patent.fyi
Applications  /  US 2026/0355970
United States Patent Application Publication

US 2026/0355970 A1

Published
Nov 12, 2026
Filed
Apr 6, 2023
Earliest date on the face
Apr 6, 2023
Classification
B25J 9/1697
First published here
Nov 19, 2026

Generated automatically. Not reviewed by a person. Not legal advice or an opinion.

Section 1 · machine-generated

What if?

Questions about how individual parts of claim 1 might be changed. A question may describe something different from what the claim language recites, or a version of it; no view is offered on which.

1.3Scale from coplanar reference marks

What ifWhat if two reference marks on the measured face supplied the image scale for each workpiece?
Claim language“acquiring a height of the workpiece”

A fixture places two marks with known separation in the same plane as the feature being located. With the image sensor parallel to that plane, the controller divides the known separation by the measured pixel separation and uses the midpoint of the marks as the coordinate origin.

Section 2 · machine-generated

Variations

Further arrangements of the same subject matter, described here as of the publication date at the foot of this page.

Other field of use

Height correction for parcel sorting

A fixed overhead camera images parcel tops as the parcels stop on a conveyor. A calibration grid placed on the empty belt provides the belt-plane scale in millimeters per pixel, which a stored scale curve maps to the lens-to-belt distance. The controller looks up the rate of scale change at that distance and subtracts the product of that rate and the parcel height from the belt-plane scale. The corrected scale converts the parcel center from pixels to millimeters for a downstream pusher.

Component substitution

Analytical calibration with a pinhole model

A fixed-focus camera with its optical axis perpendicular to the loading surface uses a calibrated pinhole model in place of a table of scale-change rates. A flat grid supplies the effective focal length in pixels during intrinsic calibration, and the processor rectifies lens distortion before measuring pixel coordinates. At installation, the processor measures the loading-surface scale with a ruler and multiplies that scale by the focal length in pixels to estimate the lens-to-surface distance. The model gives the scale-change rate as the reciprocal of the focal length in pixels, and the processor subtracts the product of that rate and the workpiece height from the measured scale.

Parameter or scale variation

Height correction across a wide tray

A 4096 by 3000 pixel camera is mounted perpendicular to a 600 by 400 millimeter loading tray at a nominal working distance of 900 millimeters. A grid with 25 millimeter spacing spans the tray during installation, and the controller uses the measured scale to infer the working distance from a stored calibration curve. The controller retrieves the scale-change rate at that distance and applies it to part heights from 0 to 120 millimeters to calculate the corrected scale. The controller converts feature offsets from the calibrated image center into millimeters along the tray axes.

Combination with existing systems

Height correction for palletizing robots

A fixed overhead camera views a loading table beside a palletizing robot, and fiducials on the table register the camera axes to the robot base frame. During installation, the controller measures the fiducial spacing in pixels, infers the working distance from the measured table scale, and retrieves the corresponding scale-change rate. For each carton, the controller reads its height from the palletizing recipe and corrects the scale before converting the detected top-center coordinates to millimeters. The robot controller uses the registered planar coordinates and the carton height to position the vacuum gripper.

Automated or data-driven variant

Inline height measurement by laser triangulation

A laser triangulation displacement sensor measures the top of each part at a stationary inspection position beneath the camera. The controller records the bare-surface reading and subtracts each part reading from it to obtain the height, with the sensor configured to report distance. An installation image of a calibration grid supplies the surface scale, which the controller uses to infer the working distance and look up the scale-change rate. The controller pairs the sensor reading with the camera frame from the same inspection cycle and uses the corrected scale to compute the part position.

Scale correction from a fitted curve

A motorized vertical stage moves a planar grid through a series of measured distances from a fixed-focus camera, and software extracts the millimeters-per-pixel scale from each image. The software fits a monotonic spline to the full set of distance and scale measurements and uses its derivative as the relationship between working distance and scale-change rate. At installation, the processor inverts the fitted curve at the measured loading-surface scale to recover the working distance. During operation, the processor integrates the fitted rate from the inferred surface distance to that distance minus the part height and adds the signed result to the measured surface scale.

Specific version of a part

Fixed scale with telecentric optics

An object-side telecentric lens holds magnification within the measurement tolerance over a specified axial range. A calibration grid supplies one millimetres-per-pixel factor, and a fixture keeps the measured workpiece face within that range.

Direct measurement of camera standoff

A calibrated depth gauge measures the distance from a machined camera mounting datum to the loading surface along the optical axis. The controller applies a stored offset from that datum to the lens reference plane and uses the resulting working distance to look up the unit resolution.

Mechanical positioning at fixed distances

Two encoded vertical slide positions have stored image scales for two fixed distances from the loading surface. For each workpiece, the controller selects the nearer calibration position and adds the measured workpiece height to the slide command, restoring the corresponding viewing distance to the top face. The controller converts pixel coordinates with the stored scale for that position.

Calibration at stepped working distances

An encoded linear stage positions a dot grid at working distances spaced 10 mm apart. The controller calculates the image scale from the known dot spacing and stores the difference in scale between neighboring positions divided by the 10 mm step, indexed by the midpoint working distance.

Installation calibration with grid plate

A flat glass grid with 5 mm spacing rests on the loading surface during installation. The controller fits the pixel spacing across several grid intervals and converts it to millimetres per pixel. It interpolates a stored scale-versus-distance calibration to estimate the working distance, then interpolates the unit resolution at that distance.

Height correction with displacement sensing

A laser displacement sensor measures the distance to the loading surface and to a flat region of the workpiece, and the controller subtracts the second reading from the first. The controller multiplies that height by the estimated decrease in image scale per millimetre of height and subtracts the result from the loading-surface scale.

Section 3 · quoted from USPTO records

Other work

Passages reproduced word for word from two earlier-published U.S. patent documents, placed beside the language of claim 1. No characterization is offered. Each passage is marked A or B for the document it comes from; ¶ numbers are the paragraph numbers in the USPTO’s published full text of that document.

Document AUS 7,012,703 B2
Published
Mar 14, 2006
Filed
Aug 21, 2003
Class
G01B 11/14
Document BUS 9,769,428 B2
Published
Sep 19, 2017
Filed
Feb 1, 2013
Class
H04N 7/18
Claim 1
Document A
Document B
1.1a first step of acquiring a first relationship that is a relationship between a unit resolution serving as an amount of change in resolution per unit height and a working distance serving as a distance between a target object and a lens of the two-dimensional camera;
At this time, the size to be captured in the image is inversely proportional to the distance between the workpiece 4 and the fixed camera 2, so that the following equation holds. SZ2/SZ1=D1/D2=(Z0−Z1)/(Z0−Z2) (Equation 1)
B · US 9,769,428 B2 · ¶ 40
1.2a second step of measuring a resolution of the image of the two-dimensional camera with respect to the loading surface of the workpiece at an installation height of the two-dimensional camera when the two-dimensional camera is installed to estimate a working distance for measurement, and estimating the unit resolution based on the estimated working distance and the first relationship; and
In one embodiment, the number of encoder counts occurring between the sensing of the identifying characteristic by first sensor 110 and second sensor 114 is divided into distance 122 to determine the encoder resolution. As such, when encoder resolution is determined by dividing distance 122 that is measured in inches by the difference between the second encoder count and the first encoder count, the determined encoder resolution has a unit of measure of inches per encoder count.
A · US 7,012,703 B2 · ¶ 29
1.3a third step of acquiring a height of the workpiece and correcting the resolution of the image based on the acquired height of the workpiece and the estimated unit resolution.
The search window is calculated by way of adjusting the reference search window using the container reference position and posture and the workpiece reference height that are stored when the reference data is taught, and the container position and posture and the workpiece height that are acquired when executing the workpiece detection.
B · US 9,769,428 B2 · ¶ 34

An empty cell means only that no passage was reproduced there.

Statements in these documents
In particular, the invention relates to a system and method for automatically determining a resolution of an encoder, automatically calibrating the encoder resolution, and automatically determining a length of a workpiece from the encoder resolution and encoder count.
A · US 7,012,703 B2 · ¶ 1
However, operation systems utilizing an encoder encounter a number of conditions that cause the resolution of the encoder to be inaccurate, thereby resulting in inaccurate measurements.
A · US 7,012,703 B2 · ¶ 4
More particularly, the present invention relates to an image processing device that is directed to preventing workpieces from being misdetected or undetected and reducing the processing time for workpiece detection by adjusting the search window to use when detecting the workpieces by the image processing device.
B · US 9,769,428 B2 · ¶ 2
Then, problems occur where, for example, workpieces are not found in the search window, unwanted objects are captured in the search window and cause misdetection, the search window is too large and the detection consumes additional time, and so on.
B · US 9,769,428 B2 · ¶ 3
Cite aspatent.fyi, US 2026/0355970 A1, claim 1, first published 2026-11-19T18:11:03-05:00, https://patent.fyi/US20260355970/, SHA-256 a8429bacde684e2342aaac0fc525754c2167e24901840f06c3ab1a15d66aaf22
First published
Permanent addresshttps://patent.fyi/US20260355970/
Content fingerprint (SHA-256)a8429bacde684e2342aaac0fc525754c2167e24901840f06c3ab1a15d66aaf22
Methodv2026-11-18.2 · automated, not reviewed by a person
Correctionscorrections@patent.fyi