Systems and methods for testing an automatic perception system

Patent metadata

Field	Value
Publication number	US-10332272-B2
Application number	US-201715670108-A
Country	US
Kind code	B2
Filing date	Aug 7, 2017
Priority date	Aug 7, 2017
Publication date	Jun 25, 2019
Grant date	Jun 25, 2019

Abstract

A method for testing a perception system is disclosed. The method includes detecting, by a processor, one or more objects in a composite image, the composite image including a first image and at least a portion of a second image. A first set of objects are detected in the first image. Further, the method includes comparing, by the processor, a second set of objects of the one or more objects detected in the portion of the second image with a set of previously detected objects in the portion of the second image. Furthermore, the method includes validating, the performance of the processor in the detection of the first set of objects in the first image by ensuring the second set of objects detected in the at least a portion of a second image matches the set of previously detected objects in the portion of the second image.

First claim

What is claimed is:

1. A method for testing a perception system, the method comprising:
detecting, by a processor, objects in a composite image,
the composite image including a first image and at least a portion of a second image, and
the objects including:
a first set of objects detected in the first image, and
a second set of objects detected in the portion of the second image;

comparing, by the processor, the second set of objects, in the portion of the second image of the composite image, with a set of previously detected objects that were detected in the portion of the second image before detecting the objects in the composite image; and
validating performance of the processor in the detecting of the first set of objects, in the first image of the composite image, based on whether the second set of objects match the set of previously detected objects.

2. The method of claim 1 , further comprising:
sending a signal indicating an improper performance of the processor in the detecting of the first set of objects, in the first image of the composite image, when the second set of objects docs does not match the set of previously detected objects.

3. The method of claim 1 , further comprising:
receiving, by the processor, the first image from an image capturing device.

4. The method of claim 1 , further comprising:
appending, by the processor, the portion of the second image to the first image to create the composite image.

5. The method of claim 1 ,
wherein the composite image further includes a portion of a third image, and
wherein the third image includes one or more previously detected objects.

6. The method of claim 1 , wherein the set of previously detected objects were previously detected by the processor.

7. The method of claim 1 , wherein the composite image further comprises a portion of the first image appended to the first image.

8. The method of claim 1 ,
wherein the first image corresponds to an image frame of a first video stream,
wherein the portion of the second image corresponds to a portion of an image frame of a second video stream,
wherein the first video stream corresponds to a real-time video being captured by an image capturing device, and
wherein the second video stream corresponds to a historical video stream previously captured by the image capturing device.

9. The method of claim 8 , further comprising:
tracking, by the processor, the first set of objects and the second set of objects in the first video stream and the second video stream, respectively.

10. The method of claim 9 , further comprising:
comparing, by the processor, the tracking of the second set of objects with tracking of the set of previously detected objects in the second video stream.

11. The method of claim 10 , further comprising:
validating, by the processor, the tracking of the first set of objects in the first video stream based on the comparing of the tracking of the second set of objects with the tracking of the set of previously detected objects.

12. An image capturing device comprising:
an image sensor;
a processor, coupled to the image sensor, configured to:
detect objects in a composite image,
the composite image including a first image and at least a portion of a second image, and
the objects including:
a first set of objects detected in the first image, and
a second set of objects detected in the portion of the second image;

compare the second set of objects, in the portion of the second image of the composite image, with a set of previously detected objects that were detected in the portion of the second image before detecting the objects in the composite image; and
validate a performance of the processor in detecting the first set of objects, in the first image of the composite image, based on whether the second set of objects match the set of previously detected objects.

13. The image capturing device of claim 12 , wherein the processor is further configured to:
send a signal indicating an improper performance of the processor in the detecting of the first set of objects, in the first image of the composite image, when the second set of objects do not match the set of previously detected objects.

14. The image capturing device of claim 12 , wherein the processor is further configured to:
append the portion of the second image to the first image to create the composite image.

15. The image capturing device of claim 12 ,
wherein the composite image further includes a portion of a third image, and
wherein the third image includes one or more previously detected objects.

16. The image capturing device of claim 12 , wherein the set of previously detected objects were previously detected by the processor.

17. The image capturing device of claim 12 ,
wherein the first image corresponds to an image frame of a first video stream,
wherein the portion of the second image corresponds to a portion of an image frame of a second video stream,
wherein the first video stream corresponds to a real-time video being captured by the image capturing device, and
wherein the second video stream corresponds to a historical video stream previously captured by the image capturing device.

18. The image capturing device of claim 17 , wherein the processor is further configured to:
track the first set of objects and the second set of objects in the first video stream and the second video stream, respectively.

19. The image capturing device of claim 18 , wherein the processor is further configured to:
compare the tracking of the second set of objects with tracking of the set of previously detected objects in the second video stream.

20. The image capturing device of claim 19 , wherein the processor is further configured to:
validate the tracking of the first set of objects in the first video stream based on the comparing of the tracking of the second set of objects with the tracking of the set of previously detected objects.

Assignees

• Caterpillar Inc

Inventors

• Mianzo Lawrence Andrew
• Forcash Joseph
• Oblak Tod Andrew
• Stringer Jeffrey T
• Hartman Mark
• Carter Jr Jeffrey J

Classifications

• G06T7/246 Primary

  using feature-based methods, e.g. the tracking of corners or segments · CPC title

• G06T7/70 Primary

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

• G06F18/2411

  based on the proximity to a decision surface, e.g. support vector machines · CPC title

• G06T7/251

  involving models · CPC title

• G01V8/10

  Detecting, e.g. by using light barriers (by reflection from the object G01S17/00) · CPC title