专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明授权

US11544874B2 System and method for calibration of machine vision cameras along at least three discrete planes 有权
公开(公告)号：US11544874B2
公开(公告)日：2023-01-03
申请号：US16882273
申请日：2020-05-22
申请人： Cognex Corporation
发明人： Gang Liu , Guruprasad Shivaram , Cyril C. Marrion, Jr.
IPC分类号： G06T7/80 , H04N5/232 , H04N5/225 , H04N17/00
摘要： This invention provides a system and method for generating camera calibrations for a vision system camera along three discrete planes in a 3D volume space that uses at least two (e.g. parallel) object planes at different known heights. For any third (e.g. parallel) plane of a specified height, the system and method then automatically generates calibration data for the camera by interpolating/extrapolating from the first two calibrations. This alleviates the need to set the calibration object at more than two heights, speeding the calibration process and simplifying the user's calibration setup, and also allowing interpolation/extrapolation to heights that are space-constrained, and not readily accessible by a calibration object. The calibration plate can be calibrated at each height using a full 2D hand-eye calibration, or using a hand-eye calibration at the first height and then at a second height with translation to a known position along the height (e.g. Z) direction.

2. 发明授权

US11049280B2 System and method for tying together machine vision coordinate spaces in a guided assembly environment 有权
公开(公告)号：US11049280B2
公开(公告)日：2021-06-29
申请号：US16410672
申请日：2019-05-13
申请人： Cognex Corporation
发明人： Guruprasad Shivaram , Cyril C. Marrion, Jr. , Lifeng Liu , Tuotuo Li
IPC分类号： G06T7/80 , B25J9/16 , G06T7/73 , G06T7/33 , G06K9/46 , H04N5/247
摘要： This invention provides a system and method that ties the coordinate spaces at the two locations together during calibration time using features on a runtime workpiece instead of a calibration target. Three possible scenarios are contemplated: wherein the same workpiece features are imaged and identified at both locations; wherein the imaged features of the runtime workpiece differ at each location (with a CAD or measured workpiece rendition available); and wherein the first location containing a motion stage has been calibrated to the motion stage using hand-eye calibration and the second location is hand-eye calibrated to the same motion stage by transferring the runtime part back and forth between locations. Illustratively, the quality of the first two techniques can be improved by running multiple runtime workpieces each with a different pose, extracting and accumulating such features at each location; and then using the accumulated features to tie the two coordinate spaces.

3. 发明申请

US20150104068A1 SYSTEM AND METHOD FOR LOCATING FIDUCIALS WITH KNOWN SHAPE 有权
标题翻译：用于定位具有已知形状的纤维的系统和方法
公开(公告)号：US20150104068A1
公开(公告)日：2015-04-16
申请号：US14515481
申请日：2014-10-15
申请人： Cognex Corporation
发明人： Guruprasad Shivaram , Lowell D. Jacobson , David Y. Li
IPC分类号： G06T7/00
CPC分类号： G06T7/74 , G06K9/4604 , G06K9/6206 , G06T2207/20081 , G06T2207/30164 , G06T2207/30204
摘要： This invention provides a system and method for determining the pose of shapes that are known to a vision system that undergo both affine transformation and deformation. The object image with fiducial is acquired. The fiducial has affine parameters, including degrees of freedom (DOFs), search ranges and search step sizes, and control points with associated DOFs and step sizes. Each 2D affine parameter's search range and the distortion control points' DOFs are sampled and all combinations are obtained. The coarsely specified fiducial is transformed for each combination and a match metric is computed for the transformed fiducial, generating a score surface. Peaks are computed on this surface, as potential candidates, which are refined until a match metric is maximized. The refined representation exceeding a predetermined score is returned as potential shapes in the scene. Alternately the candidate with the best score can be used as a training fiducial.
摘要翻译：本发明提供一种用于确定进行仿射变换和变形的视觉系统已知的形状姿态的系统和方法。获取具有基准的物体图像。基准具有仿射参数，包括自由度（DOF），搜索范围和搜索步长，以及具有相关DOF和步长的控制点。对每个2D仿射参数的搜索范围和失真控制点的DOF进行采样，并获得所有组合。为每个组合转换粗略指定的基准，并为变换的基准计算匹配度量，生成分数表面。在该表面上计算峰值作为潜在候选，其在精度匹配度量最大化之前进行细化。超过预定分数的精细表示作为场景中的潜在形状返回。或者，具有最佳成绩的候选人可以用作培训基准。

4. 发明授权

US10664994B2 System and method for calibration of machine vision cameras along at least three discrete planes 有权
公开(公告)号：US10664994B2
公开(公告)日：2020-05-26
申请号：US13776617
申请日：2013-02-25
申请人： Cognex Corporation
发明人： Gang Liu , Guruprasad Shivaram , Cyril C. Marrion, Jr.
IPC分类号： G06T7/80 , H04N5/232 , H04N5/225 , H04N17/00
摘要： This invention provides a system and method for generating camera calibrations for a vision system camera along three discrete planes in a 3D volume space that uses at least two (e.g. parallel) object planes at different known heights. For any third (e.g. parallel) plane of a specified height, the system and method then automatically generates calibration data for the camera by interpolating/extrapolating from the first two calibrations. This alleviates the need to set the calibration object at more than two heights, speeding the calibration process and simplifying the user's calibration setup, and also allowing interpolation/extrapolation to heights that are space-constrained, and not readily accessible by a calibration object. The calibration plate can be calibrated at each height using a full 2D hand-eye calibration, or using a hand-eye calibration at the first height and then at a second height with translation to a known position along the height (e.g. Z) direction.

5. 发明申请

US20170132807A1 SYSTEM AND METHOD FOR TYING TOGETHER MACHINE VISION COORDINATE SPACES IN A GUIDED ASSEMBLY ENVIRONMENT 审中-公开
公开(公告)号：US20170132807A1
公开(公告)日：2017-05-11
申请号：US15223436
申请日：2016-07-29
申请人： Cognex Corporation
发明人： Guruprasad Shivaram , Cyril C. Marrion, Jr. , Lifeng Liu , Tuotuo Li
IPC分类号： G06T7/80 , B25J9/16 , G06T7/73 , G06T7/33 , H04N5/247 , G06K9/46
CPC分类号： G06T7/80 , B25J9/1692 , B25J9/1697 , G05B2219/39039 , G06K9/4604 , G06T7/337 , G06T7/344 , G06T7/74 , G06T2207/30164 , H04N5/247 , Y10S901/47
摘要： This invention provides a system and method that ties the coordinate spaces at the two locations together during calibration time using features on a runtime workpiece instead of a calibration target. Three possible scenarios are contemplated: wherein the same workpiece features are imaged and identified at both locations; wherein the imaged features of the runtime workpiece differ at each location (with a CAD or measured workpiece rendition available); and wherein the first location containing a motion stage has been calibrated to the motion stage using hand-eye calibration and the second location is hand-eye calibrated to the same motion stage by transferring the runtime part back and forth between locations. Illustratively, the quality of the first two techniques can be improved by running multiple runtime workpieces each with a different pose, extracting and accumulating such features at each location; and then using the accumulated features to tie the two coordinate spaces.

6. 发明授权

US09569850B2 System and method for automatically determining pose of a shape 有权
标题翻译：用于自动确定形状姿态的系统和方法
公开(公告)号：US09569850B2
公开(公告)日：2017-02-14
申请号：US14515481
申请日：2014-10-15
申请人： Cognex Corporation
发明人： Guruprasad Shivaram , Lowell D. Jacobson , David Y. Li
IPC分类号： G06K9/00 , G06T7/00 , G06K9/46 , G06K9/62
CPC分类号： G06T7/74 , G06K9/4604 , G06K9/6206 , G06T2207/20081 , G06T2207/30164 , G06T2207/30204
摘要： This invention provides a system and method for determining the pose of shapes that are known to a vision system that undergo both affine transformation and deformation. The object image with fiducial is acquired. The fiducial has affine parameters, including degrees of freedom (DOFs), search ranges and search step sizes, and control points with associated DOFs and step sizes. Each 2D affine parameter's search range and the distortion control points' DOFs are sampled and all combinations are obtained. The coarsely specified fiducial is transformed for each combination and a match metric is computed for the transformed fiducial, generating a score surface. Peaks are computed on this surface, as potential candidates, which are refined until a match metric is maximized. The refined representation exceeding a predetermined score is returned as potential shapes in the scene. Alternately the candidate with the best score can be used as a training fiducial.
摘要翻译：本发明提供一种用于确定进行仿射变换和变形的视觉系统已知的形状姿态的系统和方法。获取具有基准的物体图像。基准具有仿射参数，包括自由度（DOF），搜索范围和搜索步长，以及具有相关DOF和步长的控制点。对每个2D仿射参数的搜索范围和失真控制点的DOF进行采样，并获得所有组合。为每个组合转换粗略指定的基准，并为变换的基准计算匹配度量，生成分数表面。在该表面上计算峰值作为潜在候选，其在精度匹配度量最大化之前进行细化。超过预定分数的精细表示作为场景中的潜在形状返回。或者，具有最佳成绩的候选人可以用作培训基准。

7. 发明申请

US20140240520A1 SYSTEM AND METHOD FOR CALIBRATION OF MACHINE VISION CAMERAS ALONG AT LEAST THREE DISCRETE PLANES 审中-公开
标题翻译：用于在最少三个离散平面上校准机器视觉摄像机的系统和方法
公开(公告)号：US20140240520A1
公开(公告)日：2014-08-28
申请号：US13776617
申请日：2013-02-25
申请人： Cognex Corporation
发明人： Gang Liu , Guruprasad Shivaram , Cyril C. Marrion, JR.
IPC分类号： H04N17/00
摘要： This invention provides a system and method for generating camera calibrations for a vision system camera along three discrete planes in a 3D volume space that uses at least two (e.g. parallel) object planes at different known heights. For any third (e.g. parallel) plane of a specified height, the system and method then automatically generates calibration data for the camera by interpolating/extrapolating from the first two calibrations. This alleviates the need to set the calibration object at more than two heights, speeding the calibration process and simplifying the user's calibration setup, and also allowing interpolation/extrapolation to heights that are space-constrained, and not readily accessible by a calibration object. The calibration plate can be calibrated at each height using a full 2D hand-eye calibration, or using a hand-eye calibration at the first height and then at a second height with translation to a known position along the height (e.g. Z) direction.
摘要翻译：本发明提供了一种用于在3D体积空间中的三个离散平面上为视觉系统照相机生成相机校准的系统和方法，其使用不同已知高度的至少两个（例如平行）物体平面。对于指定高度的任何第三（例如并行）平面，系统和方法然后通过从前两个校准进行内插/外插来自动生成相机的校准数据。这减轻了将校准对象设置在两个以上高度的需要，加速校准过程并简化用户的校准设置，并且还允许插值/外推到空间受限的高度，并且不容易被校准对象访问。可以使用全2D手眼校准在每个高度校准校准板，或者使用在第一高度处然后在第二高度处的手眼校准，沿着高度（例如Z）方向平移到已知位置。

8. 发明申请

US20190176335A1 CALIBRATION AND OPERATION OF VISION-BASED MANIPULATION SYSTEMS 审中-公开
公开(公告)号：US20190176335A1
公开(公告)日：2019-06-13
申请号：US15840495
申请日：2017-12-13
申请人： COGNEX CORPORATION
发明人： Guruprasad Shivaram , Gang Liu
IPC分类号： B25J9/16 , H04N7/18
摘要： A vision-based manipulations system can be configured for, and can be operated with methods including, performing hand-eye calibrations at multiple workstations of the system, performing a cross-station calibration for the system, and determining relationships between the hand-eye calibrations and the cross-station calibration. In some embodiments, the system can be used to move a work object between the workstations based on the cross-station calibration.

9. 发明授权

US10290118B2 System and method for tying together machine vision coordinate spaces in a guided assembly environment 有权
公开(公告)号：US10290118B2
公开(公告)日：2019-05-14
申请号：US15223436
申请日：2016-07-29
申请人： Cognex Corporation
发明人： Guruprasad Shivaram , Cyril C. Marrion, Jr. , Lifeng Liu , Tuotuo Li
IPC分类号： G06T7/80 , B25J9/16 , G06T7/73 , G06T7/33 , G06K9/46 , H04N5/247
摘要： This invention provides a system and method that ties the coordinate spaces at the two locations together during calibration time using features on a runtime workpiece instead of a calibration target. Three possible scenarios are contemplated: wherein the same workpiece features are imaged and identified at both locations; wherein the imaged features of the runtime workpiece differ at each location (with a CAD or measured workpiece rendition available); and wherein the first location containing a motion stage has been calibrated to the motion stage using hand-eye calibration and the second location is hand-eye calibrated to the same motion stage by transferring the runtime part back and forth between locations. Illustratively, the quality of the first two techniques can be improved by running multiple runtime workpieces each with a different pose, extracting and accumulating such features at each location; and then using the accumulated features to tie the two coordinate spaces.

10. 发明授权

US11072074B2 Calibration and operation of vision-based manipulation systems 有权
公开(公告)号：US11072074B2
公开(公告)日：2021-07-27
申请号：US15840495
申请日：2017-12-13
申请人： COGNEX CORPORATION
发明人： Guruprasad Shivaram , Gang Liu
IPC分类号： B25J9/16 , H04N7/18
摘要： A vision-based manipulations system can be configured for, and can be operated with methods including, performing hand-eye calibrations at multiple workstations of the system, performing a cross-station calibration for the system, and determining relationships between the hand-eye calibrations and the cross-station calibration. In some embodiments, the system can be used to move a work object between the workstations based on the cross-station calibration.

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式