公开(公告)号：US08934721B2

公开(公告)日：2015-01-13

申请号：US13696982

申请日：2011-04-25

Applicant: Guangjun Zhang ,  Zhenzhong Wei ,  Weixian Li ,  Zhipeng Cao ,  Yali Wang

Inventor： Guangjun Zhang ,  Zhenzhong Wei ,  Weixian Li ,  Zhipeng Cao ,  Yali Wang

IPC: G06K9/36 ,  G01B11/00 ,  G02B21/36 ,  G02B21/22

CPC classification number: G01B11/005 ,  G02B21/22 ,  G02B21/365

Abstract: The present disclosure provides a microscopic vision measurement method based on the adaptive positioning of the camera coordinate frame which includes: calibrating parameters of a microscopic stereo vision measurement model (201); acquiring pairs of synchronical images and transmitting the acquired images to a computer through an image acquisition card (202); calculating 3D coordinates of feature points in a scene according to the matched pairs of feature points in the scene obtained from the synchronical images and the calibrated parameters of the microscopic stereo vision measurement model (203); and performing specific measurement according to the 3D coordinates of the feature points in the scene (204). With the method, the nonlinearity of the objective function in the microscopic vision calibration optimization is effectively decreased and a better calibration result is obtained.

Abstract translation: 本公开提供了一种基于相机坐标系的自适应定位的微观视觉测量方法，其包括：校准微观立体视觉测量模型（201）的参数; 获取成对的同步图像并通过图像采集卡（202）将所获取的图像发送到计算机; 根据从同步图像获取的场景中的特征点匹配对和微观立体视觉测量模型（203）的校准参数，计算场景中特征点的3D坐标; 以及根据场景（204）中的特征点的3D坐标进行特定测量。 利用该方法，有效降低了微观视觉校准优化中目标函数的非线性，获得了更好的校准结果。

公开(公告)号：US08019147B2

公开(公告)日：2011-09-13

申请号：US12025741

申请日：2008-02-04

Applicant: Guangjun Zhang ,  Junhua Sun ,  Zhenzhong Wei ,  Qianzhe Liu

Inventor： Guangjun Zhang ,  Junhua Sun ,  Zhenzhong Wei ,  Qianzhe Liu

IPC: G06K9/00

CPC classification number: G06K9/209 ,  G06K2209/40 ,  G06T7/593 ,  G06T7/85 ,  G06T2207/10012

Abstract: The present disclosure is directed to a three-dimensional data registration method for vision measurement in flow style based on a double-sided target. An embodiment of the disclosed method that comprises A. Setting up two digital cameras which can observe the entire measured object; B. Calibrating intrinsic parameters and a transformation between the two digital camera coordinate frames; C. A double-sided target being placed near the measured area of the measured object, the two digital cameras and a vision sensor taking images of at least three non-collinear feature points of the double-sided target; D. Removing the target, measuring the measured area by using the vision sensor; E. Respectively computing the three dimensional coordinates of the feature points in the global coordinate frame and in the vision sensor coordinate frame; F. Estimating the transformation from the vision sensor coordinate frame to the global coordinate frame through the three dimensional coordinates of the three or more non-collinear feature points obtained at step E, then transforming the three dimensional data of the measured area to the global coordinate frame; and G. Repeating step C, D, E, F, then completing three dimensional data registration for all measured areas. The present disclosure improves three dimensional data registration precision and efficiency.

Abstract translation: 本公开涉及一种用于基于双面目标的流动式视觉测量的三维数据登记方法。 所公开的方法的实施例包括A.设置可以观察整个测量对象的两个数码相机; B.校准固有参数和两个数码相机坐标系之间的变换; C.一个双面目标放置在测量对象的测量区域附近，两个数字照相机和视觉传感器拍摄双面目标的至少三个非共线特征点的图像。 D.去除目标，使用视觉传感器测量测量面积; E.分别计算全局坐标系和视觉传感器坐标系中特征点的三维坐标; F.通过在步骤E获得的三个或多个非共线特征点的三维坐标来估计从视觉传感器坐标系到全局坐标系的变换，然后将测量区域的三维数据变换为全局坐标 帧; 和G.重复步骤C，D，E，F，然后完成所有测量区域的三维数据登记。 本公开改进了三维数据配准精度和效率。

公开(公告)号：US07768527B2

公开(公告)日：2010-08-03

申请号：US11561696

申请日：2006-11-20

Applicant: Guangjun Zhang ,  Suqi Li ,  Zhenzhong Wei ,  Zhen Liu

Inventor： Guangjun Zhang ,  Suqi Li ,  Zhenzhong Wei ,  Zhen Liu

IPC: G09G5/00 ,  G03B21/26

CPC classification number: G01B11/2504 ,  G06T7/85 ,  G06T19/006 ,  G06T2207/10012

Abstract: The disclosure relates to a hardware-in-the-loop simulation system and method for computer vision. An embodiment of the disclosed system comprises a software simulation and a hardware simulation. The software simulation includes a virtual scene and an observed object that are generated by virtual reality software. The virtual scene images are obtained at different viewpoints. The hardware simulation includes the virtual scene images being projected onto a screen by a projector, wherein the projected scene images are shot by a camera, and where in the direction of the camera is controlled by a pan-tilt.

Abstract translation: 本公开涉及一种用于计算机视觉的硬件在环仿真系统和方法。 所公开的系统的实施例包括软件仿真和硬件仿真。 软件模拟包括由虚拟现实软件生成的虚拟场景和观察对象。 以不同的观点获得虚拟场景图像。 硬件模拟包括通过投影仪投影到屏幕上的虚拟场景图像，其中投影的场景图像被照相机拍摄，并且通过摇摄来控制相机的方向。

公开(公告)号：US20090059011A1

公开(公告)日：2009-03-05

申请号：US12022681

申请日：2008-01-30

Applicant: Junhua Sun ,  Guangjun Zhang ,  Zhenzhong Wei

Inventor： Junhua Sun ,  Guangjun Zhang ,  Zhenzhong Wei

IPC: H04N17/00

CPC classification number: G01B11/2504

Abstract: This disclosure provides a calibration method for structure parameters of a structured-light vision sensor, which includes setting up the coordinate frames of a camera, image plane and target for calibration. The calculation of coordinates in the camera coordinate frame of stripes, projected by structured-light, on the planar target and a structured-light equation fitting according to the coordinates in the camera coordinate frame of the stripes on the planar target, by moving the planar target arbitrarily multiple times. The calibration method of the structured-light vision sensor provided by the disclosure is easy to operate and no auxiliary apparatus is needed, which can not only promote the efficiency of the calibration of structured-light, but also extend the application scope of calibration of structured-light.

Abstract translation: 本公开提供了一种用于结构光视觉传感器的结构参数的校准方法，其包括设置相机，图像平面和用于校准的目标的坐标系。 通过平面目标上的结构光投影的条纹的相机坐标系中的坐标的计算和通过平面目标上的条纹的相机坐标系中的坐标拟合的结构光方程， 目标任意多次。 本发明提供的结构光视觉传感器的校准方法易于操作，不需要辅助设备，这不仅可以提高结构光的校准效率，而且可以扩展结构光校准的应用范围 -光。

公开(公告)号：US20080317283A1

公开(公告)日：2008-12-25

申请号：US11967772

申请日：2007-12-31

Applicant: Guangjun Zhang ,  Jie Jiang ,  Qiaoyun Fan

Inventor： Guangjun Zhang ,  Jie Jiang ,  Qiaoyun Fan

IPC: G01C9/06 ,  G01C11/04

CPC classification number: G01J1/42 ,  G01C21/02 ,  G01J1/04 ,  G01J1/0437 ,  G01J2001/4266 ,  G01S3/784 ,  G01S3/7862

Abstract: The disclosure relates to a signal processing method for multi aperture sun sensor comprising the following steps: reading the information of sunspots in a row from a centroid coordinate memory, judging the absence of sunspots in that row, identifying the row and column index of the sunspots in the complete row, selecting the corresponding calibration parameter based on the row and column index, calculating attitude with the attitude calculation module the corresponding to identified sunspots, averaging the accumulated attitude of all sunspots and outputting the final attitude. At the same time, a signal processing device for multi aperture sun sensor is also presented. It is comprised of a sunspot absence judgment and an identification module and an attitude calculation module. The disclosure implements the integration of sun sensors without additional image processor or attitude processor, reduces field programmable gate array resource and improves the reliability of sun sensors.

Abstract translation: 本发明涉及一种用于多孔径太阳传感器的信号处理方法，包括以下步骤：从质心坐标存储器读取一行中的太阳黑子的信息，判断该行中没有太阳黑子，识别太阳黑子的行和列索引 在完整的行中，根据行和列索引选择相应的校准参数，使用姿态计算模块计算对应于确定的太阳黑子的姿态，对所有太阳黑子的累积姿态进行平均并输出最终姿态。 同时，还提出了一种用于多光圈太阳传感器的信号处理装置。 它由无太阳黑子判断和识别模块和姿态计算模块组成。 该公开实现了太阳传感器的集成，而没有附加的图像处理器或姿态处理器，减少了现场可编程门阵列资源并且提高了太阳传感器的可靠性。

公开(公告)号：US09524555B2

公开(公告)日：2016-12-20

申请号：US14360388

申请日：2011-12-12

Applicant: Guangjun Zhang ,  Zhenzhong Wei ,  Wei Wang

Inventor： Guangjun Zhang ,  Zhenzhong Wei ,  Wei Wang

IPC: G06T7/00 ,  G06T3/00 ,  H04N13/02

CPC classification number: G06T7/0046 ,  G06T3/0075 ,  G06T7/75 ,  G06T2207/10012 ,  H04N13/239

Abstract: A method and software for the simultaneous pose and points-correspondences determination from a planar model are disclosed. The method includes using a coarse pose estimation algorithm to obtain two possible coarse poses, and using each one of the two coarse poses as the initialization of the extended TsPose algorithm to obtain two candidate estimated poses; selecting one from the two candidate estimated poses based on the cost function value. Thus, The method solves the problem of pose redundancy in the simultaneous pose and points-correspondences determination from a planar model, i.e., the problem that the numbers of estimated poses increase exponentially as the iterations go. The disclosed embodiment is based on the coplanar points, and does not place restriction on the shape of a planar model. It performs well in a cluttered and occluded environment, and is noise-resilient in the presence of different levels of noise.

Abstract translation: 公开了一种用于从平面模型确定同时姿势和点对应关系的方法和软件。 该方法包括使用粗略姿态估计算法获得两种可能的粗略姿态，并且使用两个粗略姿势中的每一个作为扩展TsPose算法的初始化来获得两个候选估计姿态; 基于成本函数值从两个候选估计姿势中选择一个。 因此，该方法解决了从平面模型的同时姿态和点对应确定中的姿态冗余的问题，即估计姿态的数量随着迭代而呈指数级增加的问题。 所公开的实施例基于共面点，并且不对平面模型的形状施加限制。 它在杂乱无章的环境中表现良好，并且在存在不同程度的噪声的情况下具有抗噪声能力。

公开(公告)号：US08964027B2

公开(公告)日：2015-02-24

申请号：US13393555

申请日：2011-08-09

Applicant: Guangjun Zhang ,  Zhen Liu ,  Zhenzhong Wei ,  Junhua Sun ,  Meng Xie

Inventor： Guangjun Zhang ,  Zhen Liu ,  Zhenzhong Wei ,  Junhua Sun ,  Meng Xie

IPC: H04N7/18 ,  G01C25/00 ,  G06T7/00

CPC classification number: G01C25/00 ,  G06T7/85 ,  G06T2207/10012 ,  G06T2207/30208

Abstract: The present disclosure provides a global calibration method based on a rigid bar for a multi-sensor vision measurement system, comprising: step 1, executing the following procedure for at least nine times: placing, in front of two vision sensors to be calibrated, a rigid bar fasten with two targets respectively corresponding to the vision sensors; capturing images of the respective targets by their corresponding vision sensors; extracting coordinates of feature points of the respective targets in their corresponding images; and computing 3D coordinates of each feature points of the respective targets under their corresponding vision sensor coordinate frames; and Step 2, computing the transformation matrix between the two vision sensors, with the constraint of the fixed position relationship between the two targets. The present disclosure also provides a global calibration apparatus based on a rigid bar for a multi-sensor vision measurement system. Putting the present disclosure to use can increase the accuracy of the global calibration, and also be suitable for the calibration of the multi-sensor vision system in the extremely large working space, which enlarges the application range of the present disclosure.

Abstract translation: 本公开提供了一种基于用于多传感器视觉测量系统的刚性杆的全局校准方法，包括：步骤1，执行以下程序至少九次：将待校准的两个视觉传感器前面放置一个 刚性杆用两个目标分别对应于视觉传感器; 通过其相应的视觉传感器捕获各个目标的图像; 提取其对应图像中各个目标的特征点的坐标; 并计算其对应的视觉传感器坐标系下各个目标的每个特征点的3D坐标; 和步骤2，计算两个视觉传感器之间的变换矩阵，与两个目标之间的固定位置关系的约束。 本公开还提供了一种基于用于多传感器视觉测量系统的刚性杆的全局校准装置。 使用本公开可以提高全局校准的精度，并且还适合于在极大的工作空间中校准多传感器视觉系统，这扩大了本发明的应用范围。

公开(公告)号：US20140321735A1

公开(公告)日：2014-10-30

申请号：US14360388

申请日：2011-12-12

Applicant: Guangjun Zhang ,  Zhenzhong Wei ,  Wei Wang

Inventor： Guangjun Zhang ,  Zhenzhong Wei ,  Wei Wang

IPC: G06T7/00

CPC classification number: G06T7/0046 ,  G06T3/0075 ,  G06T7/75 ,  G06T2207/10012 ,  H04N13/239

Abstract: A method and software for the simultaneous pose and points-correspondences determination from a planar model are disclosed. The method includes using a coarse pose estimation algorithm to obtain two possible coarse poses, and using each one of the two coarse poses as the initialization of the extended TsPose algorithm to obtain two candidate estimated poses; selecting one from the two candidate estimated poses based on the cost function value. Thus, The method solves the problem of pose redundancy in the simultaneous pose and points-correspondences determination from a planar model, i.e., the problem that the numbers of estimated poses increase exponentially as the iterations go. The disclosed embodiment is based on the coplanar points, and does not place restriction on the shape of a planar model. It performs well in a cluttered and occluded environment, and is noise-resilient in the presence of different levels of noise.

Abstract translation: 公开了一种用于从平面模型确定同时姿势和点对应关系的方法和软件。 该方法包括使用粗略姿态估计算法获得两种可能的粗略姿态，并且使用两个粗略姿势中的每一个作为扩展TsPose算法的初始化来获得两个候选估计姿态; 基于成本函数值从两个候选估计姿势中选择一个。 因此，该方法解决了从平面模型的同时姿态和点对应确定中的姿态冗余的问题，即估计姿态的数量随着迭代而呈指数级增加的问题。 所公开的实施例基于共面点，并且不对平面模型的形状施加限制。 它在杂乱无章的环境中表现良好，并且在存在不同程度的噪声的情况下具有抗噪声能力。

公开(公告)号：US20130209751A1

公开(公告)日：2013-08-15

申请号：US13698728

申请日：2011-05-18

Applicant: Guangjun Zhang ,  José Zimmer ,  He Feng ,  Gerhard Lautenschlaeger ,  Armin Vogl

Inventor： Guangjun Zhang ,  José Zimmer ,  He Feng ,  Gerhard Lautenschlaeger ,  Armin Vogl

IPC: C03C3/112

CPC classification number: C03C3/112 ,  C03C3/095 ,  Y10T428/24628 ,  Y10T428/315

Abstract: An alkali aluminosilicate glass for 3D precision molding and thermal bending is provided. The glass has a working point lower than 1200° C. (104 dPas) and a transition temperature Tg lower than 610° C. The glass has, based on a sum of all the components in percentage by weight, 51-63% of Si02; 5-18% of Al203; 8-16% of Na20; 0-6% of K20; 3.5-10% of MgO; 0-5% of B203; 0-4.5% of Li20; 0-5% of ZnO; 0-8% of CaO; 0.1-2.5% of Zr02; 0.01-

Abstract translation: 提供用于3D精密成型和热弯曲的碱性硅铝酸盐玻璃。 该玻璃具有低于1200℃（104dPas）的工作点和低于610℃的转变温度Tg。该玻璃基于所有组分的总和，以重量百分比计，51-63％的SiO 2 ; 5-18％Al2O3; 8-16％的Na2O; 0〜6％K20; 3.5-10％的MgO; 0〜5％的B203; 0-4.5％的Li20; 0-5％ZnO; 0-8％的CaO; 0.1-2.5％的ZrO2; 0.01-0.2％的CeO2; 0-0.5％的F2; 0.01-0.5％的SnO 2; 0-3％的BaO; 0-3％的SrO; 0〜0.5％的Yb203; 其中SiO 2 + Al 2 O 3的总和为63〜81％，CaO + MgO的总和为3.5〜18％，Na 2 O /（Li 2 O + Na 2 O + K 2 O）的比例为0.4〜1.5。

公开(公告)号：US20120162414A1

公开(公告)日：2012-06-28

申请号：US13393555

申请日：2011-08-09

Applicant: Guangjun Zhang ,  Zhen Liu ,  Zhenzhong Wei ,  Junhua Sun ,  Meng Xie

Inventor： Guangjun Zhang ,  Zhen Liu ,  Zhenzhong Wei ,  Junhua Sun ,  Meng Xie

IPC: H04N7/18

CPC classification number: G01C25/00 ,  G06T7/85 ,  G06T2207/10012 ,  G06T2207/30208

Abstract: The present disclosure provides a global calibration method based on a rigid bar for a multi-sensor vision measurement system, comprising: step 1, executing the following procedure for at least nine times: placing, in front of two vision sensors to be calibrated, a rigid bar fasten with two targets respectively corresponding to the vision sensors; capturing images of the respective targets by their corresponding vision sensors; extracting coordinates of feature points of the respective targets in their corresponding images; and computing 3D coordinates of each feature points of the respective targets under their corresponding vision sensor coordinate frames; and Step 2, computing the transformation matrix between the two vision sensors, with the constraint of the fixed position relationship between the two targets. The present disclosure also provides a global calibration apparatus based on a rigid bar for a multi-sensor vision measurement system. Putting the present disclosure to use can increase the accuracy of the global calibration, and also be suitable for the calibration of the multi-sensor vision system in the extremely large working space, which enlarges the application range of the present disclosure.

Abstract translation: 本公开提供了一种基于用于多传感器视觉测量系统的刚性杆的全局校准方法，包括：步骤1，执行以下程序至少九次：将待校准的两个视觉传感器前面放置一个 刚性杆用两个目标分别对应于视觉传感器; 通过其相应的视觉传感器捕获各个目标的图像; 提取其对应图像中各个目标的特征点的坐标; 并计算其对应的视觉传感器坐标系下各个目标的每个特征点的3D坐标; 和步骤2，计算两个视觉传感器之间的变换矩阵，与两个目标之间的固定位置关系的约束。 本公开还提供了一种基于用于多传感器视觉测量系统的刚性杆的全局校准装置。 使用本公开可以提高全局校准的精度，并且还适合于在极大的工作空间中校准多传感器视觉系统，这扩大了本发明的应用范围。