激光直写光刻中线条轮廓的分析

考虑了光刻胶对光吸收作用,在已有描述胶层内光场分布模型的基础上,较为准确地推导出光刻胶层内不同深度位置的光场分布.使用迭代方法计算得到了胶层内曝光量空间分布曲线,分析了不同曝光量下胶层内的线条轮廓,为直写光刻中曝光量的选择提供了依据.实验结果分析与理论分析的结果一致.     

数字全息成像系统的景深和焦深分析

根据全息理论,分析了数字全息成像系统的景深和焦深.针对数字全息不同的记录光路结构,分别给出了焦深的近似表达式.结果表明：数字全息系统的景深和焦深不仅与记录波长以及记录时的数值孔径有关,还与记录时参考光波的偏置情况有关;在记录距离和CCD参量一定的条件下,离轴无透镜傅里叶变换全息对称偏置下的焦深比非对称偏置下的稍小;显微成像情况下离轴无透镜傅里叶变换全息系统的焦深大于同轴菲涅耳数字全息系统的焦深,计算机模拟表明了结果的正确性.     

Optimal resolution with extreme depth of focus

When the focal depth is required to be much larger than the wavelength, λ, the effective NA of the beam with optimal resolution is much less than unity. An aperture that is much larger than this beam's footprint is then of no consequence. Such beams that maximize the mean encircled energy fraction within a cylindrical focal region are shown to depend on only a single parameter, ω, that is proportional to the ratio of the square of the cylinder's radius to the product of its length and λ. A linear combination of Hermite– or Laguerre–Gaussian modes is used to represent these fields in two and three dimensions, respectively. For small ω, the results are compared both to asymptotic expansions and to optimal Gaussian and Bessel–Gauss beams.     

基于深度约束的声速估算方法*

高精度的声速剖面数据是提高水下定位精度的关键。在无法精确获得声速剖面数据的情况下,该文提出了一种基于深度约束,同时根据浮标与水下目标的空间几何距离进行有效声速估算的方法。通过仿真模拟,将该文提出的方法与加权平均声速法、泰勒级数展开法和等效声速剖面法的定位结果进行对比分析。结果表明,新方法不需要依靠声速剖面数据,通过深度约束可有效地估算出声速值,其定位精度与等效声速结果相当,极大地提高了无声速剖面数据下的水下定位精度。     

基于信道估计的自适应连续变量量子密钥分发方法

为使连续变量量子密钥分发协议获得稳定的密钥生成率,需要根据信道变化动态调整发送光脉冲的强度.将光纤量子信道看作加性玻色量子高斯信道,给出高斯态通过玻色量子高斯信道仍得到高斯态的证明过程.通过平衡零差检测后,采用最大似然估计法得到了信道参数,进而根据估计的噪声大小自适应调整Alice发送的光脉冲的强度,从而获得稳定的密钥生成率. 关键词： 量子密钥分发 连续变量 玻色量子高斯信道 信道估计     

基于路径选择的深海水下运动目标被动深度估计

深海水下运动目标的深度估计是目标分辨与识别中的重要一环。基于深海运动目标在深海中的直达波与海面反射波的到达时延与位置之间的关系,寻找时延图上运动目标的候选路径,利用候选路径模拟目标的运动路径,提出了一种可用单水听器实现的水下运动目标深度估计方法。通过用候选路径的估计时延与实际接收时延的比较对路径进行筛选,从而减少了需要的位置先验信息。2018年在南中国海进行的深海实验数据处理的结果表明该估计方法可行,对于实验中距离15 km以内的拖曳声源,深度相对估计误差可以达到15%以内。     

A post-processing technique for extending depth of focus in conventional optical microscopy

In this paper, we propose a post-processing technique to obtain optical microscope images with extended depth of focus using a conventional microscope. With the proposed technique, we collect a sequence of images focused at different depths. We then combine the in-focus regions of each acquired frame to compose a single all-in-focus image. That is, a new image with extended depth of focus is obtained. The key to such an algorithm is in selecting the “in-focus” regions from each frame. In this paper, we describe the technique used to identify the in-focus region on every depth slice. Quantitative simulation results are presented where mean absolute error is used as a metric to assess the algorithm performance. Results using real imagery are also presented for subjective evaluation. Based on subjective evaluation and the quantitative simulation results, we believe that the proposed algorithm provides useful depth of focus extension.     

不同2×2集束聚焦方式下高功率激光靶面光强分布特性

大口径高功率激光装置为提高激光靶面的光强强度, 通常采用2×2集束聚焦的模式进行打靶. 大口径楔形透镜是组成2×2集束聚焦系统的核心元件, 可分为二维离轴楔形透镜、一维离轴楔形透镜和非离轴楔形透镜3类. 为了获得理想靶面光强分布, 基于这3 类楔形透镜, 对比研究相应2×2集束聚焦系统下的靶面光强分布特性. 研究结果表明: 相比离轴楔形透镜, 采用基于非离轴楔形透镜的2×2集束聚焦系统时, 容易在激光靶面获得更窄的主瓣宽度、更强的峰值强度、更高的能量集中度. 研究结果对高功率激光靶场聚焦系统的配置选择有重要参考价值.     

Formation of sub-half-wavelength focal spot with ultra long depth of focus

We theoretically demonstrate the creation of sub-wavelength focal spot with ultra long depth of focus (DOF) non-diffracting. To achieve a sub-half-wavelength focal spot (0.4λ) with modulation, extended DOF (over 9.5λ) in vacuum, an integration of amplitude phase encoding, cylindrical polarization and the combination of dual-beams is utilized. Further analysis also predicts that a flattop intensity distribution along z axis can be observed around the ideal focus point.     

光纤耦合LD输出光场特性研究

根据光纤传输的模式理论,分析了单芯光纤输出端面的光场分布.利用单模光纤测量了纤端光场的强度分布,拟合了光纤输出端面光强分布的近似表达式.根据此数值计算和实验测量了多芯光纤输出端面的强度分布,结果基本吻合.表明光纤出射端面的光强分布应是介于高斯分布和几何分布之间的一种准高斯分布.     

Transverse superresolution with extended depth of focus using binary phase filters for optical storage system

This paper reports the study of superresolution and extension of depth of focus (DOF) of three-zone binary phase filters (BPFs) under high numerical aperture (NA) focusing. Two three-zone BPFs (BPF1 and BPF2) are specially designed to achieve transverse superresolution and extension of DOF simultaneously for the optical disc system with an objective lens of NA = 0.65. Numerical simulations based on the vectorial diffraction theory indicate that the transverse focal spot size of this system can be as small as that of NA = 0.85 and its DOF is as large as that of NA = 0.6 with acceptable Strehl ratios (> 0.4 for BPF1 and > 0.35 for BPF2) and low side-lobe intensity ratios (< 10%) by using these BPFs. It suggests that the storage capacity of this optical storage system may be improved to the same as that of a system with an objective lens of NA = 0.85. Moreover, the strict requirement of the focusing servo system caused by decreasing wavelength or employing higher NA objective lens can be mitigated. Therefore, these BPFs presented in this paper should be highly interesting for applications of high density optical data storage.     

极坐标激光直写系统原点定位的实验研究

分析了旋转线性光栅法对准光轴与旋转机械轴中心的理论准确度,实验研究了极坐标激光直写系统的原点定位.在极坐标激光直写系统下进行了圆环线条写入实验,得出了原点定位的准确度,原点定位准确度优于0.5 μm,实验与理论定位准确度基本一致.     

基于红外和可见光图像逐级自适应融合的场景深度估计

从图像中恢复场景的深度是计算机视觉领域中的一个关键问题。考虑到单一类型图像在深度估计中受场景不同光照的限制,提出了基于红外和可见光图像逐级自适应融合的场景深度估计方法（PF-CNN）。该方法包括双流滤波器部分耦合网络、自适应多模态特征融合网络以及自适应逐级特征融合网络。在双流卷积中红外和可见光图像的滤波器部分耦合使两者特征得到增强;自适应多模态特征融合网络学习红外和可见光图像的残差特征并将两者自适应加权融合,充分利用两者的互补信息;逐级特征融合网络学习多层融合特征的结合,充分利用不同卷积层的不同特征。实验结果表明:PF-CNN在测试集上获得了较好的效果,将阈值指标提高了5%,明显优于其他方法。     

基于双阈值视觉注意模型的图像关注焦点检测

图像关注焦点（FOA）检测是基于人眼视觉关注模型的图像感兴趣区提取的关键技术。为了更加精确、合理地搜索图像关注焦点,提出一种基于双阈值视觉关注模型的FOA检测算法。算法首先提取图像的亮度、颜色、方向和离散矩变换（DMT）特征,生成各个特征对应的特征图;然后将各特征图合并为一张包含多种特征的显著图;最后根据显著图的灰度直方图建立静态阈值与动态阈值,确定图像关注焦点的位置与数量。实验结果表明：新算法在图像关注焦点检测的准确性上较Itti模型有更为优秀的表现,更符合人眼视觉的关注习惯。     

A novel approach to design intraocular lenses with extended depth of focus in a pseudophakic eye model

Eye model is firstly used to design and assess the performance of intraocular lenses (IOLs) with extended depth of focus (DOF), including aspherical IOL, refractive multifocal IOL and diffractive multifocal IOL. The details of design and optimization are given, and the optical performance of the pseudophakic eye with the designed IOLs is assessed with the spot diagram and the visual acuity. For the pseudophakic eye with 3 mm pupil, when the spherical aberration is fully corrected by the aspherical IOL, the best visual acuity reaches 1.2 with a DOF of only 1.4D. Whereas when the spherical aberration is 0.4λ, the best visual acuity is 0.9 with a DOF as much as 2.2D. With the implantation of refractive or diffractive multifocal IOL, the pseudophakic eye has fairly good distant and near vision, while the intermediate vision is worse. Diffractive multifocal IOL diverts 81% of the input light to two primary focuses equally, with the additional 19% of the light wasted as higher order diffraction. Refractive multifocal IOL diverts all the light to two focuses but the light distribution varies with the pupil diameter.     

基于注意力机制与图卷积神经网络的单目红外图像深度估计

对场景中的物体进行深度估计是无人驾驶领域中的关键问题,红外图像有利于在光线不佳的情况下解决深度估计问题.针对红外图像纹理不清晰与边缘信息不丰富的特点,提出了将注意力机制与图卷积神经网络相结合来解决单目红外图像深度估计问题.首先,在深度估计问题中,图像中每个像素点的深度信息不仅与其周围像素点的深度信息相关,还需考虑更大范...     

抵消Wigner-Ville分布交叉项的窄带运动声源无源测速

利用声波的多普勒频移可以对窄带运动声源进行单传感器无源测速,其性能很大程度上取决于能否精确地估计出声波的瞬时频率.Wigner-Ville分布虽然时频分辨率高,但存在交叉项干扰,很少被直接用于瞬时频率估计。对此,提出了抵消Wigner-Ville分布交叉项的单传感器窄带声源无源测速方法。利用交叉项与声源速度的关系构造一个抵消项,引入到Wigner-Ville分布中,通过对声源速度估计值进行迭代更新,使抵消项与交叉项相位相反,从而约掉交叉项。经实测噪声数据验证,对一辆以6.07 m/s匀速运动的卡车(信噪比约为29 dB)测速误差为0.1 m/s,运行时间为4.6 s,对一架以28.90 m/s匀速运动的直升机(信噪比约为16 dB)测速误差为0.46 m/s,运行时间为1.2 s,均优于匹配Wigner变换和多普勒线性调频小波变换测速方法.     

Photoetching of spherical microlenses on glasses using a femtosecond laser

We fabricated spherical microlenses on optical glasses by femtosecond laser direct writing (FLDW) in ambient air. To achieve good appearances of the microlenses, a meridian-arcs scanning method was used after a selective multilayer removal process with spiral scanning paths. A positive spherical microlens with diameter of 48 μm and height of 13.2 μm was fabricated on the surface of the glass substrate. The optical performances of the microlens were also tested. Compared to the conventional laser direct writing (LDW) technique, this work could provide an effective method for precise shape-controlled fabrication of three-dimensional (3D) microstructures with curved surfaces on difficult-to-cut materials for practical applications.     

Effect of binary phase plate on incident radially polarized hollow Gaussian beam

Effect of a pure binary phase filter on the three-dimensional light intensity distributions in focal region is theoretically investigated in this paper. The results show that the proposed binary phase filter may induce flat top profile with large depth of focus and reduced focal spot by properly adjusting the geometrical parameters of the binary phase filter of incident radially polarized hollow Gaussian beam (HGB) beam. On the other hand, the HGB beam benefits the most from the use of an annulus. Such kind of system is potentially useful for lithography, imaging, optical data storage, optical trapping, optical excitation of molecules, or coupling to optical fibers.     

百万像素照相手机CMOS模组调焦方法的探讨

简要介绍了CMOS模组调焦的基本原理。在分析传统调焦方法的基础上,提出了用双光源标板进行调焦的新方法,同时测量CMOS中心像场近距离和无穷远处的MTF,调节镜头和CMOS的相对位置使两处的MTF相等时的位置为最佳对焦位置,实验结果证明该方法调焦范围分布窄、调焦精准、检查位置设置合理。对百万像素照相手机CMOS模组调焦、检测有较大的实用价值。