Detecting objects with a depth sensor

• US 8,831,285 B2
• Filed: 07/26/2012
• Issued: 09/09/2014
• Est. Priority Date: 07/26/2012
• Status: Active Grant

• Current Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Kang, Jinman
• Primary Examiner(s): AZARIAN, SEYED H
• Application Number:
  US13/558,538
• Publication Number:
  US 20140029788A1
• Time in Patent Office: 775 Days
• Field of Search: 382/100, 382/103, 382106-107, 382/113, 382/162, 382/168, 382/173, 382/177, 382/181, 382212-213, 382219-220, 382/232, 382/123, 382/254, 382/276, 382286-295, 382/305, 382/312, 382/104, 382/154, 382/132, 348/661, 348/708
• US Class Current: 382/103
• CPC Class Codes: G06F 2218/04 Denoising
• CPC Class Codes: G06V 10/803 of input or preprocessed data
• CPC Class Codes: G06V 10/84 using probabilistic graphical models from image or video features, e.g. Markov models or Bayesian networks

Abstract

Detecting an object includes receiving depth data and infrared (IR) data from a depth sensor. A first background subtraction is performed on the IR data to create a first mask, and a second background subtraction is performed on the IR data to create a second mask. The first and second masks and the depth data are merged to create a third mask.

Claims

1. A method, comprising:

receiving depth data and infrared (IR) data from a depth sensor;

performing a first background subtraction on the IR data to create a first mask;

performing a second background subtraction on the IR data to create a second mask;

merging the first and second masks and the depth data to create a third mask, wherein merging the first and second masks and the depth data to create a third mask includes an adaptive weighted average process; and

detecting an object using the third mask.

2. The method of claim 1, wherein at least one of the first and second background subtractions is an Adaptive Gaussian Mixture Model background subtraction.

3. The method of claim 1, wherein at least one of the first and second background subtractions is an accumulated weight frame difference background subtraction.

4. The method of claim 1, further comprising:

creating a depth map from the depth data.

5. The method of claim 4, wherein creating a depth map includes

monitoring a tangible reference at a reference distance from the depth sensor;

and

maintaining a virtual reference approximately on calibrated three dimensional coordinates derived from said reference distance.

6. The method of claim 1, further comprising:

rescaling the IR data into gray scaled frames.

7. A system, comprising:

an input to receive depth data and infrared (IR) data from a depth sensor;

an object segmentation module to;

perform a first background subtraction on the IR data to create a first mask;

perform a second background subtraction on the IR data to create a second mask;

merge the first and second masks and the depth data to create a third mask, wherein the object segmentation module is to merge the first and second masks and the depth data to create a third mask using at least an adaptive weighted average process; and

an output to output the third mask for detecting an object using the third mask.

8. The system of claim 7, further comprising:

a memory accessible by the processor, the memory to store depth data and IR data received from a depth sensor.

9. The system of claim 7, further comprising:

a depth sensor to obtain and send the depth data and IR data to the processor.

10. The system of claim 7, wherein at least one of the first and second background subtractions is an Adaptive Gaussian Mixture Model background subtraction.

11. The system of claim 7, wherein at least one of the first and second background subtractions is an accumulated weight frame difference background subtraction.

12. A non-transitory computer readable storage medium storing instructions that when executed cause a system to:

receive depth data and infrared (IR) data from a depth sensor;

perform a first background subtraction on the IR data to create a first mask;

perform a second background subtraction on the IR data to create a second mask; and

merge the first and second masks and the depth data to create a third mask, wherein the instructions to merge the first and second masks and the depth data to create a third mask include instructions to conduct an adaptive weighted average process.

13. The computer readable storage medium of claim 12, further comprising instructions to:

detect an object using the third mask.

14. The computer readable storage medium of claim 12, wherein at least one of the first and second background subtractions is an Adaptive Gaussian Mixture Model background subtraction.

15. The computer readable storage medium of claim 12, wherein at least one of the first and second background subtractions is an accumulated weight frame difference background subtraction.

16. The computer readable storage medium of claim 12, further comprising instructions to:

create a depth map from the depth data.

17. The computer readable storage medium of claim 16, wherein the instructions to create a depth map include instructions to:

monitor a tangible reference at a reference distance from the depth sensor; and

maintain a virtual reference approximately on calibrated three dimensional coordinates derived from said reference distance.

18. The computer readable storage medium of claim 12, further comprising instructions to:

rescale the IR data into gray scaled frames.