利用阶跃滤波器实现二值图像边缘增强

介绍利用阶跃滤波器实现二值图像边缘增强的光学方法，阶跃滤波器在频域的适当的位移（偏置）既可以改善输出图像的衬底，也能改善其轮廓视锐度，从而有利于图像边缘的检测。     

基于灰度冗余的红外图像自适应输出窗技术

根据红外图像的特点，提出了一种基于灰度冗余的红外图像自适应输出窗技术.该技术通过对图像灰度直方图的统计，设定适当的阈值，将灰度像素分布为零的灰度级进行完全压缩，将小于阈值的灰度级映射到阈值灰度级，然后将有效灰度级在整个可显示灰度级范围内作等间距排列.该算法在压缩灰度冗余的同时，实现了对有效图像灰度级的无损均衡，提高了图像质量，并能实现实时处理.     

固态体积式真三维显示中的灰度级修正研究

基于聚合物稳定胆甾相液晶(PSCT)光阀的固态体积式真三维(3D)显示体,在状态转换瞬间其雾度和透射率的非线性变化会引起成像亮度变化,并引起深度图像间的串扰。通过分析PSCT光阀的光电响应特性和显示体驱动方式,推导出显示体成像亮度的变化函数,并据此计算出子帧亮度和灰度级偏差。通过调整光阀的驱动时序和重新分配数字微镜器件子帧,结合伽马校正,实现了灰度级的修正。给出了在真3D样机上显示的灰度级修正前后的3D成像,实验结果表明,该方法在亮度损失较小(约6%)的前提下,有效地修正了真三维成像的灰度级偏差,并基本消除了深度图像间的串扰,在一定程度上提高了3D显示的成像质量。     

一种抖动与孤立像素点分离相结合的灰度级增强方法

为了使由等离子体显示器(PDP)反γ校正和有限灰度级显示方法引起的灰度损失得到改善,在分析灰度损失原因及有序抖动和随机抖动特点的基础上,提出了一种抖动与孤立像素点分离相结合的灰度级增强方法。该方法首先采用有序抖动和孤立像素点分离对γ校正后的图像进行灰度增强;然后把图像数据映射到有限灰度级,计算有限灰度级与需要显示的灰度级之间的误差;最后采用随机抖动和孤立像素点分离对有限灰度级图像进行灰度增强。实验结果表明,反γ校正和有限灰度级图像的灰度均得到了显著增强,低灰度级和中高灰度级过渡平滑,没有周期性抖动噪声,孤立像素点的亮度显著降低,图像显示效果明显优于仅采用随机抖动或有序抖动的图像。     

基于区域分割的遥感图像融合方法

提出了一种基于区域分割的图像融合方法，先将待融合的图像按空间特性分割成相似度不同的区域，然后根据具体应用目的对每个区域采用不同的融合规则.对多光谱与全色图像、多光谱与雷达图像两类遥感图像对进行了融合实验，实验结果表明，引入区域分割后的融合结果不但性能更优，而且可以很方便地控制不同图像源的成分对融合结果的贡献，满足特定应用的需求.     

一种凹凸边界上特征点的提取方法

提出了一种新方法用于提取用于匹配计算的可靠特征点——凹凸(ridges/troughs)边界上的特征点：沿凹凸边界上的高曲率点，凹凸边界线的交叉点以及图像表面上的最小值点.实验表明，与传统的沿阶跃边界提取特征点的方法相比，该算法提取出的特征点可靠性更高，大大减少了误匹配率.在凹凸边界点检测阶段，只需要对图像扫描一遍即可,而凹凸边界交叉点和高曲率点的提取要消耗更多的时间.因此本方法可用于对特征点可靠性要求很高但数量需求不大的情况.     

束-离子通道电磁模式的边界效应分析

考虑束-离子通道边界上等离子体电子可能发生的扰动，导出了TM模本征方程的理论式.通过对理论结果的数值模拟计算，在阶跃边界情况下将束-离子通道与一般介质波导的电磁关系进行了比较，发现束-离子通道可以通过改变等离子体频率来控制其工作模式.分析对照扰动电荷边界与阶跃边界对束-离子通道电磁模式的影响，观察到在扰动电荷边界情况下，束-离子通道在低频区域(ω＜ω_p,ω_p为等离子体频率)内截止频率显著提高，并在高频区域(ω>ω_p)内出现了新的电磁模式.研究结果对离子通道激光(ICL)和离子通道电子回旋脉塞(ICECM)的设计提供重要的理论依据. 关键词： 束-离子通道 阶跃边界 扰动电荷边界 电磁模式     

基于分段直方图变换的图像非线性增强

针对可见光图像的特性,依据图像的灰度性质与概率密度,提出一种分段直方图灰度图像非线性增强法.根据图像灰度的性质,将图像灰度级分成背景段、过渡段和目标段,由灰度级的累计概率密度,分别构造三部分灰度级的变换函数,压缩背景段灰度差,保持过渡段灰度差,提高目标段灰度,避免图像过度增强,提高了图像的清晰度.以低对比度的可见光图像进行验证,实验结果和理论分析证实了此方法的合理性和有效性,表明此方法不仅实现了图像增强,保留了图像零灰度级附近的灰度,而且减少了细节信息的损失,避免了直方图均衡化和直方图双向均衡化方法中图像过度增强的现象,提高了图像清晰度,获得更加符合人眼视觉的增强效果.     

用于光学系统分析的质心配准法

将图像区域像素的统计特征引入活动轮廓模型,提出使分割轮廓所包括区域的质心对应角均方根误差极小的最优搜索方法——质心配准法。质心配准法有效地避免了散斑噪声等图像污染对图像区域分割精度的影响,实现了模糊图像精确的区域分割,求得图像最佳配准参数。对实际光学系统输出的多幅具有较大空时互补性的低分辨率图像进行了仿真实验,实验结果表明：该配准算法具有较高的精度,并在光学系统分析上得到了很好的应用。     

一种改进的球栅阵列封装焊点射线图像阈值分割算法

球栅阵列封装焊点的射线图像具有信噪比差、背景不均匀等特点,故传统的阈值分割方法无法将目标焊点与背景图像很好的分割.本文通过对球栅阵列封装焊点射线图像直方图的分析,利用了自适应维纳滤波对阈值分割前的图像进行了预处理.根据图像的差异来调整该滤波器的参量,对局部差异大的地方进行小的平滑操作,对局部差异小的地方进行大的平滑操作.在最大类间方差法的基础上,对分割后的图像进行了进一步的分析并提出了改进的二次分割方法.改进的方法为并不直接通过OTSU法进行二值化处理来去除背景,而是在阈值分割得到的两个灰度级内通过计算中值和统计最大灰度像素的方法得到了更优化的阈值,使得去除背景后的焊点图像整体更加清晰和均匀.在背景灰度级内寻找了一个合适的灰度级作为处理后的灰度图像新背景,实验证明该方法明显改进了传统最大类间方差法对球栅阵列封装焊点射线图像的阈值分割效果.     

小型线形可变滤光片分光的可见成像光谱仪及其特性研究

本文设计了一种使用线形可变干涉滤光片的小型成像光谱仪。使用CCD摄像头作图像传感器，通过微动平台带动线形可变干涉滤光片，横向扫描通过面阵CCD的每个像元，每扫描一步所得到的总体图像中不同列的像元是由不同波长的单色部分像元组成，随着扫描步进，这些单色部分像元的波长随之变化，最后采用图像重构的文教是到一系列单色图像（或见光波段400－700mm）。扫描间隔由所需的单色图像的光谱分辨率和各单色图像间的波长间隔来确定。经过试验证明，此成像光谱仪充分利用了线型干涉滤光片的性能，其光谱分辨率为16nm，同后者的光谱带宽相差无几，仪器的空间分辨率由成像系统和CCD来确定，论仪器装置具有结构简单，高空间分辨率，较主光谱分辨率，扫描速度快等特点，通过一个伪装识别试验，验证了仪器的性能指标。     

Mode structure of a semiconductor laser subject to filtered optical feedback

We present an analytical study of the mode structure of a semiconductor laser subject to filtered optical feedback (FOF). First, we derive expressions for the solution sets that define the structure of the modes. We then employ continuation techniques to investigate how this structure changes as filter width, filter detuning and feedback phase are varied. This leads to a comprehensive picture of the FOF laser’s mode structure. In particular, we find an equation for the boundary of the parameter region in which one finds two separate components of modes.     

含油岩心显微荧光成像光谱研究

发展了一种显微荧光光谱成像技术,并将其应用于天然岩心进行显微荧光成像光谱研究。利用这种技术同时采集含油岩心表面的荧光光谱信息和空间信息．并对获得的光谱图像给予光谱学和地质学解释。结果表明,不但能显示岩心形貌和组分的大致趋势,而且能揭示其精细细节．为石油地质研究提供了一种新方法,为今后的石油勘探开发工作提供了一种先进的指导手段。     

Impact of valley polarization on the resistivity in two dimensions

We examine the temperature dependence of resistivity in a two-dimensional electron system formed in a silicon-on-insulator quantum well. The device allows us to tune the valley splitting continuously in addition to the electron density. Our data provide a global picture of how the resistivity and its temperature dependence change with valley polarization. At the boundary between valley-polarized and partially polarized regions, we demonstrate that there is an insulating contribution from spin-degenerate electrons occupying the upper valley-subband edge.     

The compressible hybrid RANS/LES formulation using an additive operator

In this work, the compressible governing equations for the hybrid Reynolds-averaged/large-eddy simulations are formally derived by applying a hybrid filter to the Navier–Stokes equations. This filter is constructed by linearly combining the Reynolds-average (RANS) and large-eddy simulation (LES) operators with a continuous blending function. The derived hybrid equations include additional terms that represent the interactions between RANS and LES formulations. The relevance of these terms is investigated in flat-plate turbulent boundary layer simulations and indicate that these additional terms play a fundamental role in compensating for the turbulence that is neither modeled nor resolved in the transition region between RANS and LES. Results also show that when the additional terms are included, the calculations are not very sensitive to the blending function implemented in the hybrid filter. In the contrary, when these terms are neglected and a step function is implemented in the hybrid filter, nonphysical discontinuities are predicted in the flow statistics.     

Kinetics of the destruction of type I superconductivity by a current

The dynamics of the destruction of type I superconductivity in a wire by an overcritical current are studied. In a first time interval a normal region grows from the surface surrounding a superconducting cylindrical core which contracts until its radius reaches the radius of the intermediate state core according to London's picture of the static state. In a second step the intermediate state is built up between the normal shell and the superconducting core, which finally disappears. The onset of the intermediate state is a consequence of an instability of the interphase boundary.     

Valley polarization in Si(100) at zero magnetic field

The valley splitting, which lifts the degeneracy of the lowest two valley states in a SiO(2)/Si(100)/SiO(2) quantum well, is examined through transport measurements. We demonstrate that the valley splitting can be observed directly as a step in the conductance defining a boundary between valley-unpolarized and -polarized regions. This persists to well above liquid helium temperature and shows no dependence on magnetic field, indicating that single-particle valley splitting and valley polarization exist in (100) silicon even at zero magnetic field.     

Direct Form I Realization of Active Photonic Filters

An integrated photonic architecture is introduced and used to realize an optical filter with direct form I realization. The architecture offers gain from semiconductor optical amplifiers, and this gain results in an active optical filter whose filter response depends on the individual gains. The presence of gain provides advantages in filter performance, and tunable and adaptive functionality. The optical filter is modeled as a discrete time system and the z-transform is used in its analysis and design. A low-pass filter design example is presented and the filter coefficients are derived in terms of gains and coupler splitting ratios. The region of stable operations is derived by applying the Schur-Cohn stability test.     

Direct Form I Realization of Active Photonic Filters

Abstract

An integrated photonic architecture is introduced and used to realize an optical filter with direct form I realization. The architecture offers gain from semiconductor optical amplifiers, and this gain results in an active optical filter whose filter response depends on the individual gains. The presence of gain provides advantages in filter performance, and tunable and adaptive functionality. The optical filter is modeled as a discrete time system and the z-transform is used in its analysis and design. A low-pass filter design example is presented and the filter coefficients are derived in terms of gains and coupler splitting ratios. The region of stable operations is derived by applying the Schur-Cohn stability test.     

基于干扰对消的红外焦平面非均匀性校正算法

红外焦平面器件的非均匀性产生机理复杂,难以准确拟合探测元响应曲线。提出了一种基于相关干扰抵消的非均匀性校正算法,以预先采集到的一帧黑体面源图像做为自适应干扰对消器的参考输入图像,自适应滤波器由参考输入图像迭代计算出待校正红外图像的空间噪声的最佳估计,实现从空间噪声中提取真实图像信号。自适应滤波算法采用变步长最小均方误差算法,减少了算法的运算量,提高了算法的收敛速度。理论分析以及针对实际红外图像的仿真结果表明,提出的算法校正效果好,收敛速度快,更易于工程实现。