神经网络余／十转换的数学原理

本文提出了孙子定理的微分方程求解法，讨论了系数为求余算子的非线性微分方程的稳定平衡点和最佳解的误差。本文构造的非线性微分方程可以唯一地收敛于孙子定理的解，其误差可任意小。该微分方程可用人工神经网络实时计算，实现余／十转换。计算机模拟结果证实了本文理论的正确性。     

STI process steps for sub-quarter micron CMOS

This paper presents Shallow Trench Isolation (STI) process steps for sub-1/4 μ CMOS technologies. Dummy active areas, vertical trench sidewalls, excellent gap filling, counter mask etch step and CMP end point detection, have been used for a 0.18 μm CMOS technology. Electrical results obtained with a 5.5 nm gate oxide thickness show good isolation down to 0.3 μm spacing. Good transistor performances have been demonstrated.     

高功率激光器的光束质量及其对激光加工的影响

本文介绍了光束的模式、评价光束质量的几种方法,着重介绍了国际标准化组织(ISO)提出的光束传输系数M作为评价光束质量的方法。文中还讨论了光束质量对焊接、切割加工的影响;热透镜效应对加工的影响等。     

离散Hopfield神经网络和联想记忆的开关电流技术实现

研究了离散Ｈｏｐｆｉｅｌｄ神经网络（ＤＨＮＮ）和联想记忆神经网络的开关电流技术实现，利用多权输入跨导，开关电流延迟器（ＳＩＤ）和可编程电流比较器（ＰＣＣ）实现了离散Ｈｏｐｉｅｌｄ神经网络，并提出了利用离散Ｈｏｐｆｉｅｌｄ神经网络实现自联想记忆时相应的开关电流电路，所提出了开关电流神经网络适宜于超大规模集成，能在低电压（如３．３Ｖ）下工作。     

Design and performance analysis of beam steering birefringent mercurous chloride acousto-optic device

I.IntroductionThepcrformanccsofAcousto-optic(AO)dcviccareoftenlimitcdbythepropertiesofAOmcdiawhenitisusedinAOsigna1processingsystems.Thespectrumanalyzersandcorre1ationreceivers,forinstancc,areexpcctcdtoimprovcthereso1utionanddynamicrangcwhilereducingtheproccssorvolume.Rccent1y,ancwclassofmateria1s,themercurousha1ides,promiscstocontributeimportantsolutionstosuchnccdsandisrefincdtopractica1uti1izations.Themecurousha1ideweconsidcrinthispapcrismecurouschloride(Hg,C1,).Thesing1ecrystalofHg,Cl…     

基于CCD采集的Mach-Zehnder干涉条纹图的处理算法

Mach-Zehnder干涉仪适用于研究气体密度迅速变化的状态。由于气体折射率的变化与其密度的变化成正比,而折射率的变化将使通过气体的光线有不同的光程,因此可通过干涉臂变化对干涉条纹图像效果的影响得到气体密度。实测中,采用图像采集卡和CCD来接收Mach-Zehnder干涉仪产生的条纹图像,再通过计算机对条纹图像的条纹间距进行处理,从而得到气体密度的变化状态。从光干涉理论出发,对Mach-Zehnder干涉条纹图像特征进行了分析,建立了Mach-Zehnder干涉条纹的数学模型,并根据此模型设计了处理Mach-Zehnder干涉条纹图像的算法。算法包括图像的预处理(即图像的噪声提取)、图像的二值化及图像的细化。     

Enhanced Image Detection on an ARM based Embedded System

This paper presents a new technique for thedetection of Integrated Circuits within images of Printed Circuit Boards autonomouslyand without the need to be assisted by CAD data. The technique is a key partof a suite of algorithms targeted for an embedded System On Chip architecturebased on the ARM7 platform for real time detection of PCB images for diagnosticpurposes. The technique has a significant reduction in complexity when comparedto conventional approaches such as the Hough Transform. The reduction in complexitymakes the approach ideal for an embedded vision application suchas the one described in this paper. This paper presents the technique, thetarget embedded architecture and results showing the reduction in complexitywhen compared to a Hough Transform.     

长江堤防工程地质信息处理模型及技术方法

长江中下游的主要防洪工程是长江干堤 ,长江堤防线路长、工程浩大、堤基工程地质水文地质条件复杂 ,涉及大量的已有地质数据和资料。今后在整治加固堤防的工程地质勘察工作中 ,还需随时进行数据的收集、存储、检索、分析、统计和维护 ,随时调用这些数据进行综合处理、编制报告和图件 ;在防汛期间可以针对堤防险工险段的险情 ,随时查询和调用有关的资料、数据或者图件 ,提供采取抢险措施决策的地质依据。这既提高了地质数据的利用价值和利用率 ,又保证了信息的准确性、完整性和共享性。因此 ,建立合理的长江堤防工程地质信息处理模型及其成熟的技术方法具有重要的意义。     

狭缝式高速摄影机胶片的计算机判读

根据狭缝摄影机胶片的成象原理,运用数字图象处理技术,对狭缝式高速摄影机胶片判读工作的计算机化进行研究.应用结果表明,该技术先进实用,极大地提高了判读的精度和效率,将是这种胶片判读技术的发展方向.     

Performance and simulation of an on-board guided-wave optical processor for airborne SAR applications

This paper describes the performance and simulation of a compact integrated optical processor for the real-time reconstruction of two-dimensional images in airborne stripmap synthetic aperture radar applications. The functional behavior of the processor is explained in some detail. The design criteria are briefly given. The simulation step allowed the main processor characteristics and properties to be identified. A number of comparisons were obtained in airborne SAR mission scenarios between the predictions of the optical device and those achieved by the modern electronic digital approach, based on the wavefront reconstruction method by matched filtering.