光纤传像束断丝率的自动检测

介绍了光纤传像束在试验鉴定中断丝率指标的自动检测方法.将一束均匀的平行白光注入光纤传像束输入端,在输出端成像于高分辨率数码相机的CCD面阵上,采集的检测图像传输到计算机中.利用图像预处理和模糊平均方法,通过自适应阈值比较,标识出光纤传像束断丝像点的大体位置,然后以最短欧氏距离为判据,采用系统聚类法对断丝像点进行聚类和识别,统计出断丝数和断丝率.此图像处理算法可以克服传像束胶缝暗纹的干扰,具有计算量小,可靠性高的优点,能够快速地对光纤传像束进行质量评估和综合鉴定.     

提高光纤成像系统传像质量的研究方法

针对提高超长异型光纤传像束系统的传像质量,提出一种研究方法,综合分析了光纤的排列、光纤填充系数、数值孔径以及后续图像合成算法.制造光纤时光纤束采用六角形的排列.使用折射率小的包层.在图像数据处理时,使用抽样定理、三次样条图像信号重建和亚像元配准等图像处理算法,最终提高了光纤成像系统的传像质量.实验证明,通过信号的重建以及错位的校正等方法,能够有效地改善成像光纤束的传像质量.     

光纤耦合的亚像元超分辨率扫描成像图像处理

遥感成像时,采用六角形排列叠层结构的光纤束耦合图像,微扫描工作方式使目标图像依次通过各层光纤。六角形排列的光纤束邻近层光纤之间的错位属亚像元量级,采用亚像元图像处理能获取超分辨率图像。亚像元处理时,首先分离邻近层光纤耦合的图像,以其中一幅图像为基准对另一图像进行配准处理得到中间图像,然后对中间图像和基准图像进行傅里叶变换。考虑仪器及人眼分辨极限,设定图像信号为带限信号,将光纤束邻近层半个像元的错位转化为两幅图像频谱的相位差,从而确定图像融合的插值函数。实验结果表明,针对光纤束六角形排列的亚像元处理获取了超高空间分辨率的图像。     

二维综合孔径辐射计阵元位置误差分析与校正

在二维Y型天线阵组成的综合孔径辐射计系统,阵元存在位置误差引起接收信号相位误差,从而导致反演图像的图像质量下降。为了校正位置误差带来的图像误差,本文在天线阵元存在位置误差的情况下,对反演图像质量的影响进行定量分析。6个辅助源中3个位于方位角为0°,估计阵元的x坐标;3个位于方位角为90°,估计阵元的y坐标。在得到较精确的阵元位置估计基础上,对存在误差的反演图像进行图像校正。     

基于双目视觉的医疗机器人摆位系统测量方法

为了实现精确放疗过程中精确定位和肿瘤位置的精确测量,建立了基于双目视觉的医疗机器人摆位测量系统,并对系统所采用的摄像机标定方法、标记点识别及其三维坐标计算、摆位系统的位置验证等算法进行研究。建立基于双目视觉的医疗机器人摆位测量系统;提出了一种对摄像机采用基于平面棋盘格和立体标定模板相结合的摄像机标定新方法;采用Roberts梯度算子对图像分割的方法,识别标记点中心并计算其三维坐标;通过比较基于双目视觉计算和三维坐标测量仪器测量的各个标记点三维坐标的摆位误差,实现放疗过程中位置验证和精确摆位。实验结果表明:摄像机的标定精度为36.5×10-3 mm和各个标记点的三维坐标平均偏差为δX=0.573mm、δY=0.495mm、δZ=0.430mm,测量方法可获得较高的标定精度和摆位精度,能满足精确放疗对高精度摆位系统的临床需求。     

基于GA-LM融合的星敏感器参数标定算法

针对传统标定算法精度存在受到初始值影响较大的问题,提出一种基于GA-LM融合的参数标定算法。首先分析了星敏感器系统误差对星点坐标位置的影响,确定各标定参数。接着充分利用了GA算法的全局搜索以及LM算法的局部寻优能力,提出一种融合算法解决了最小二乘算法依赖初值、易陷入局部收敛以及GA算法标定效率低、标定结果差的问题,通过半实物仿真系统实验表明,相比于优化前精度提高了88.099%,相比于传统最小二乘法的精度提高了68.015%,体现出了融合算法较大的优越性。     

基于相机标定的非对称裁剪检测算法

该文提出一种针对图像非对称裁剪的取证方法,算法通过相机标定的方法来估计主点位置,并根据主点的位置对图像中存在的非对称裁剪进行指证。该文根据图像取证需求改进了相机标定模型,通过选取不共面的规则图形作为标定物,实现了在不用对标定物进行测量和建模的前提下通过单幅图像对主点坐标进行估计。实验表明,该文提出的算法对非对称裁剪篡改具有较好的检测效果。     

基于感知功能和变步长的蚁群图像边缘检测

针对传统蚁群算法对图像边缘检测需要重复计算、步长单一问题,提出了一种具有感知功能和变步长的蚁群图像边缘检测算法。让觅食蚁具有感知功能和灵活的步长,这样,可以避免大量的重复计算同时让觅食蚁快速搜索到图像边缘。实验仿真结果表明,该算法可以有效地检测出图像边缘,运行时间明显缩短。     

图像目标质心快速搜索算法

提出一种利用x轴、y轴对称平方距离模板并行搜索跟踪二值图像目标质心的新算法,并针对该方法提出了快速实现法.在导弹跟踪目标过程中,当目标发生平移、旋转、缩放时,利用目标的差分内积性质,可快速计算出目标质心位置.仿真结果表明,新算法相比改进前的算法,质心误差始终控制在1.0个像素内,搜索速度有很大提高,而新算法的快速实现更是极大的提高了质心跟踪速度.     

基于红外多目标图像序列的自动判读技术

提出了一种红外多目标图像序列的自动判读方法,先着重阐述一种目标序列段自动变步长搜索方法精确定位有用目标序列段,然后对图像二值化分割批处理方法以及判读脱靶量计算等关键性技术进行说明,最后输出准实时脱靶量;探讨目标搜索中步长和信号量阈值对搜索时长以及精度的影响.实验结果表明,对于低SNR的红外弱小多目标图像序列,该判读方法能缩短判读时间,提高判读的准确性.     

线列光纤传像束的对比度调制传递函数

根据对比度传递函数的定义, 推导出线列光纤传像束对比度调制传递函数的表达式, 采用数值计算的方式, 仿真分析了线列光纤传像束的对比度调制传递函数。仿真结果表明, 输入信号与光纤束位相匹配对对比度调制传递函数结果有一定的影响, 对比度调制传递函数值随位相呈周期性变化。在输入的空间频率小于Nyquist 频率时, 对比度调制传递函数值随频率的增大而减小。     

High Power Fiber Bundle Array Coupled LDA Module

1 Introduction High power laser diode arrays (LDA) have many advan- tages such as small volume, long working life, high slope efficiency and high optical density, so they have many applications in medical treatment, material pro- cessing, and for the pumping source of solid laser and etc. But unfortunately, the LDA can not be easy to use directly in these fields because of their poor beam quality and extremely asymmetric divergent beams (!x≈ 5°～10°and !y≈20°～35°, for example), so it …     

采用多准直光束测量六自由度微位移新方法

为实现星载三线阵CCD相机的相机参数在轨测量,提出了一种六自由度微位移测量方法。高亮度LED输出光被准直并耦合到输入光纤。输出光纤末端固定在可移动的被测物体上。光纤输出由多个光纤准直器（ 4）准直,并由系统固定部分中的多个区域阵列的CCD相机（ 4）捕获。根据CCD图像中光点的位置变化来求解被测物体的六自由度位移。为了验证系统模型和六自由度位移计算程序,对该方法进行了理论分析和仿真。结果表明:当平移位移小于100 m且旋转位移小于6'时,典型的4准直测量系统求解误差小于10-5 m和10-4'。并且,当准直器的光斑位置的两个坐标方向上添加-0.5~0.5 m的随机量时,平移误差和旋转误差的3分别为0.9 m和0.012'。     

高功率光纤激光信号耦合器

实验研究了由三根信号光纤锥形束和一根多模光纤组成的3×1信号耦合器.通过仿真,发现锥形长度越长传输效率和输出光束质量越好,同时也验证了对于低阶模场的吸收要高于高阶模场.实验中,制作了锥区长度为10 mm的3×1信号耦合器,在单纤注入信号功率分别为258 W和365 W的情况下获得转换效率为96.6%的信号输出,总输出功率602W,光束质量为Mx2 = 10.5, My2 = 9.7.     

New biconically tapered fiber star coupler fabricated by indirect heating method

A new biconically tapered fiber star coupler is developed. The fiber bundle is inserted in a quartz protection tube and the tube is heated with an oxyhydrogen burner while the fibers are twisted and pulled. The tube ends are then sealed with adhesive agent. Excess loss for the new coupler is approximatelylog_{10}mfor input/output fiber port numberm. Furthermore, the coupling ratio is almost independent of wavelength and polarization of incident light. The new fabrication method offers high reproducibility and stability.     

High-power fiber laser combination technology

Research on fiber laser combination is discussed briefly in this paper.High-power double clad-fiber laser beam combination technology is introduced and different kinds of fiber laser beam combination and coherent combination technologies are evaluated.Tapered fused bundle (TFB) couplers are used in laser combine for higher power.In this paper, the theory and progress in TFB coupling are introduced.The experiment on our selffabricated TFB is presented.The efficiency of the fiber coupler exceeded 70% and increased as the input power went up.A maximum total output power of 689 W was obtained, with an efficiency of 74%.The fiber coupler displayed stability during the course of the experiment,without any cooling provided.     

用于光学向量矩阵乘法器的光源阵列系统

对用于光学向量矩阵乘法器的16路阵列光源和驱动电路进行了系统的研究。提出了采用商用分布式反馈激光器（DFB）和16路光纤阵列耦合的方式构造阵列光源模块,并且在驱动电路中设计了功率反馈自校正调节算法, 解决了由于当前激光器制造工艺条件的限制而造成的各路激光器之间的阈值电流和P-I转换效率等参数差异问题。实现了光源系统各个通道间的输入向量数据和输出光强之间的一致性映射。实验结果证明,所研究开发的光源阵列不仅成本低,而且数据源信号的高频响应性能良好,保证了光学向量矩阵乘法器运算性能的稳定性和准确性。     

1×N fiber bundle scanning switch

Conventional optomechanical 1×N fiber-optic switches that use individually lensed fibers do not scale gracefully to N>100. We present a 1×N switch where the input and all output fibers are held in a fiber bundle and share a common imaging lens. Output is switched by a reflective scanner that uses prism rotation to achieve 0.1-μm spatial resolution. The fiber locations in a disordered bundle are mapped during a setup procedure and stored in a lookup table for operation. We demonstrate a 1×160 switch in an 11"×3"×4.25" optomechanical package with 20 ms response, 3.5 dB maximum insertion loss, and ±0.02 dB switched connection stability. We discuss how this prototype can be developed into a compact and inexpensive 1×4096 switch     

Liquid-crystal polarization stabilizers on fiber arrays

We have built polarization stabilizer arrays of ultrathin liquid-crystal (LC) polarization controller (PC) arrays and a tap-type photodetector array on a fiber array. The ultrathin (35-/spl mu/m-thick) LC cells are inserted into trenches cut across the fiber array, which is affixed to V grooves in a glass plate and are used to change the phase of 1.55-/spl mu/m lights from 0.5 to 3.9/spl pi/ with the application of less than 3 Vrms. Two cascaded ultrathin LC cells, where the directions of alignment of the LCs are rotated by 45/spl deg/ relative to each other, are capable of converting light with an arbitrary input polarization to the transverse electric (TE) or transverse magnetic (TM) polarization. We fabricated a multichannel TE- and TM-polarization stabilizer of the previously mentioned 16-channel LC PC array and a feedback system made up of a multichannel tap-type photodetector array, sheet polarizer, and a computer and analog-to-digital converters. Feedback control was applied to maximize and minimize the tapped light, to convert an arbitrary input polarization to TM or TE output polarization, and stabilize it in the given state.     

Pressure sensing recognition of FBG array based on random forest algorithm

In order to improve the precision of static load pressure recognition and identify the position of the applied force accurately, a fiber Bragg grating (FBG) flexible sensor array is proposed in this work. Numerical analysis for the package thickness (4 mm) and package position (2 mm from the bottom) of the FBG flexible sensor is performed using COMSOL, and optimal package thickness (4 mm) and package position (2 mm from the bottom) are selected in the analysis. By using 12-FBGs layout method and random forest algorithm, the position and load prediction model is established. The results show that the average error of the distance between the prediction points of coordinates X-Y and static load F and the real sample points is 0.092. Finally, to verify the proposed models, the pressure sensing experiments of the flexible FBG array are carried out on this basis. The weights of 100 g to 1 000 g are applied to different regions of the flexible sensor array one by one in accordance with a certain trajectory. The variation of each FBG wavelength was taken as the input of the stochastic forest prediction model, and the coordinate position and the static load size F were taken as the output to establish the prediction model. The minimum distance error between the actual point and the predicted point was calculated by experiment as 0.03491. The maximum is 0.2481, and the mean error is 0.1515. It is concluded that the random forest prediction model has a good prediction effect on the pressure sensing of the flexible FBG sensing array.