公开(公告)号：WO2008147355A1

公开(公告)日：2008-12-04

申请号：PCT/US2007/012637

申请日：2007-05-29

Applicant: COGNEX TECHNOLOGY AND INVESTMENT CORPORATION ,  MICHAEL, David, J. ,  WALLACK, Aaron, S.

Inventor： MICHAEL, David, J. ,  WALLACK, Aaron, S.

IPC: G06T7/00

CPC classification number: G06T7/0006 ,  G06T7/593 ,  G06T2207/10012 ,  G06T2207/30164

Abstract: A method and apparatus for assembly verification is disclosed. A measurement of the 3D position of each subcomponent is performed using triangulation from three cameras acquiring images simultaneously. An operator trains one 2D model, correspond to the same subcomponent of the assembly, per camera. At run-time, models are registered in each camera view so as to provide measured 3D position of the subcomponents. Then, measured 3D positions are compared with expected nominal 3D positions, and differences in 3D position are checked against tolerances. The invention simplifies the task of assembly verification, requiring only multiple cameras fixed above an assembly line. After minor operator activity, the invention can then perform assembly verification automatically. Since the invention can perform fixtureless assembly verification, a part can be presented to the machine vision system with arbitrary 3D position and orientation. Stroboscopic illumination can be used to illuminate parts on a rapidly moving assembly line.

Abstract translation: 公开了一种装配验证的方法和装置。 使用三个同时采集图像的摄像机的三角测量来执行每个子组件的3D位置的测量。 每个摄像机操作员训练一个2D模型，对应于组件的相同子组件。 在运行时，在每个摄像机视图中注册模型，以提供子组件的测量3D位置。 然后，将测量的3D位置与预期的标称3D位置进行比较，并且针对公差检查3D位置的差异。 本发明简化了装配验证的任务，仅需要在装配线上固定多个相机。 在轻微的操作员活动之后，本发明可以自动执行装配验证。 由于本发明可以执行无夹具组装验证，所以可以将具有任意3D位置和取向的部件呈现给机器视觉系统。 频闪照明可用于照亮快速移动的装配线上的部件。

公开(公告)号：WO1995030307A1

公开(公告)日：1995-11-09

申请号：PCT/US1995005320

申请日：1995-05-01

Applicant: COGNEX CORPORATION ,  PETRY, John, P. ,  MICHAEL, David, J. ,  GARAKANI, Arman

Inventor： COGNEX CORPORATION

IPC: H04N07/18

CPC classification number: H01L21/67288 ,  G01R31/2853 ,  G01R31/311 ,  H01L21/67271 ,  H01L24/48 ,  H01L24/49 ,  H01L24/78 ,  H01L24/85 ,  H01L2224/05553 ,  H01L2224/05554 ,  H01L2224/48465 ,  H01L2224/49171 ,  H01L2224/78 ,  H01L2224/78301 ,  H01L2224/85 ,  H01L2224/859 ,  H01L2924/00014 ,  H01L2924/01039 ,  H01L2924/14 ,  H04N7/18 ,  H01L2224/45099 ,  H01L2224/05599 ,  H01L2924/00

Abstract: This invention provides a method and apparatus for automatically inspecting the connection of a wire (50) to a lead (30) on a lead frame (10) containing a semiconductor chip (20) or similar device. Using an image processor (06) to locate the general position of a soldered lead (30) in a digitized image, the present invention creates, in Step (A), a template (120) of an idealized optical indentation left by a good bond; determines parameters such as wire angle, idealized position and shape thresholds for applying the template (120); conducts a normalized correlation search of a digitized image in Step (C); compares the results returned to the parameters and reports, in Step (D) the resulting signals generated by this comparison to a host controller (08) or other control module.

Abstract translation: 本发明提供了一种用于自动检查包含半导体芯片（20）或类似装置的引线框架（10）上的导线（30）与引线（30）的连接的方法和装置。 使用图像处理器（06）来定位焊接引线（30）在数字化图像中的一般位置，本发明在步骤（A）中创建由良好的键合留下的理想光学压痕的模板（120） ; 确定用于应用模板（120）的诸如线角度，理想位置和形状阈值的参数; 在步骤（C）中进行数字化图像的归一化相关搜索; 将返回到参数和报告的结果与步骤（D）中的结果信号进行比较，该结果由该比较生成的信号与主机控制器（08）或其他控制模块进行比较。

公开(公告)号：WO2008153721A1

公开(公告)日：2008-12-18

申请号：PCT/US2008/006535

申请日：2008-05-22

Applicant: COGNEX CORPORATION ,  WALLACK, Aaron, S. ,  MICHAEL, David, J.

Inventor： WALLACK, Aaron, S. ,  MICHAEL, David, J.

IPC: G06T7/00

CPC classification number: G06K9/32 ,  G06T7/73 ,  G06T2207/30164

Abstract: This invention provides a system and method for determining position of a viewed object in three dimensions by employing 2D machine vision processes on each of a plurality of planar faces of the object, and thereby refining the location of the object. First a rough pose estimate of the object is derived. This rough pose estimate can be based upon predetermined pose data, or can be derived by acquiring a plurality of planar face poses of the object (using, for example multiple cameras) and correlating the corners of the trained image pattern, which have known coordinates relative to the origin, to the acquired patterns. Once the rough pose is achieved, this is refined by defining the pose as a quaternion (a, b, c and d) for rotation and a three variables (x, y, z) for translation and employing an iterative weighted, least squares error calculation to minimize the error between the edgelets of trained model image and the acquired runtime edgelets. The overall, refined/optimized pose estimate incorporates data from each of the cameras' acquired images. Thereby, the estimate minimizes the total error between the edgelets of each camera's/view's trained model image and the associated camera's/view's acquired runtime edgelets. A final transformation of trained features relative to the runtime features is derived from the iterative error computation.

Abstract translation: 本发明提供一种用于通过在对象的多个平面中的每一个上采用2D机器视觉处理来确定三维视图对象的位置的系统和方法，从而改善对象的位置。 首先推导出对象的粗略姿态估计。 这种粗略姿态估计可以基于预定的姿态数据，或者可以通过获取对象的多个平面脸部姿势（使用例如多个照相机）并且将已训练图像图案的角部相关联而得到，其具有已知的坐标相对 到原产地，获得的模式。 一旦实现了粗糙姿态，就通过将姿态定义为旋转的四元数（a，b，c和d）和用于平移的三个变量（x，y，z）来进行改进，并采用迭代加权最小二乘误差 计算以最小化训练模型图像的边缘与所获取的运行时间边缘之间的误差。 整体，精细/优化的姿态估计包含来自每个摄像机所获取图像的数据。 因此，估计使每个相机/视图的训练模型图像的边缘与相关联的相机/视图的获取的运行时间边缘之间的总误差最小化。 相对于运行时特征的训练特征的最终变换是从迭代误差计算得出的。

公开(公告)号：WO2011079258A1

公开(公告)日：2011-06-30

申请号：PCT/US2010/061989

申请日：2010-12-23

Applicant: COGNEX CORPORATION ,  YE, Xiangyun ,  LI, David, Y. ,  SHIVARAM, Guruprasad ,  MICHAEL, David, J.

Inventor： YE, Xiangyun ,  LI, David, Y. ,  SHIVARAM, Guruprasad ,  MICHAEL, David, J.

IPC: G06T7/00

CPC classification number: G06T7/85 ,  G06T2207/10012 ,  G06T2207/30108

Abstract: This invention provides a system and method for runtime determination (self- diagnosis) of camera miscalibration (accuracy), typically related to camera extrinsics, based on historical statistics of runtime alignment scores for objects acquired in the scene, which are defined based on matching of observed and expected image data of trained object models. This arrangement avoids a need to cease runtime operation of the vision system and/or stop the production line that is served by the vision system to diagnose if the system's camera(s) remain calibrated. Under the assumption that objects or features inspected by the vision system over time are substantially the same, the vision system accumulates statistics of part alignment results and stores intermediate results to be used as indicator of current system accuracy. For multi-camera vision systems, cross validation is illustratively employed to identify individual problematic cameras. The system and method allows for faster, less- expensive and more-straightforward diagnosis of vision system failures related to deteriorating camera calibration.

Abstract translation: 本发明提供了一种用于基于场景中获取的对象的运行时对准分数的历史统计信息，其通常与摄像机外部特征相关的摄像机误差（精度）的运行时间确定（自诊断）的系统和方法， 训练对象模型的观察和预期图像数据。 这种安排避免了停止视觉系统的运行时操作和/或停止由视觉系统服务的生产线来诊断系统的相机是否保持校准的需要。 在视觉系统随时间检查的对象或特征基本相同的假设下，视觉系统累积部分对准结果的统计信息并存储中间结果，以用作当前系统精度的指标。 对于多摄像机视觉系统，交叉验证被说明性地用于识别单个有问题的摄像机。 该系统和方法允许对与相机校准恶化相关的视觉系统故障进行更快，更便宜和更直接的诊断。

公开(公告)号：WO1995030204A1

公开(公告)日：1995-11-09

申请号：PCT/US1995005273

申请日：1995-05-01

Applicant: COGNEX CORPORATION ,  KOLJONEN, Juha ,  MICHAEL, David, J.

Inventor： COGNEX CORPORATION

IPC: G06K09/62

CPC classification number: G06K9/4633 ,  G06T7/0004 ,  G06T7/62 ,  G06T7/66 ,  G06T7/70 ,  G06T2207/30152

Abstract: The invention automatically inspects the bond of a wire to a contact pad on a semiconductor chip. The apparatus includes a movable platform (03) for holding semiconductor chips (20) situated in leadframes (10); a video camera (04) for sensing images; illumination means (05) for illuminating a chip (20) in a lead frame (10); an image processor (06) to digitize and analyze the images; a bonding mechanism (07); and a host controller (08) electronically connected to bonding mechanism (07), movable platform (03), video camera (04) and image processor (06). Image processor (06) locates a bond on a pad in a digitized image and provides a first nominal center of ball bond image. The invention utilizes a polar coordinate transform image to evaluate the ball bond image by creating a polar projection histogram array. A peak in the histogram array is then dectected. The peak gives an indication of the size and position of the bond.

Abstract translation: 本发明自动检查导线与半导体芯片上接触焊盘的接合。 该装置包括用于保持位于引线框架（10）中的半导体芯片（20）的可移动平台（03）。 用于感测图像的摄像机（04） 照明装置（05），用于照亮引线框架（10）中的芯片（20）; 图像处理器（06），用于数字化和分析图像; 粘合机构（07）; 以及主机控制器（08），其电连接到接合机构（07），可移动平台（03），摄像机（04）和图像处理器（06）。 图像处理器（06）将数字化图像中的焊盘定位在焊盘上，并提供第一标称中心的焊球图像。 本发明利用极坐标变换图像，通过产生极性投影直方图阵列来评估球接合图像。 然后检查直方图阵列中的峰。 峰值表示债券的大小和位置。