Automated 360-degree dense point object inspection

Patent metadata

Field	Value
Publication number	US-12001191-B2
Application number	US-202117335882-A
Country	US
Kind code	B2
Filing date	Jun 1, 2021
Priority date	Nov 17, 2017
Publication date	Jun 4, 2024
Grant date	Jun 4, 2024

Abstract

A system and method for performing real-time quality inspection of objects is disclosed. The system and method include a transport to move objects being inspected, allowing the inspection to be performed in-line. At least one optical acquisition unit is provided that captured optical images of the objects being inspected. The captured optical images are matched to CAD models of objects, and the matched CAD model is extracted. A laser with an illumination light beam has a wavelength in the violet or ultraviolet range then conducts scans of the objects, which are formed into three-dimensional point clouds. The point clouds are compared to the extracted CAD models for each object, where CTF are compared to user input or CAD model information and the object is determined to be acceptable or defective based on the extent of deviation between the point cloud and the CAD model.

First claim

The invention claimed is:

1. An object inspection system, the system comprising: a scanning unit configured to scan a calibration target to obtain a point cloud representing the target; and a processor configured to receive the point cloud representing the target from the scanning unit, and to: tessellate the point cloud into one or more tiles containing a plurality of points of the point cloud; for each of the one or more tiles, fit a plane model to the points contained in the tile, wherein the plane model includes value of a three-dimensional vector normal to the surface of the plane and the distance of the plane from a coordinate system reference point; add the normal vectors from each of the one or more tiles and calculate the average of the collection of normal vectors; use average normal vector orientation to define a reference plane outside the point cloud and calculate each cloud point to reference plane; and use a histogram to find local maximums, and create a group of points each with a distance to the defined reference plane that falls within a tolerance threshold to the local maximum; and monitor the performance of the scanning unit.

2. The system of claim 1 wherein the one or more tiles are non-overlapping.

3. The system of claim 1 wherein the processor is further configured to:
create the histogram of distances from points of the point cloud to the reference plane.

4. The object inspection system of claim 1 , wherein the calibration target includes fiducial holes at known locations and positions, and wherein the processor is further configured to analyze the one or more plane models to detect the location and position of representations of these fiducial holes in the one or more plane models.

5. The object inspection system of claim 4 , wherein the processor is further configured to determine a transformation matrix, based on the detected locations and positions of the representations of fiducial holes, that determines the orientation of the calibration target in three-dimensional space.

6. The system of claim 1 wherein the processor is further configured to:
compare each of the normal vectors to the calculated average and remove any of the normal vectors that are outliers.

7. The system of claim 6 wherein the comparing step is performed as a dot product of the average normal vector and the normal vector being compared.

8. The system of claim 6 wherein the processor is further configured to:
recalculate the average when any of the normal vectors are removed as outliers.

9. An object inspection system, the system comprising:
a scanning unit; and
a processor configured receive the captured point clouds from the scanning unit, and to:
determine a system calibration transform;
determine a recommended scan set-up parameters for an object to be scanned;
instruct the scanning unit to perform at least one scan of the object, wherein the at least one scan includes capturing two or more point clouds representing the same view of the scanned object and two or more point clouds representing different views of the scanned object;
receive the captured point clouds from the scanning unit;
determine a three-dimensional registration between the two or more point clouds representing the same view of the object;
determine a three-dimensional registration between the two or more point clouds representing different views of the object; and
monitor the performance of the scanning unit.

10. A method for calibrating an object inspection system comprising: scanning a calibration target to obtain a point cloud representing the target; tessellating the point cloud into one or more tiles containing a plurality of points of the point cloud; for each of the one or more tiles, fitting a plane model to the points contained in the tile, wherein the plane model includes value of a three-dimensional vector normal to the surface of the plane and the distance of the plane from a coordinate system reference point; adding the normal vectors from each of the one or more tiles and calculating the average of the collection of normal vectors; using average normal vector orientation to define a reference plane outside the point cloud and calculating each cloud point to reference plane; using a histogram to find local maximums, and creating a group of points each with a distance to the defined reference plane that falls within a tolerance threshold to the local maximum, wherein each of these groups is a planar segment of the point cloud corresponding to a detected plane of the target; for each planar segment, fitting a plane model, and calculating the normal value of the segmented points; calculating the average of the normal values weighted by the population of points in each planar segment; and applying the average normal vector to each of the planar segments, and recalculate the distance from the coordinate origin for each plane; and monitor the performance of the scanning.

11. The method of claim 10 wherein the one or more tiles are non-overlapping.

12. The method of claim 10 further comprising:
creating the histogram of distances from points of the point cloud to the reference plane.

13. The method of claim 10 further comprising:
comparing each of the normal vectors to the calculated average and removing any of the normal vectors that are outliers.

14. The method of claim 13 wherein the comparing step is performed as a dot product of the average normal vector and the normal vector being compared.

15. The method of claim 13 further comprising:
recalculating the average when any of the normal vectors are removed as outliers.

Assignees

• Kodak Alaris Inc

Inventors

• Link Bruce A
• Johnson Robert W
• Zegarra Jose Zvietcovich
• Stubler Peter O
• Barton Mark
• Brockler Frank R
• Axman Michael S
• Das Madirakshi
• Ferencz Jr Gabor J

Classifications

• G05B19/4097 Primary

  characterised by using design data to control NC machines, e.g. CAD/CAM (G05B19/4093 takes precedence) · CPC title

• G06T7/001 Primary

  using an image reference approach · CPC title

• G06T7/33

  using feature-based methods · CPC title

• G06T7/70

  Determining position or orientation of objects or cameras (camera calibration G06T7/80) · CPC title

• G06T19/20

  Editing of three-dimensional [3D] images, e.g. changing shapes or colours, aligning objects or positioning parts · CPC title