Method and system for creating a three dimensional representation of an object

• US 9,204,130 B2
• Filed: 02/06/2013
• Issued: 12/01/2015
• Est. Priority Date: 02/06/2013
• Status: Active Grant

• Current Assignee: Caterpillar Incorporated
• Original Assignee: Caterpillar Incorporated
• Inventors: Pawelski, Paul C.
• Primary Examiner(s): Tran, Trang U
• Application Number:
  US13/760,355
• Publication Number:
  US 20140218477A1
• Time in Patent Office: 1,028 Days
• Field of Search: 348/46, 348/48, 348/50, 348/51
• US Class Current: 1/1
• CPC Class Codes: H04N 13/204 using stereoscopic image cameras stereoscopic photography G03B35/00
• CPC Class Codes: H04N 13/254 in combination with electromagnetic radiation sources for illuminating objects

Abstract

A method and a system for creating three dimensional representation of an object are provided. The method includes projecting a pattern image comprising one or more pattern elements and capturing an image of the object with the pattern image projected onto the object. The method includes extracting intensity values corresponding to at least two color components of each pixel of the captured image. For each color component, a probability map is generated using the corresponding intensity values. The probability map for a color component represents a probability that a pixel of the captured image is representing a part of a pattern element. A joint probability distribution is determined by combining the probability maps. Using the joint probability distribution, specific pattern elements are located in the captured image. The method includes determining three dimensional coordinates representing points on surface of the object based on the location of the specific pattern elements.

Claims

1. A method comprising:

projecting a pattern image on an object by a projector, the pattern image comprising one or more pattern elements;

capturing an image of the object with the pattern image projected onto the object;

extracting intensity values corresponding to at least two color components of each pixel of the captured image;

generating a probability map for each color component using the corresponding extracted intensity values, the probability map for a color component representing a probability that a pixel of the captured image is representing a part of a pattern element;

determining a joint probability distribution using the probability maps for the color components, the joint probability distribution representing a probability that a pixel of the captured image is representing part of any pattern element;

locating specific pattern elements in the captured image using the joint probability distribution; and

determining three dimensional coordinates representing points on surface of the object based, at least in part, on the location of the specific pattern elements.

2. The method of claim 1, further comprising:

projecting a blank image on the object;

capturing a reference image of the object corresponding to the projected blank image;

extracting intensity values corresponding to the at least two color components of each pixel of the reference image; and

generating a component image for each color component by performing pixel by pixel subtraction of intensity values of the reference image from the corresponding intensity values of the captured image, wherein generating the probability map for each color component comprises generating the probability map using intensity values of the component image.

3. The method of claim 1, wherein the color components are obtained using one of RGB color model or CMYK color model.

4. The method of claim 1, wherein the pattern image includes at least one of an array of dots, an array of stripes, an array of rectangles, and a grid.

5. The method of claim 1, wherein a timing associated with the capture of the image is synchronized with a timing associated with the projection of the pattern image.

6. The method of claim 1, wherein the determining the joint probability distribution includes combining the probability maps using Bayesian joint probability distribution.

7. The method of claim 1, wherein the three dimensional coordinates are determined using one of a lookup table or a mathematical formula for translating the location of the specific pattern elements into the three dimensional coordinates.

8. A system for generating a three dimensional (3D) representation of an object, the system comprising:

a projector configured to project a pattern image on an object, the pattern image comprising one or more pattern elements;

an image capturing module configured to capture an image of the object with the pattern image projected onto the object; and

a controller coupled to the image capturing module and the projector, the controller is configured to;

extract intensity values corresponding to at least two color components of each pixel of the captured image;

generate a probability map for each color component using the corresponding extracted intensity values, the probability map for a color component representing a probability that a pixel of the captured image is representing a part of a pattern element;

determine a joint probability distribution using the probability maps for the color components, the joint probability distribution representing a probability that a pixel of the captured image is representing part of any pattern element;

locate specific pattern elements in the captured image using the joint probability distribution; and

determine three dimensional coordinates representing points on surface of the object based, at least in part, on the location of the specific pattern elements.

9. The system of claim 8 further comprising a mount, which is attached to the projector and the image capturing module so as to hold the image capturing module in a fixed spatial relation to the projector.

10. The system of claim 8, wherein the projector is a liquid crystal display (LCD) projector.

11. The system of claim 8, wherein the image capturing module is a charge coupled device (CCD) camera.

12. The system of claim 8, wherein:

the projector is further configured to project a blank image on the object;

the image capturing module is further configured to capture a reference image of the object corresponding to the projected blank image;

the controller is further configured to;

extract intensity values corresponding to at least two color components of each pixel of the reference image; and

generate a component image for each color component by performing pixel by pixel subtraction of intensity values of the reference image from the corresponding intensity values of the captured image, wherein the controller generates the probability map for each color component using intensity values of the component image.

13. The system of claim 8, wherein the projector is configured to project the pattern image including at least one of an array of dots, an array of stripes, an array of rectangles, and a grid.

14. The system of claim 8 further comprising a timer configured to synchronize a timing associated with the capture of the image with a timing associated with the projection of the pattern image.

15. The system of claim 8, wherein the controller is configured to determine three dimensional coordinates using one of a lookup table or a mathematical formula for translating the location of the specific pattern elements into the three dimensional coordinates.

16. The system of claim 8, wherein the controller is configured to determine the joint probability distribution by combining the probability maps using Bayesian joint probability distribution.