Angeles. He was on the faculties of Department of Neurology and Biomathematics at UCLA until he joined the University of Pittsburgh. Dr. Sclabassi is currently a Professor of Neurological Surgery, Psychiatry, Electrical Engineering, Mechanical Engineering, Psychiatry, and Behavioral Neuroscience at the University of Pittsburgh. Dr. Sclabassi has published over 400 papers, chapters and conference proceedings. Dr. Sclabassi is a Registered Professional Engineer.     

VoIP技术--语音和数据的集成

介绍了VoIP(Voice over IP)的基本组成构件即网关(Gateway)、网守(Gatekeeper)的概念和用途。讲述了语音在IP网上传输的基本原理，语音和数据、数据和IP包之间的转换和传送过程。在H．323协议栈的基础上详细讲述了H．248，H．225，H．245等通信协议和语音编码G．729，G．23l，G．7ll等协议。最后对VoIP的语音服务质量作了较为详细的阐述。