System and method for runtime determination of camera miscalibration

A system for determining camera miscalibration in a system of at least three cameras, comprising:

• a) at least three cameras, each including respective intrinsic calibration parameters, extrinsic calibration parameters, and respective error statistics estimated prior to a runtime vision inspection process and during a first calibration process using a calibration object, each of the at least three cameras is located at a different orientation with respect to an inspection object in a three-dimensional space, wherein the inspection object is different from the calibration object;
• b) a first plurality of the at least three cameras that find a first feature on the inspection object in three-dimensional space, during the runtime vision inspection process in which the inspection object is aligned or inspected based upon a training model, intrinsic calibration parameters, extrinsic calibration parameters, and the error statistics, and determine a first estimated location of the first feature;
• c) a second plurality of the at least three cameras that find a second feature on the inspection object in three-dimensional space, during the runtime vision inspection process in which an inspection object is aligned or inspected based upon a training model, intrinsic calibration parameters, extrinsic calibration parameters, and the error statistics, and determine a second estimated location;
• d) a comparison process that compares the first estimated location with a first predicted location from the training model to generate an alignment score to be used in alignment or inspection of the inspection object, compares the second estimated location with a second predicted location from the training model to generate a second alignment score to be used in alignment or inspection of the inspection object, and compares the first alignment score and the second alignment score with respect to at least one of (i) an accuracy associated with the extrinsic calibration parameters or (ii) a desired system accuracy;
• e) a recalibration determination process that, in response to the comparison process d) and during the runtime vision process, determines that a second calibration process is desired based upon a result of the step of comparing exceeding at least one of the (i) accuracy determined during step (a) and (ii) the desired system accuracy;
• and
• f) a recalibration process that ceases the runtime vision inspection process, and performs the second calibration process to at least one of the at least three cameras, wherein the second calibration process is performed with respect to a different object than the runtime vision inspection process, and wherein the second calibration process comprises redetermining the respective intrinsic calibration parameters, the extrinsic calibration parameters, and the respective error statistics for at least one of the at least three cameras.

A method for determining camera miscalibration in a system of at least three cameras, each camera located at a different orientation with respect to a calibration object in a three-dimensional space, the method comprising the steps of:

• a) during a first calibration process, receiving calibration data including at least one of:
• estimated respective intrinsic calibration parameters for each of the at least three cameras, estimated respective extrinsic calibration parameters for each of the at least three cameras, or respective error statistics for each of the at least three cameras;
• b) during the runtime vision inspection process in which an inspection object is aligned or inspected based upon a training model and the at least one of the estimated intrinsic calibration parameters, the estimated extrinsic calibration parameters, or the error statistics, finding a first feature on the inspection object in the three-dimensional space with a first plurality of the at least three cameras and determining a first estimated location of the first feature, wherein the inspection object is different from the calibration object;
• c) during the runtime vision inspection process, comparing the first estimated location with a first predicted location from the training model to generate a first alignment score to be used in alignment or inspection of the inspection object;
• d) during the runtime vision inspection process, comparing the first alignment score with respect to at least one of (i) an accuracy determined during step (a), or (ii) a desired system accuracy;
• e) in response to step (d) and during the runtime vision inspection process, determining that a second calibration process is desired based upon a result of the step of comparing exceeding at least one of the (i) accuracy determined during step (a) and (ii) the desired system accuracy;
• f) in response to step (e), ceasing the runtime vision inspection process;
• and
• g) in response to step (f), performing the second calibration process to recalibrate at least one of the at least three cameras, wherein the second calibration process is performed with respect to a different object than the runtime vision inspection process, and wherein the second calibration process comprises at least one of:
• estimating the respective intrinsic calibration parameters, estimating the extrinsic calibration parameters for at least one of the at least three cameras, or redetermining the respective error statistics for at least one of the at least three cameras.