电路噪声对星敏感器星点定位精度的影响

提出了一种测量星敏感器电路模块带来的图像噪声的方法。在实验室条件下,用星敏感器拍摄多幅背景图像,分析表明:不同背景图像上相同像素的灰度差值代表了电路模块带来的图像噪声,该噪声服从正态分布。模拟产生该正态分布的随机矩阵作为图像噪声加到理想的星点上,采用亚像素细分质心算法计算出带噪声的星点质心坐标,从而得到噪声带来的星点位置偏差。计算结果表明:实验室条件下,星敏感器实测6等星的质心测量精度为1/25 Pixel,而电路噪声将引起1/200 Pixel量级的标准偏差;电路噪声,尤其是DC/DC电源纹波噪声对星敏感器星点定位带来的误差是不容忽视的。     

检校场标定长焦航测相机参数

为实现长焦航测相机内参数及畸变参数标定,提出利用室外检校场对相机进行标定的方法。介绍了该方法由初值到精确值的两步法的求解步骤、计算过程及实验过程。首先通过给定已知特征点位置信息以及不同角度不同位置下拍摄的影像,通过特征点提取及匹配,根据像点坐标与世界坐标的线性对应关系,基于直接线性变换算法建立约束方程求取相机参数初值,进而通过混合LMQN(Levenberg-Marquardt与Qusai-Newton)迭代优化加速求解精确值。与精密测角法比无需精密设备的操作及记录,只需不同位置不同姿态拍摄多张包含精确位置信息的检校场影像即可。最后进行标定实验并对结果分析,基于检校场的标定方法在参数解算中特征点最大投影误差为2.471pixel,标定参数主点精度为9.7μm(2pixel),主距精度为4.3μm(1pixel),解算相机拍摄位置精度0.035m,满足测绘应用的精度需求。     

基于粒子滤波的彩色图像跟踪算法的研究

摘 要;在以往的视频跟踪中,目标检测和跟踪常常需要两个计算法则,过程比较复杂,实现耗时较多。为了实现序列彩色图像的实时检测与跟踪,本文以目标有无信息和目标位置信息为变量建立联合状态向量,利用粒子滤波方法实现目标检测及跟踪。此外为了减少计算量并且充分考虑到跟踪区域各像素的权重,建立基于颜色信息的特征直方图作为观测向量,用于后验估计。实验证明本文提出的方法能够在14.37ms内检测并跟踪目标,并且对目标的旋转变化和尺度变化具有一定的鲁棒性     

飞行器仿真测试中合作目标投影光斑的跟踪识别

针对飞行器仿真测试中合作目标投影光斑的跟踪识别问题,提出了一种投影光斑跟踪识别新方法,该方法主要由预测、识别及修正3个阶段组成。预测阶段主要结合投影光斑运动特点对传统卡尔曼滤波进行改进,提高投影光斑的位置预测精度。识别阶段则根据投影光斑位置的预测值,分两种情况对光斑进行处理:如果下一时刻投影光斑在视场内,则根据相应的判别准则和匹配策略在下一时刻图像中快速搜索投影光斑的最优匹配光斑;如果下一时刻投影光斑在视场外,则根据测量系统相关信息,对视场外投影光斑在图像平面上的位置进行求解,并将求解结果加入相应的运动轨迹,实现对视场外投影光斑的跟踪识别。在完成投影光斑的跟踪识别后,根据跟踪识别结果对投影光斑相关参数进行修正。仿真实验和实际实验结果表明,本文方法能够有效跟踪识别飞行器仿真测试中合作目标的投影光斑,其最大跟踪识别误差不超过2.5pixel,即使跟踪识别过程中存在投影光斑进出视场的情况也不受影响。     

基于图像处理的轴类零件尺寸测量技术研究

采用图像检测技术对轴类零件尺寸进行非接触检测,实现了动态测量。在获取零件图像后,采用Sobel算子完成对图像边缘的初步定位,在此基础上,通过对图像边缘的离散点作最小二乘曲线拟合,达到边缘的亚像素定位,基于精度分析,得出了误差最小时的尺寸求解方法。通过标定系数k的求解,计算出精确的零件尺寸。实验表明,该方法能有效地提高检测精度,在CCD摄像机分辨率为1392×1040,像元尺寸为4.65μm时,检测误差<±20μm。     

星点质心亚像元定位的高精度误差补偿法

在CCD星敏感器设计过程中,星点坐标的提取精度不仅影响星图识别的正确性,也直接影响星敏感器的最终姿态输出精度。本文通过分析传统质心法亚像元提取算法误差模型,证明该模型存在系统误差和随机误差。针对系统误差,提出了用最小二乘拟合法来估计质心位置和系统误差大小的对应关系,利用这个关系对系统误差进行补偿;针对随机误差,提出用边缘阈值法和星图整体相消法来消除。本文还分析了CCD成像数学模型,星光成像的能量分布近似服从二维高斯分布,星点高斯分布弥散半径不同,用于误差补偿的曲线方程也不一样,根据不同的高斯弥散半径,设计了误差补偿模板,用于补偿质心法的系统误差。实验表明,采用误差补偿模板,可以将质心法精度从1/20pixel提高到1/200pix-el,满足了星敏感器高精度星点质心提取的要求。     

基于变分贝叶斯的自适应 PF-SLAM 方法研究

针对移动机器人即时定位与地图构建中时变观测噪声及粒子位置分布对SLAM精度的影响,本文提出基于变分贝叶斯的自适应PF-SLAM算法,采用高斯混合模型对时变的观测噪声建模,并通过变分贝叶斯方法,迭代估算出混合模型中的未知参数;同时根据粒子权值将粒子划分为固定粒子和优化粒子,通过粒子间的近邻拓扑位置关系调整粒子分布,处理时变观测噪声与优化粒子的位置分布,使得优化的粒子集可以更好地表示机器人位置概率分布,实现观测噪声及粒子位置分布自适应。仿真实验表明本算法对比传统PF-SLAM算法定位与地图构建误差降低了76.45%。实际实验表明本算法处理下的环境轮廓误差对比传统PF-SLAM算法的环境轮廓误差减小了61.87%。该算法有效提高了移动机器人的状态估计精度,为移动机器人即时定位与地图构建提供了新的参考。     

Subpixel position measurement using 1D, 2D and 3D centroid algorithms with emphasis on applications in confocal microscopy

The accuracy with which centroid algorithms in 1D, 2D and 3D can estimate an object's position has been investigated. Three factors that can influence the method's accuracy have been investigated: systematic error of the algorithm, influence of photon noise and the influence of perturbations such as scanning nonlinearity. The variation of the accuracy with parameters that are relevant for confocal microscopy, such as object diameter and photon noise/pixel, has been considered. Theory and simulations presented show that the variation of the accuracy with respect to such parameters can differ drastically between the 1D, 2D and 3D cases. Experiments performed using microspheres show that the magnitudes of the three types of error can be approximately the same under normal operating conditions and that it is therefore necessary to take all three into account when assessing the total error.     

Errors in particle tracking velocimetry with high-speed cameras

Velocity errors in particle tracking velocimetry (PTV) are studied. When using high-speed video cameras, the velocity error may increase at a high camera frame rate. This increase in velocity error is due to particle-position uncertainty, which is one of the two sources of velocity errors studied here. The other source of error is particle acceleration, which has the opposite trend of diminishing at higher frame rates. Both kinds of errors can propagate into quantities calculated from velocity, such as the kinetic temperature of particles or correlation functions. As demonstrated in a dusty plasma experiment, the kinetic temperature of particles has no unique value when measured using PTV, but depends on the sampling time interval or frame rate. It is also shown that an artifact appears in an autocorrelation function computed from particle positions and velocities, and it becomes more severe when a small sampling-time interval is used. Schemes to reduce these errors are demonstrated.     

Methods for Digital Particle Image Sizing (DPIS): Comparisons and improvements

This paper explores the accuracy of particle image sizing using direct processing of digitally recorded images. Traditional methods for particle image sizing were considered and, four new algorithms were developed to deliver improved accuracy and robustness. Statistical error analysis was performed using Monte Carlo simulations in order to quantify the dependence of these methods on noise, discretization, and particle size distribution. The performance of these methods were compared against Phase Doppler Analyzer measurements of spray atomization.We introduce a novel two-dimensional four-point Gaussian estimator and an alternative Gaussian estimator based on a local least squares (LLS) fit. These methods were further advanced to account for pixel discretization effects using integral formulations (continuous methods). All new methods were compared against conventional pixel counting and the established three-point Gaussian estimator. The new methods significantly reduced the total error in the diameter estimation compared to the three-point Gaussian estimator and pixel counting. The least squares Gaussian estimator and its continuous version demonstrated almost identical results and superior performance for diameters over 4 pixels. For smaller diameters, the continuous four-point Gaussian estimator delivered the highest accuracy. For uniform particle size distribution between 2–14 pixels image diameter, the least squares estimators delivered error less than 5% with respect to the true diameter for 80% of the particles. The remaining methods demonstrated error of 5% (or better) for less than 60% of the particles. Validation in an experiment of high-pressure spray atomization showed that the Gaussian local least squares methods and the continuous four-point method delivered similar particle size distribution compared to PDA. The particle mean diameter estimated by the two methods differed only by 3% and 6% respectively with respect to the PDA measurements.The novel particle image sizing schemes developed here can deliver accurate, robust, and computationally efficient apparent diameter measurement, thus providing a viable, simple and inexpensive solution for performing sizing on conventional particle image velocimetry images. This capability enables simultaneous measurements of both velocity and particle size for a wide range of multi-phase flows.     

基于多峰拟合的直线电机动子位置精密测量方法

为满足直线电机动子位置的高精度测量要求,提出一种基于多峰拟合相位相关算法的直线电机动子位置精密测量方法。通过直线电机动子上的高速相机采集相邻栅栏图像;采用相位相关算法得到相邻栅栏图像的相位相关函数;根据相位相关频谱的峰值分布特性,采用多峰拟合获得相邻栅栏图像的亚像素位移值;根据测量系统标定系数,进而获得直线电机动子位置的精确位置。实验结果表明,该亚像素测量方法用于直线电机动子位置检测能达到精密定位,测量精度为1/100 pixel,且具有较强的鲁棒性和较高的实时性。     

兼顾热轧工艺润滑的工作辊磨损预报模型

通过对某1700热连轧厂润滑轧制中轧辊磨损主要影响因素的研究,提出了包含润滑项的工作辊磨损预报模型,并采用模拟退火遗传算法估算模型主要参数。现场运用结果表明,该工作辊磨损模型结构及其模型参数能够兼顾热轧润滑和非润滑工况,提高了工作辊磨损的预报精度,满足在线运用需求。
     

6UPS并联机构静态误差的矢量法标定

分析了不同误差对并联机构的影响，通过对机构进行位姿误差分析，建立了采用并联机构运动学反解模型的矢量法误差求解模型；研究了常用的机构标定方法及相应的测量工具，使用激光干涉仪完成了误差测量实验，并按照矢量法误差模型对测量结果进行了分析计算，使用多组姿态数据的不同组合进行误差计算，并对误差影响区域及区域重心进行拟合处理，获得了优化的机构静平台、动平台铰链的位置误差数据。使用位置误差数据可以获得更好的补偿效果。     

基于ICT图像与设计模型的制造误差检测方法

使用二维CT图像与设计模型通过比对进行制造误差分析时,必须知道CT图像在设计模型中的位置,而位置的确定往往需要在工件上找到一个和设计模型相匹配的基准位置,并对该位置进行精加工,这增加了工件加工的周期和制造成本.针对该问题,本文提出一种自动搜索CT图像的设计模型来完成工件制造误差检测的方法.该方法首先提取CT图像的边缘,...     

第3代轮毂轴承单元轴向游隙影响因素的正交试验

利用正交试验法分析沟位置差、径向游隙、配合过盈量及锁紧力矩等参数对轮毂轴承单元轴向游隙的影响,给出了沟位置差和径向游隙值的控制范围,可为第3代汽车轮毂轴承单元的装配及零配件工艺的优化提供数据支持。     

Particle positioning techniques for dusty plasma experiments

Video microscopy is a widely applied diagnostic to investigate the structure and the dynamics of particles in dusty plasmas. Reliable algorithms are required to accurately recover particle positions from the camera images. Here, four different particle positioning techniques have been tested on artificial and experimental data of dusty plasma situations. Two methods that rely on pixel-intensity thresholds were found to be strongly affected by pixel-locking errors and by noise. Two other methods-one applying spatial bandpass filters and the other fitting polynomials to the intensity pattern-yield subpixel resolution under various conditions. These two methods have been shown to be ideally suited to recover particle positions even from small-scale fluctuations that are used to derive the normal mode spectra of finite dust clusters.     

Analysis and evaluation of objective functions in kinematic calibration of parallel mechanisms

This paper analyses different objective functions for the parameter identification of parallel mechanisms and studies the influence in the position and orientation errors to improve their accuracy. A new objective function considering deviation terms is presented. This function is compared with other widely used functions and the advantages and disadvantages of each function are presented. The geometric parameter identification is performed by external calibration by means of the direct kinematic model. First, the objective functions are defined considering error position, error orientation and deviations in measurement. These functions compare the measured and calculated moving platform coordinates in order to obtain the identified model parameters that minimize this difference. The measured coordinates are obtained by measuring three sphere centres, fixed to the moving platform of a parallel mechanism, and the computed coordinates are given by the kinematic model. Second, the model is solved by the Levenberg–Marquardt algorithm for a number of identification positions. Finally, the calibration is verified in test positions. The results obtained show that the consideration of the deviations in measurement in the objective function with respect to classical approaches allows us to better identify those kinematic parameters corresponding with passive joints that cannot be measured. These findings confirm that a suitable objective function can improve the mechanism accuracy by more than one order of magnitude in both position and orientation errors.     

共面点的摄像机非线性畸变校正

采用传统的Tsai两步法进行摄像机标定时,标定精度会受一阶径向畸变模型的限制。本文提出了一种同时考虑摄像机镜头径向畸变和切向畸变的摄像机模型并研究了模型求解方法来提高畸变校正精度。考虑图像中心区域畸变较小,故用中心附近点列出线性方程组计算了摄像机的部分参数;建立了综合畸变模型,将摄像机参数代入模型计算畸变参数的初始值。由于焦距和平移分量在标定板与相机平面的距离深度变化不够时难以一次性准确标定,故将其代入综合畸变模型重新计算,并运用两步迭代法逐步逼近精确解。最后,对世界坐标系进行空间几何变换、透视变换和成像变换得到的重投影图像的像素坐标并与实际测得的像素坐标值进行比较,得到校正误差。结果表明,本文的畸变校正方法平均像素误差可以达到0.114 9pixel,优于Tsai校正方法的0.367 0pixel,且重复性较好。     

Position referencing in optical microscopy thanks to sample holders with out-of-focus encoded patterns

This article introduces smart sample holders for optical microscopy. Their purpose is to allow the absolute determination of the position of the observed zone with respect to the sample holder itself and with a high accuracy. It becomes then straightforward to find a given zone of interest by positioning coarsely the microscope slide to the same position coordinates. Furthermore images recorded during different observation sessions; i.e. for slightly different positions; can be processed numerically in order to superimpose them with a high accuracy. Thus the slight deviations of the microscope slide position and orientation due to the different observations are compensated numerically and a perfect superimposition of the recorded images is performed. Then accurate site-by-site image comparisons become possible even for images recorded during different observation sessions and over a long period of time. The subpixel capability of the proposed method is demonstrated and those smart microscope slides constitute a new tool for live cell experiment.
In practise, an encoded geometrical pattern used as position reference is inserted in a plane parallel to the surface receiving the tissue section or sample. Then the transition of the focus position from the tissue section to the position reference requires only a vertical adjustment and does not affect the lateral coordinates of observation. The numeric processing of the image of the position reference pattern allows the retrieval of the lateral coordinates that are also used for the tissue section image. Thus each image is recorded with a set of position coordinates that defines accurately the position of the observed area with respect to the sample holder itself.     

V型焊接坡口中心的高精度提取方法

焊缝跟踪精度对焊缝成形质量有很重要的影响,V型坡口是一种最常见的焊接接头。提出了一种提高激光视觉机器人跟踪V型坡口精度的软件措施,介绍了提取V型坡口中心亚像素级坐标的方法。采用多尺度的LOG(Laplacian of Gaussian)算子检测焊接坡口特征边缘,在分析图像灰度直方图的基础上选取适当的阈值进行二值化。用曲率极值的角点检测法从细化后的图像中提取焊接坡口中心点像素级图像坐标,然后对此点邻域的梯度值进行拉格朗日多项式插值求得亚像素级坐标。实验结果表明,该方法简单可行,能满足机器人焊缝跟踪的实时性要求。