Methods and apparatus for generating a dense field of three dimensional data for machine vision

Patent metadata

Field	Value
Publication number	US-10846563-B2
Application number	US-201816129121-A
Country	US
Kind code	B2
Filing date	Sep 12, 2018
Priority date	Sep 12, 2018
Publication date	Nov 24, 2020
Grant date	Nov 24, 2020

Abstract

The techniques described herein relate to methods, apparatus, and computer readable media configured to test a pose of a model in three-dimensional data. Three-dimensional data of an object is received, the three-dimensional data comprising a set of data entries. The three-dimensional data is converted to a field comprising a set of cells that each have an associated value, comprising determining, for each cell value, representative data based on one or more data entries from the set of data entries of the three-dimensional data. A pose of the model is tested with the field to determine a score for the pose.

First claim

The invention claimed is:

1. A computerized method for testing a pose of a model of an object in three-dimensional data, the method comprising:
receiving three-dimensional data of the object, the three-dimensional data comprising a set of data entries, wherein the set of data entries of the three-dimensional data comprises a list of points;
converting the three-dimensional data to a field comprising a set of cells, each cell of the set of cells having an associated value, wherein converting the three-dimensional data comprises determining, for the associated value for each cell of the set of cells, representative data based on one or more data entries from the set of data entries of the three-dimensional data, wherein determining, for the associated value for each cell of the set of cells, the representative data based on the one or more data entries comprises determining a vector based on one or more points in the list of points; and
testing a pose of the model with at least some of the representative data for the associated value for each cell of the set of cells of the field to determine a score for the pose of the model.

2. The method of claim 1 , wherein converting the three-dimensional data to the field further comprises generating a three-dimensional array of associated values for each cell of the set of cells.

3. The method of claim 1 , wherein converting the three-dimensional data to the field further comprises generating a densely-populated lattice, wherein the densely-populated lattice comprises associated values for each cell of the set of cells.

4. The method of claim 1 , wherein determining the vector based on the one or more points in the list of points comprises:
determining, based on the list of points, that the cell is associated with an interior portion of the object; and
based on determining that the cell is associated with an interior portion of the object, setting the vector to zero.

5. The method of claim 4 , wherein testing the pose of the model with at least some of the representative data for the associated value for each cell of the set of cells of the field to determine the score for the pose of the model comprises:
testing a set of probes of the model to at least some of the representative data for the associated value for each cell of the set of cells of the field to determine the score, each probe comprising a point of interest and associated data, the representative data for the associated value for each cell including an associated vector, the testing the set of probes of the model comprising summing a dot product of each probe and the associated vector in the field.

6. The method of claim 1 , wherein converting the three-dimensional data to the field comprising the set of cells, each cell of the set of cells having the associated value, comprising determining, for the associated value for each cell of the set of cells, the representative data based on the one or more data entries from the set of data entries of the three-dimensional data, the representative data including a representative vector, comprising:
determining a set of vectors based on the one or more data entries from the set of data entries of the three-dimensional data;
generating an accumulated matrix comprising accumulating an outer product of each vector of the set of vectors with itself; and
extracting eigenvectors, eigenvalues, or both, from the accumulated matrix to determine the representative vector.

7. The method of claim 1 , wherein determining the vector based on the one or more points in the list of points comprises:
determining one or more vectors based on the one or more points in the list of points; and
determining the vector based on the one or more vectors.

8. A system for testing a pose of a model of an object in three-dimensional data, the system comprising one or more processors configured to:
receive three-dimensional data of the object, the three-dimensional data comprising a set of data entries, wherein the set of data entries of the three-dimensional data comprises a list of points;
convert the three-dimensional data to a field comprising a set of cells, each cell of the set of cells having an associated value, wherein converting the three-dimensional data comprises determining, for the associated value for each cell of the set of cells, representative data based on one or more data entries from the set of data entries of the three-dimensional data, wherein determining, for the associated value for each cell of the set of cells, the representative data based on the one or more data entries comprises determining a vector based on one or more points in the list of points; and
test a pose of the model with at least some of the representative data for the associated value for each cell of the set of cells of the field to determine a score for the pose of the model.

9. The system of claim 8 , wherein converting the three-dimensional data to the field further comprises generating a three-dimensional array of associated values for each cell of the set of cells.

10. The system of claim 8 , wherein converting the three-dimensional data to the field further comprises generating a densely-populated lattice, wherein the densely-populated lattice comprises associated values for each cell of the set of cells.

11. The system of claim 8 , wherein determining the vector based on the one or more points in the list of points comprises:
determining, based on the list of points, that the cell is associated with an interior portion of the object; and
based on determining that the cell is associated with an interior portion of the object, setting the vector to zero.

12. The system of claim 11 , wherein testing the pose of the model with at least some of the representative data for the associated value for each cell of the set of cells of the field to determine the score for the pose of the model comprises:
testing a set of probes of the model to at least some of the representative data for the associated value for each cell of the set of cells of the field to determine the score, each probe comprising a point of interest and associated data, the representative data for the associated value for each cell including an associated vector, the testing the set of probes of the model comprising summing a dot product of each probe and the associated vector in the field.

13. The system of claim 8 , wherein converting the three-dimensional data to the field comprising the set of cells, each cell of the set of cells having the associated value, comprising determining, for the associated value for each cell of the set of cells, the representative data based on the one or more data entries from the set of data entries of the three-dimensional data, the representative data including a representative vector, comprising:
determining a set of vectors based on the one or more data entries from the set of data entries of the three-dimensional data;
generating an accumulated matrix comprising accumulating an outer product of each vector of the set of vectors with itself; and

extracting eigenvectors, eigenvalues, or both, from the accumulated matrix to determine the representative vector.

14. The system of claim 8 , wherein determining the vector based on the one or more points in the list of points comprises:
determining one or more vectors based on the one or more points in the list of points; and
determining the vector based on the one or more vectors.

15. At least one non-transitory computer-readable storage medium storing processor-executable instructions that, when executed by at least one computer hardware

Assignees

• Cognex Corp

Inventors

• Liu Zihan Hans
• Bogan Nathaniel
• Hoelscher Andrew

Classifications

• G06V30/2504

  Coarse or fine approaches, e.g. resolution of ambiguities or multiscale approaches · CPC title

• G06T7/73 Primary

  using feature-based methods · CPC title

• G06V20/64

  Three-dimensional [3D] objects · CPC title

• G06T1/0014

  Image feed-back for automatic industrial control, e.g. robot with camera (robots B25J19/023) · CPC title

• G06F17/16

  Matrix or vector computation {, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization (matrix transposition G06F7/78)} · CPC title