神光Ⅱ终端光学组件的频率变换逆问题研究

在激光惯性约束核聚变(ICF)精确打靶中,对激光束变换、整形和控制要求苛刻,通过数字在线操控技术能够在线实现激光系统的自动校准,而频率变换是数字在线操控技术中的关键一环。在不进行逼近迭代计算的情况下,基于可逆的傅里叶变换和龙格-库塔算法,对神光II终端靶场系统中负单轴晶体KDP的"I+II"类频率变换的逆问题进行了探讨和相应的算法研究,并给出了对应的模拟计算结果。最后,与神光装置运行的实验数据进行比对,验证了频率变换逆问题算法的正确性和稳定性。     

Search for singlet-triplet bistability or biradicaloid properties in polycyclic conjugated hydrocarbons: a valence-bond analysis

Two classes of molecules displaying singlet-triplet biradical bistability (i.e. species having significant biradicaloid properties) can be designed as follows. Alternant conjugated polycyclic hydrocarbons with numerous fixed double bonds (double bonds that remain unchanged in all its Kekulé resonance structures), a large number of Dewar resonance structures which measures the corresponding diradical resonance, and a small HOMO-LUMO band gap which measures the ease of thermal spin inversion are candidates for singlet triplet biradical bistability. Chichibabin's hydrocarbon ( 1 ) is an example. In addition, in the search for candidate molecules having singlet triplet bistability, one should also examine polycyclic conjugated systems having nonalternant induced spin frustration. Spin frustrated nonalternant polycyclic conjugated hydrocarbons will display singlet-triplet bistability (biradicaloid properties) and are generated from alternant valence-bond diradicals or Hückel molecular orbital diradicals having classical Kekulé structures by appropriate intramolecular joining of two starred or nonstarred positions with bonds, respectively.     

Contrast enhancement of optical coherence tomography images using least squares fitting and histogram matching

A method of contrast enhancement of optical coherence tomography (OCT) images based on least squares fitting and histogram matching is presented. Several different functions are adopted as the probability density functions of the gray levels to fit the normalized histogram of an OCT image and histogram matching is used to enhance the OCT image automatically. The effectiveness of the method is proved by the experimental results.     

Accurate Design of Finline Tapers for Millimeter Wave Applications

The design of nonuniform bilateral finlines on anisotropic substrates for millimeter wave applications is presented. The taper consists of a planar circuit with smooth variation of the slot width profile along the structure. The spectral domain method and Galerkin procedure are combined with the transmission line theory for studying the behavior of tapered bilateral finlines and evaluation of their design parameters. The influence of the substrate anisotropy on the propagation characteristics of these structures is also examined. This technique is general and can be applied to investigate a broad class of planar transmission line tapers.     

颜色视觉匹配中明度阈值的评价

为了对颜色视觉匹配中的明度阈值特性进行评价,采用在阴极射线管显示器上获得的颜色刺激,在5个不同明度的中性色背景下,标准刺激的明度L*从5到95以5个CIELAB单位为变化步长,共对19个不同明度的中性标准颜色进行视觉匹配实验。通过对实验结果的处理和分析,验证了关于颜色刺激视觉辨别明度阈值与背景明度之间相关性的韦伯法则和crispening效应,同时利用颜色视觉匹配数据对基于CIELAB的相关色差公式进行了评价和比较,CIEDE2000具有最好的明度差预测性能,CMC次之,CIELAB和CIE94最差。     

相位匹配法在毛细管波导中产生高次谐波的分析

在理论上研究了超短强激光脉冲与毛细管中惰性气体相互作用产生高次谐波的最佳相位匹配条件,在计算中发现了特定级次 相位匹配中截止驱动光强的存在。分析表明要想使高次谐波获得相位匹配,需要超短脉宽的驱动脉冲以及电离能大的气体工作介质。讨论了毛细管波导中的强场高次谐波选择性并指出了改善谐波选择性的措施。     

准方波脉冲形成网络的理论分析与设计

针对微波驱动源、固态调制器等对宽平顶脉冲源的应用需求,开展了基于脉冲形成网络输出准方波的理论研究与分析设计。首先采用Prony算法获得准方波波形的解析表达式,然后基于最佳一致逼近的优化控制思路,列出极点控制方程,采用数值方法求解非线性方程组,获得优化波形参数。在此基础上采用阻抗函数匹配的算法求解出脉冲形成网络的元件参数初始值,根据工程设计的实际情况舍弃部分参数并对个别参数进行优化后,获得最终的元器件参数。通过理论分析与数值模拟相结合的方法,系统地给出了获取长脉宽、低纹波准方波脉冲形成网络的设计方法以及设计准则。该方法可用于设计任意阶低纹波系数的准方波脉冲源,也可用于设计其他输出波形要求的脉冲源。     

The Soft Point and Its Applications in Body Falling*

From the mesoscopic point of view, a definition of soft point is introduced by considering the attributes of geometric profile and mass distribution. After that, this concept is used to develop the soft matching technique to simulate the chaotic behaviors of the equations. Especially, a tennis model with deformation factor a(t) is proposed to derive a generalized Newton-Stokes equation v′(t) = λ(v T-a(t)v(t)). Furthermore, a concept of duality of deformation factor a(t) and velocity v(t) with re...     

基于灰度曲面匹配的虹膜识别方法

目前的虹膜识别都是采用在图像上提取特征点,并将特征点编码为固定长度的特征数据用于匹配的方式。这种方式使虹膜识别系统易受攻击。为了进一步提高虹膜识别系统的安全性和识别速度,提出了一种基于灰度曲面匹配的虹膜识别方法。该方法抛弃了特征提取和编码等传统操作,在特征分析的基础上直接利用了灰度曲面匹配的思想,首先对两幅图像中的像素灰度做差,得到灰度差曲面,然后求出该灰度差曲面的方差。将此方差作为衡量两个虹膜特征曲面之间距离的依据,并据此判定两个虹膜是否来自同一只眼睛。在给定阈值为40的前提下,正确识别率为96.89%,识别时间为53.2 ms。实验结果证明,该方法识别准确率高,识别速度快。