Sensing cargo using an imaging device

• US 20040125217A1
• Filed: 12/31/2002
• Published: 07/01/2004
• Est. Priority Date: 12/31/2002
• Status: Active Grant

• Current Assignee: Asset Intelligence LLC (PowerFleet, Inc.)
• Original Assignee: Asset Intelligence LLC (PowerFleet, Inc.)
• Inventors: Calver, Andrew J., Jesson, Joseph E., Cobello, Rick
• Granted Patent: US 7,746,379 B2
• Time in Patent Office: Days
• US Class Current: 348/231.3
• CPC Class Codes: G06Q 10/087 Inventory or stock management, e.g. order filling, procurement or balancing against orders

Abstract

A camera is used to sense cargo in a cargo space. An image of at least a portion of a cargo space is captured using the camera, and a digital signal processor classifies the image as representing an empty cargo space or a non-empty cargo space. Additionally, a type of cargo represented by the captured image may also be classified by the digital signal processor and an indication of the type of cargo output.

Claims

1. A method comprising:

capturing, using a light sensor, an image of at least a portion of a cargo space;

classifying, using a learning classifier, the captured image as representing either the presence of cargo or the absence of cargo; and

outputting, based on the classifying, an indication as to whether cargo is present in the cargo space.

2. A method as recited in claim 1, wherein the cargo space comprises an interior of a trailer.

3. A method as recited in claim 1, wherein capturing the image comprises capturing the image using an infrared (IR) imaging device.

4. A method as recited in claim 1, wherein capturing the image comprises capturing the image using an ultraviolet (UV) imaging device.

5. A method as recited in claim 1, wherein the learning classifier comprises a back-propagation of errors artificial neural network.

6. A method as recited in claim 1, wherein the learning classifier comprises a self-organizing Kohonen neural network.

7. A method as recited in claim 1, wherein the learning classifier comprises a relaxed-energy self-organizing artificial neural network.

8. A method as recited in claim 1, wherein the captured image is modified prior to the classifying.

9. A method as recited in claim 1, further comprising:

combining a plurality of captured images into a single image; and

wherein the classifying comprises using the single image as the captured image.

10. A method as recited in claim 1, wherein the classifying further comprises classifying a type of cargo present in the cargo space.

11. A system comprising:

a camera to capture an image of at least a portion of a cargo space; and

a digital signal processor, coupled to receive the image from the camera, to classify the image as representing an empty cargo space or a non-empty cargo space.

12. A system as recited in claim 11, wherein the digital signal processor is implemented in a single integrated circuit.

13. A system as recited in claim 11, wherein the digital signal processor comprises:

a classification module to extract characteristics from the image and determine, based on learned characteristics from training images, whether the image is to be classified as representing an empty cargo space or a non-empty cargo space.

14. A system as recited in claim 11, wherein the camera comprises an infrared (IR) imaging device.

15. A system as recited in claim 11, wherein the camera comprises an ultraviolet (UV) imaging device.

16. A system as recited in claim 11, wherein the digital signal processor implements a back-propagation of errors artificial neural network.

17. A system as recited in claim 11, wherein the digital signal processor implements a self-organizing Kohonen neural network.

18. A system as recited in claim 11, wherein the digital signal processor implements a relaxed-energy self-organizing artificial neural network.

19. A system as recited in claim 11, wherein the digital signal processor is further to classify a type of cargo present in the cargo space.

20. One or more computer readable media having stored thereon a plurality of instructions that, when executed by a processor, causes the processor to:

access at least a portion of an image of at least a portion of a cargo space, wherein the image was generated by sensing light at a particular frequency range;

classify the captured image as representing an empty cargo space or a non-empty cargo space; and

output, based on the classification, an indication as to whether the cargo space is empty or non-empty.

21. One or more computer readable media as recited in claim 20, wherein the particular frequency range comprises the infrared (IR) range.

22. One or more computer readable media as recited in claim 20, wherein the particular frequency range comprises the ultraviolet (UV) range.

23. A method comprising:

capturing, using a light sensor, an image of at least a portion of a cargo space;

classifying, based on one or more training images, a type of cargo represented by the captured image; and

outputting, based on the classifying, an indication of the type of cargo present in the cargo space.

24. A method as recited in claim 23, wherein the type of cargo comprises an amount of cargo.

25. A method as recited in claim 23, wherein the light sensor comprises an infrared (IR) imaging device.

26. A method as recited in claim 23, wherein the light sensor comprises an ultraviolet (UV) imaging device.

27. A method as recited in claim 23, wherein the classifying comprises classifying the type of cargo represented by the captured image using a back-propagation of errors artificial neural network.

28. A method as recited in claim 23, wherein the classifying comprises classifying the type of cargo represented by the captured image using a self-organizing Kohonen neural network.

29. A method as recited in claim 23, wherein the classifying comprises classifying the type of cargo represented by the captured image using a relaxed-energy self-organizing artificial neural network.

30. A system comprising:

a camera to capture an infrared (IR) or ultraviolet (UV) image of at least a portion of a cargo space; and

a digital signal processor implementing a learning classifier to classify the image as representing either the presence of cargo in the cargo space or the absence of cargo in the cargo space, wherein the digital signal processor is coupled to receive the image from the camera.

31. A system as recited in claim 30, wherein the learning classifier comprises a back-propagation of errors artificial neural network.

32. A system as recited in claim 30, wherein the learning classifier comprises a self-organizing Kohonen neural network.

33. A system as recited in claim 30, wherein the learning classifier comprises a relaxed-energy self-organizing artificial neural network.