Cognex Corporation

Huawei Technologies Co., Limited

Future Wei Technologies, Inc.

Lifeng Liu

Cyril C Marrion

Aaron S Wallack

Xiangyun Ye

Yingxuan Zhu

Jian Li

Guruprasad Shivaram

Jun Zhang

Xiaotian Yin

Tuotuo Li

David J Michael

Tian Gan

David Y Li

Nigel John Foster

Yongxiang Tao

Han Xiao

Dayao Liang

A data management system for collecting data of users from sensors at an event. An event-specific token corresponding to the event attended by the users is generated, where the event-specific token includes event-specific signatures about the users attending the event, and the signature of the event venue or facility. The public data is sent to public storage as an aggregation of the collected data and tagged with the event-specific token. The public data is statistically analyzed to track interactions and assess risks of the user's potential infection of a communicable disease based on anomalous activity detected at the event, or based on the notifications from the event attendees, which are matched and validated anonymously using the event-specific tokens. Based on a result of the statistical analysis, an advisory alert is sent to the users to warn them of a level of risk of harm based on attendance of the event.

Data management system to track and distribute public data collection with user privacy protection and event token exchange

A node in a distributed network computes a hash of content for a service received in a data packet. The node verifies the data packet by comparing the hash of the content of a service received from a neighboring node to a hash of the content computed by the node. An amount of content of the service having a same identification is accumulated in a trusted execution environment (TEE) of the node, and a signature based on code stored in a TEE of the node is generated. The node then sends the data packet to the next neighboring node, where the service-related information includes the service ID, a hash of the service content and the signature. The service records with the accumulated amount of service content, accumulated hash values, and nodes' signatures are sent to the validation nodes to reach consensus for the service provided.

System and method of establishing a trusted relationship in a distributed system

A computer implemented method for self-learning of a control system. The method includes creating an initial knowledge base. The method learns first principles using the knowledge base. The method creates initial control commands derived from the knowledge base. The method generates constraints for the control commands. The method performs constrained reinforcement learning by executing the control commands with the constraints and observing feedback to improve the control commands. The method enriches the knowledge base based on the feedback.

System and method of efficient, continuous, and safe learning using first principles and constraints

This invention provides a system and method for training and performing runtime 3D pose determination of an object using a plurality of camera assemblies in a 3D vision system. The cameras are arranged at different orientations with respect to a scene, so as to acquire contemporaneous images of an object, both at training and runtime. Each of the camera assemblies includes a non-perspective lens that acquires a respective non-perspective image for use in the process. The searched object features in one of the acquired non-perspective image can be used to define the expected location of object features in the second (or subsequent) non-perspective images based upon an affine transform, which is computed based upon at least a subset of the intrinsics and extrinsics of each camera. The locations of features in the second, and subsequent, non-perspective images can be refined by searching within the expected location of those images. This approach can be used in training, to generate the training model, and in runtime operating on acquired images of runtime objects. The non-perspective cameras can employ telecentric lenses.

System and method for training a model in a plurality of non-perspective cameras and determining 3D pose of an object at runtime with the same

A computer-implemented method of answering questions comprises: receiving, by one or more processors, a query; based on the query, generating, by the one or more processors, a matrix; based on the matrix, modifying, by the one or more processors, a dynamic memory; based on the matrix, determining, by the one or more processors, a first response from the dynamic memory; based on the matrix, determining, by the one or more processors, a second response from a database; based on the first response and the second response, determining, by the one or more processors, a third response; and in response to the query, providing, by the one or more processors, the third response.

Dynamic semantic networks for language understanding and question answering

A system and method for robustly calibrating a vision system and a robot is provided. The system and method enables a plurality of cameras to be calibrated into a robot base coordinate system to enable a machine vision/robot control system to accurately identify the location of objects of interest within robot base coordinates.

System and method for robust calibration between a machine vision system and a robot

Disclosed herein are systems and methods for machine vision. A machine vision system includes a motion rendering device, a first image sensor, and a second image sensor. The machine visions system includes a processor configured to run a computer program stored in memory that is configured to determine a first transformation that allows mapping between the first coordinate system associated with the motion rendering device and the second coordinate system associated with the first image sensor, and to determine a second transformation that allows mapping between the first coordinate system associated with the motion rendering device and the third coordinate system associated with the second image sensor.

Machine vision system calibration

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.

System and method for tying together machine vision coordinate spaces in a guided assembly environment

Described are machine vision systems and methods for simultaneous kinematic and hand-eye calibration. A machine vision system includes a robot and a 3D sensor in communication with a control system. The control system is configured to move the robot to poses, and for each pose: capture a 3D image of calibration target features and robot joint angles. The control system is configured to obtain initial values for robot calibration parameters, and determine initial values for hand-eye calibration parameters based on the initial values for the robot calibration parameters, the 3D image, and joint angles. The control system is configured to determine final values for the hand-eye calibration parameters and robot calibration parameters by refining the hand-eye calibration parameters and robot calibration parameters to minimize a cost function.

Simultaneous kinematic and hand-eye calibration

Methods, apparatus, and systems are provided for integrated driver expression recognition and vehicle interior environment classification to detect driver condition for safety. A method includes obtaining an image of a driver of a vehicle and an image of an interior environment of the vehicle. Using a machine learning method, the images are processed to classify a condition of the driver and of the interior environment of the vehicle. The machine learning method includes general convolutional neural network (CNN) and CNN with adaptive filters. The adaptive filters are determined based on influence of filters. The classification results are combined and compared with predetermined thresholds to determine if a decision can be made based on existing information. Additional information is requested by self-motivated learning if a decision cannot be made, and safety is determined based on the combined classification results. A warning is provided to the driver based on the safety determination.

Integrated system for detection of driver condition

Various embodiments include systems and methods structured to provide recognition of an object in an image using a learning module trained using decomposition of the object into components in a number of training images. The training can be based on an overall objectness score of the object, an objectness score of each component of the object, a pose of the object, and a pose of each component of the object for each training image input. Additional systems and methods can be implemented in a variety of applications.

Recognition and reconstruction of objects with partial appearance

Described are methods, systems, and apparatus, including computer program products for finding correspondences of one or more parts in a camera image of two or more cameras. For a first part in a first camera image of a first camera, a first 3D ray that is a first back-projection of a first feature coordinate of the first part in the first camera image to a 3D physical space is calculated. For a second part in a second camera image of a second camera, a second 3D ray that is a second back-projection of a second feature coordinate of the second part in the second camera image to the 3D physical space is calculated, wherein the first feature coordinate and the second feature coordinate correspond to a first feature as identified in a model. A first distance between the first 3D ray and the second 3D ray is calculated.

Multi-part corresponder for multiple cameras

Described are machine vision systems and methods for simultaneous kinematic and hand-eye calibration. A machine vision system includes a robot or motion stage and a camera in communication with a control system. The control system is configured to move the robot or motion stage to poses, and for each pose: capture an image of calibration target features and robot joint angles or motion stage encoder counts. The control system is configured to obtain initial values for robot or motion stage calibration parameters, and determine initial values for hand-eye calibration parameters based on the initial values for the robot or motion stage calibration parameters, the image, and joint angles or encoder counts. The control system is configured to determine final values for the hand-eye calibration parameters and robot or motion stage calibration parameters by refining the hand-eye calibration parameters and robot or motion stage calibration parameters to minimize a cost function.

Growing community of inventors

Sudbury, MA, United States of America

Lifeng Liu