激光投影成像式运动目标位姿测量与误差分析

为了实现室内运动目标位姿的高精度测量,建立了一套激光投影成像式位姿测量系统.该系统利用两两共线且交叉排列在同一平面上的点激光投射器作为合作目标捷联在运动目标上,通过与光斑接收幕墙的配合共同组成运动目标位姿测量基线放大系统,利用高速摄像机实时记录幕墙上投影光斑的位置,利用摄像机标定结果求解投影光斑的世界坐标,利用投影光斑之间构成的单位向量建立运动目标位姿解算模型.最后,根据测量原理推导了图像坐标提取、摄像机外部参数标定、光束直线度与目标位姿解算结果之间的误差传递函数.实验结果表明,当摄像机的视场范围为14 000mm×7 000mm时,测量系统的姿态角测量精度为1′(1δ),位置测量精度为5mm,且误差大小与目标位姿测量误差传递函数理论计算值一致,验证了本文提出的目标位姿测量方法与测量误差传递模型的准确性,能够满足目标位姿测量高精度的要求.     

Robust geometric,phase and colour structured light projection system calibration

This paper introduces a new comprehensive procedure for both geometric and colour calibration of structured light system. In order to perform both geometric and colour calibration procedure, a new calibration artifact is proposed. The intrinsic and extrinsic parameters of projector and camera are estimated by using an extended pinhole camera model with a tangential and radial distortion. Camera image plane coordinates are obtained by extracting features from images of a calibration artifact. Projector image plane coordinates are calculated on the basis of continuous phase maps obtained from a fringe pattern phase reconstruction procedure. In order to stereo calibrate camera-projector system, pairs of corresponding image plane points are calculated with subpixel accuracy. In addition, one of three pattern views is used in colour calibration. RGB values of a colour field pattern detected by camera and their reference values are compared. This comparison leads to derivation of a colour transformation matrix. The performance of the proposed method is tested by measuring plane, sphere and distance reference. Also 360 degrees complex object 3D model from a set of measurements is obtained. Residual mean errors for all tests performed are calculated.     

基于计算机视觉的大射电望远镜馈源舱动态定位系统

在FAST项目馈源舱扫描跟踪设计中 ,舱体的动态定位是实现系统闭环控制的基础。作者根据相机交汇测量原理提出了舱体空间动态定位的基本方法。推导了通过舱体上不共线三点描述舱体位姿的方法 ,介绍了相机交汇测量原理及系统标定方法 ,最后给出了大射电望远镜FAST 5 0m模型计算机视觉动态定位系统的组成 ,并讨论了CCD技术在FAST项目中应用的可行性     

Accurate ball tracking in volleyball actions to support referees

This paper proposes a method for offline accurate ball tracking for short volleyball actions in sport halls. Our aim is to detect block touches on the ball and to determinate accurate trajectory and impact positions of the ball to support referees. The proposed method is divided into two stages, namely training and ball tracking, and is based on background subtraction. Application of the Gaussian mixture model has been used to estimate a background, and a high-speed camera with a capture rate of 180 frames per second and a resolution of 1920?×?1080 are used for motion capture. In sport halls significant differences in light intensity occur between each sequence frame. To minimize the influence of these light changes, an additional model is created and template matching is used for accurate determination of ball positions when the ball contour in the foreground image is distorted. We show that this algorithm is more accurate than other methods used in similar systems. Our light intensity change model eliminates almost all pixels added to images of moving objects owing to sudden changes in intensity. The average accuracy achieved in the validation process is of 0.57 pixel. Our algorithm accurately determined 99.8% of all ball positions from 2000 test frames, with 25.4?ms being the average time for a single frame analysis. The algorithm presented in this paper is the first stage of referee support using a system of many cameras and 3D trajectories.     

基于傅里叶条纹分析的多摄像机标定方法

提出一种新的摄像机标定方法,该方法基于2D共面参照物摄像机标定方法和傅里叶条纹分析方法.将已知相位分布的平面二维正弦灰度调制条纹图作为平面标定靶,通过傅里叶条纹分析方法计算出两个截断正交相位分布,利用截断正交相位分布并结合二维正弦条纹图特点提取相应的图像特征点,建立像素坐标与2D平面坐标的对应关系.将该二维平面靶在摄像机成像空间中放置不同的位置,并完成相应的特征点提取,根据2D共面参照物摄像机标定方法即可完成摄像机标定.该方法利用平面相位测量的高准确度来提高标定特征点的提取准确度,从而提高标定准确度.实验对双摄像机系统进行了标定,标定后该系统对标定靶进行测量,标准偏差达到0.010 mm.     

Research on camera calibration using new optimization strategy

An important task for stereo vision is camera calibration, whose goal is to obtain the intrinsic and extrinsic parameters of each camera. This paper proposes a new accurate calibration method with multilevel process of camera parameters. In order to improve the calibration accuracy, a sub-pixel corner detection method is presented. We start with several views of a planar calibration to obtain some intrinsic camera parameters and to build an accurate model with lens distortion on a planar calibration target. Flexibly making use of geometry imaging theory, our algorithm obtains all the parameters through logical organization of solving order, accordingly avoids obtaining possible local optimized problem when solving the non-linear equation, gets over the relativity influence of every unknown parameters of traditional calibration way, and makes the error distributed among the constraint relation of parameters. Experiments with real images are carried out to verify the image correction effect and numerical robustness of our results. Compared with classical calibration techniques, that use expensive equipment and complicated mathematical computation, the proposed technique, which was verified by experiment, achieves high accuracy and reliable parameters.     

Calibration for stereo vision system based on phase matching and bundle adjustment algorithm

Calibration for stereo vision system plays an important role in the field of machine vision applications. The existing accurate calibration methods are usually carried out by capturing a high-accuracy calibration target with the same size as the measurement view. In in-situ 3D measurement and in large field of view measurement, the extrinsic parameters of the system usually need to be calibrated in real-time. Furthermore, the large high-accuracy calibration target in the field is a big challenge for manufacturing. Therefore, an accurate and rapid calibration method in the in-situ measurement is needed. In this paper, a novel calibration method for stereo vision system is proposed based on phase-based matching method and the bundle adjustment algorithm. As the camera is usually mechanically locked once adjusted appropriately after calibrated in lab, the intrinsic parameters are usually stable. We emphasize on the extrinsic parameters calibration in the measurement field. Firstly, the matching method based on heterodyne multi-frequency phase-shifting technique is applied to find thousands of pairs of corresponding points between images of two cameras. The large amount of pairs of corresponding points can help improve the accuracy of the calibration. Then the method of bundle adjustment in photogrammetry is used to optimize the extrinsic parameters and the 3D coordinates of the measured objects. Finally, the quantity traceability is carried out to transform the optimized extrinsic parameters from the 3D metric coordinate system into Euclid coordinate system to obtain the ultimate optimal extrinsic parameters. Experiment results show that the procedure of calibration takes less than 3 s. And, based on the stereo vision system calibrated by the proposed method, the measurement RMS (Root Mean Square) error can reach 0.025 mm when measuring the calibrated gauge with nominal length of 999.576 mm.     

Constructing feature points for calibrating a structured light vision sensor by viewing a plane from unknown orientations

To calibrate a structured light vision sensor, it is necessary to obtain at least four non-collinear feature points that fall on the light stripe plane. We propose a novel method to construct non-collinear feature points used for calibrating a structured light vision sensor with a planar calibration object. After the planar calibration object is moved freely in the range of measuring of the structured light vision sensor at least twice, all the local world coordinates of the feature points falling on the light stripe plane can be readily obtained in site. The global world coordinates of the non-collinear feature points in the local world coordinate frame can be computed through the three-dimensional (3D) camera coordinate frame. A planar calibration object is designed according to the proposed approach to provide accurate feature points. The experiments conducted on a real structured light vision sensor that consists of a camera and a single-light-stripe-plane laser projector reveal that the proposed approach has high accuracy and is practical in the vision inspection applications. The proposed approach greatly reduces the cost of the calibration equipment and simplifies the calibrating procedure. It advances structured light vision inspection one step from laboratory environments to real world use.     

基于两个正交一维物体的单幅图像相机标定

提出了一种利用两个正交一维物体构成"T"型靶标进行摄像机标定的新方法。该方法只需对"T"型靶标上已知坐标的5点投影一幅图像,然后根据柔性靶标原理计算出由虚点和标记点组成的共直线的4点,由射影变换同素性、接合性以及交比不变性标定出镜头的一阶径向畸变参数。利用已知畸变参数对图像进行畸变校正,然后由基于两个正交一维物体坐标变换的方法即可标定出相机的内外参数。该方法线性求解镜头畸变参数,避免了传统方法非线性迭代优化过程中产生的参数耦合现象。实验表明,不进行镜头畸变校正则相机标定精度随着图像噪声的增加呈不稳定状态;进行畸变校正后对简单标定计算的初始值进行优化得到稳定的高精度标定结果。整个实验设备简单,操作方便,只需一幅图像即可实现镜头畸变和相机内外参数的标定,可以达到实时的效果。     

Self-calibration of multi-camera networks without feature correspondence between different cameras

An automatic calibration approach for multi-camera networks is proposed to calibrate the intrinsic parameters of each camera and the extrinsic parameters between different cameras. Firstly, the moving objects are tracking, and the feature points are detected and calculated by a matching method from image sequences. And then we estimate the intrinsic parameters of each camera respectively by a self-calibration method from the motion of feature points, while estimating the rotation and translation of each camera with respect to the object. Thirdly, we estimate the extrinsic parameters between different cameras from the rotation and the motion of each camera with respect to the object. Our method only needs to track the motion of objects in each camera without the correspondence between different cameras. It avoids the difficulty of the correspondence between different cameras in real networks. Experiments with simulated data and real images are carried out to verify the theoretical correctness and numerical robustness.     

未标定相机的高精度位姿估计方法

针对未标定相机的位姿估计问题,提出了一种焦距和位姿同时迭代的高精度位姿估计算法。现有的未标定相机的位姿估计算法是焦距和相机位姿单独求解,焦距估计精度较差。提出的算法首先通过现有算法得到相机焦距和位姿的初始参数;然后在正交迭代的基础上推导了焦距和位姿最小化函数,将焦距和位姿同时作为初始值进行迭代计算;最后得到高精度的焦距和位姿参数。仿真实验表明提出的算法在点数为10,噪声标准差为2的情况下,角度相对误差小于1%,平移相对误差小于4%,焦距相对误差小于3%;真实实验表明提出的算法与棋盘标定方法的精度相当。与现有算法相比,能够对未标定相机进行高精度的焦距和位姿估计。     

Camera calibration with planar crossed fringe patterns

We introduced a novel crossed fringe pattern as the model plane for camera calibration. It can provide abundant control points, in extreme case each pixel on the CCD sensor can be taken as control points. Fourier Fringe Analysis is used to extract the phase distributions from the image of this pattern. Each control point consists of the locations of a pixel and the world coordinates, which can be calculated from the phase distributions. It avoids the complex extracting procedure of those classical control points and could produce abundant control points. Both computer simulation and real data have confirmed the proposed technique is easy to use, reliable and high accurate.     

Integrated calibration of multiview phase-measuring profilometry

Phase-measuring profilometry (PMP) measures per-pixel height information of a surface with high accuracy. Height information captured by a camera in PMP relies on its screen coordinates. Therefore, a PMP measurement from a view cannot be integrated directly to other measurements from different views due to the intrinsic difference of the screen coordinates. In order to integrate multiple PMP scans, an auxiliary calibration of each camera’s intrinsic and extrinsic properties is required, in addition to principal PMP calibration. This is cumbersome and often requires physical constraints in the system setup, and multiview PMP is consequently rarely practiced. In this work, we present a novel multiview PMP method that yields three-dimensional global coordinates directly so that three-dimensional measurements can be integrated easily. Our PMP calibration parameterizes intrinsic and extrinsic properties of the configuration of both a camera and a projector simultaneously. It also does not require any geometric constraints on the setup. In addition, we propose a novel calibration target that can remain static without requiring any mechanical operation while conducting multiview calibrations, whereas existing calibration methods require manually changing the target’s position and orientation. Our results validate the accuracy of measurements and demonstrate the advantages on our multiview PMP.     

A novel planar calibration method using the iterative correction of control points information

In the camera calibration using translational planar object instead of 3D target, the skew affected by the imprecision in installation causes the object to deviate from the designated position, which produces errors to the world coordinates of control points and reduces the calibration accuracy. In this paper, a skew correction model is established to correct this skew. We note and prove that the skew bias of control point which is close to the fix point is tiny. According to the projection regulation, the pixel distance between two adjacent control points which is close to the fix point in skewed situation is a value extremely approximate to that in the ideal situation. Based on this property, we utilize the pixel distance to assign the image coordinates based on the fix point. The assigned values are then employed to estimate the skew factors, by which the initial camera parameters are optimized and lens distortion in the calibration images are corrected. This process is then repeated until convergence. Experiments based on real images prove that this method is more accurate than other methods without correcting skew. The parameters obtained by our method can be applied to the 3D reconstruction directly and effectively.     

Real and Pseudo Pedestrian Detection Method with CA-YOLOv5s Based on Stereo Image Fusion

With the development of convolutional neural networks, the effect of pedestrian detection has been greatly improved by deep learning models. However, the presence of pseudo pedestrians will lead to accuracy reduction in pedestrian detection. To solve the problem that the existing pedestrian detection algorithms cannot distinguish pseudo pedestrians from real pedestrians, a real and pseudo pedestrian detection method with CA-YOLOv5s based on stereo image fusion is proposed in this paper. Firstly, the two-view images of the pedestrian are captured by a binocular stereo camera. Then, a proposed CA-YOLOv5s pedestrian detection algorithm is used for the left-view and right-view images, respectively, to detect the respective pedestrian regions. Afterwards, the detected left-view and right-view pedestrian regions are matched to obtain the feature point set, and the 3D spatial coordinates of the feature point set are calculated with Zhengyou Zhang’s calibration method. Finally, the RANSAC plane-fitting algorithm is adopted to extract the 3D features of the feature point set, and the real and pseudo pedestrian detection is achieved by the trained SVM. The proposed real and pseudo pedestrian detection method with CA-YOLOv5s based on stereo image fusion effectively solves the pseudo pedestrian detection problem and efficiently improves the accuracy. Experimental results also show that for the dataset with real and pseudo pedestrians, the proposed method significantly outperforms other existing pedestrian detection algorithms in terms of accuracy and precision.     

基于三线结构光的相机平面标定方法研究

相机标定技术是结构光三维视觉测量的关键技术之一,结构光系统的相机标定的精度对三维测量的精度有很大影响。首先对三线结构光系统图的相机标定方法进行了分析,简单介绍了工业相机成像的几何模型及标定的原理;其次利用Harris角点检测方法提取特征点坐标,并选用了BP神经网络来校正工业相机的畸变模型,以提高标定算法的优化速度和标定精度;最后采用张正友的平面标定法对校正后的摄像机模型进行标定实验,由实验结果知,该方法具有一定的准确性和有效性,在一定误差范围内,基于神经网络畸变校正的张正友相机标定能够有效提高视觉检测的精度。     

Extended gravitational pose estimation

The model-to-image registration problem is a problem of determining the position and orientation (the pose) of a three-dimensional object with respect to a camera coordinate system. When there is no additional information available to constrain the pose of the object and to constrain the correspondence of object features to image features, the problem is also known as simultaneous pose and correspondence problem, or correspondenceless pose estimation problem. In this paper, we present a new algorithm, called extended gravitational pose estimation (EGPE), for determining the pose and correspondence simultaneously. The algorithm is based on gravitational pose estimation (GPE) algorithm. In our algorithm, the original GPE has been revised to deal with the problem with false image points. For problems with both occluded object points and false image points, we firstly applied single-link agglomerative clustering algorithm to pick out occluded object points when a local minimum has been found, then the revised GPE is applied again on the clustering result to update rotation and translation of the object model. EGPE has been verified on both synthetic images and real images. Empirical results show that EGPE is faster, more stable and reliable than most current algorithms, and can be used in real applications.     

用于相机标定的球靶标投影误差分析与校正

由于球体具有轮廓连续性好等优点,在摄像机标定,尤其是多相机标定方面获得了广泛的应用。利用球作为标定靶标可以弥补平面靶标在多相机标定中出现视角过大时畸变太大甚至于观测不到的不足,但是空间球经透视投影后成像一般并非标准圆,而是一个椭圆。椭圆几何中心与球心真实成像中心并不一致,从而影响了标定精度。造成球心成像误差的因素主要有两个,即球的相对大小及相对于相机的位置。通过分析空间球成像模型,仿真研究了各因素对球心成像误差影响的大小,寻找球心的透视投影像点与其成像椭圆几何中心之间的误差变化规律,并建立了两者之间的误差校正模型,最后通过实验验证了该校正模型的可行性和有效性。通过校正,球心投影像点定位精度可达到亚像素级。     

结构光三维成像系统的计算机仿真

基于相位测量轮廓术和摄像机模型,提出一种结构照明三维成像系统的高精度计算机仿真算法。对于给定的物体三维模型,首先根据系统结构,采用Z缓冲技术消除遮挡和阴影部分,得到与摄像机像素点对应的物体表面采样点三维坐标,再使用统一的数学模型和方法处理投影过程。根据摄像机及投影仪的内、外参数,最终得到了摄像机像素点、物体表面采样点和投影仪像素坐标三者之间的对应关系,从而实现了结构照明三维成像系统仿真。为实际系统的结构设计、调整和参数校正提供了参考。     

中心折反射全方位摄像头的标定方法研究

提出一种适用于不同类型中心折反射全方位摄像头的标定方法。该方法不需标定模板的运动信息或限定全方位摄像头的类型,而只用通过摄像头观测分布在不同位置的标定模板,对所采集的标定模板图像提取角点,将包含内外部参数的非线性图像投影函数用泰勒级数展开表示,应用Levenberg-Marquardt算法计算摄像头外部参数,而通过对超定方程组的伪逆求解计算内部参数。整个标定过程简单,快速而且自动完成。利用某一中心折反射全方位摄像头对提出的方法进行了实验,实验结果表明,该方法可获得很好的标定效果。