Method and system for automatic focusing for high-resolution structured light 3D imaging

Patent metadata

Field	Value
Publication number	US-11512946-B2
Application number	US-202117159263-A
Country	US
Kind code	B2
Filing date	Jan 27, 2021
Priority date	Feb 17, 2020
Publication date	Nov 29, 2022
Grant date	Nov 29, 2022

Abstract

The disclosure provides an improvement to digital fringe projection techniques in which the optimal focal length settings are automatically determined for reconstructing a 3D profile. In a pre-calibration phase, geometric parameters of the system are calibrated using a few discrete focal length settings. These discretely calibrated geometric parameters are fitted onto a continuous function model. In a 3D autofocusing phase, a set of optimal focal length settings for a scene are determined using a 2D autofocusing technique. Calibrated geometric parameters for each optimal focal length setting are automatically calculated using the continuous geometric parameter model. Finally, a 3D profile of objects in the scene is reconstructed using the calibrated geometric parameters for each optimal focal length setting.

First claim

What is claimed is:

1. A method for measuring a three-dimensional profile of at least one object using the digital fringe projection system including a camera and a projector, the method comprising:
determining, with a processor, an optimal value for a focal length setting of the camera for capturing in-focus images of the at least one object with the camera, the camera having an electrically tunable lens;
determining, with the processor, calibrated values for a plurality of geometric parameters of the digital fringe projection system based on (i) a geometric model of the digital fringe projection system and (ii) the optimal value for the focal length setting of the camera, the plurality of geometric parameters being continuous functions of the focal length setting of the camera in the geometric model;
projecting, with the projector, structured light onto the at least one object to generate fringe patterns on a surface of the at least one object;
capturing, with the camera operated with the optimal value for the focal length setting of the camera, a first image of the surface of the at least one object that includes the fringe patterns on the surface of the at least one object; and
determining, with the processor, a three-dimensional profile of the surface of the at least one object based on (i) the fringe patterns captured in the first image and (ii) the calibrated values for the plurality of geometric parameters.

2. The method according to claim 1 , the determining the optimal value for the focal length setting of the camera further comprising:
determining, with the processor, a plurality of optimal values for the focal length setting of the camera for capturing in-focus images of the at least one object with the camera.

3. The method according to claim 2 , the capturing the first image of the surface of the at least one object further comprising:
for each respective optimal value in the plurality of optimal values for the focal length setting of the camera, capturing, with the camera operated with the respective optimal value for the focal length setting of the camera, a respective first image of the surface of the at least one object that includes the fringe patterns on the surface of the at least one object.

4. The method according to claim 3 , the determining the calibrated values for the plurality of geometric parameters of the digital fringe projection system further comprising:
for each respective optimal value in the plurality of optimal values for the focal length setting of the camera, determining, with the processor, respective calibrated values for the plurality of geometric parameters of the digital fringe projection system based on (i) the geometric model of the digital fringe projection system and (ii) the respective optimal value for the focal length setting of the camera.

5. The method according to claim 4 , the determining the three-dimensional profile of the surface of the at least one object further comprising:
for each respective optimal value in the plurality of optimal values for the focal length setting of the camera, determining, with the processor, a respective portion of the three-dimensional profile of the surface of the at least one object based on (i) the fringe patterns captured in the respective first image captured using the respective optimal value for the focal length setting of the camera and (ii) the corresponding respective calibrated values for the plurality of geometric parameters.

6. The method according to claim 2 , the determining the plurality of optimal values for the focal length setting of the camera further comprising:
capturing, with the camera, a plurality of second images of the surface of the at least one object before the projecting of the structured light onto the at least one object, each image in the plurality of second images being captured with a different one of a plurality of test values for the focal length setting of the camera; and
determining, with the processor, the plurality of optimal values for the focal length setting of the camera as a subset of the plurality of test values for the focal length setting of the camera based on the plurality of second images.

7. The method according to claim 6 , the determining the plurality of optimal values for the focal length setting of the camera further comprising:
determining, with the processor, a focal length setting map, the focal length setting map being a two-dimensional array having values corresponding to individual pixel coordinates within the plurality of second images, each value of the focal length setting map indicating a focal length setting selected from the plurality of test values for the focal length setting of the camera that makes pixels at the corresponding pixel coordinate most in-focus.

8. The method according to claim 7 , the determining the focal length setting map further comprising:
for each respective pixel coordinate within the plurality of second images, determining the focal length setting selected from the plurality of test values for the focal length setting of the camera that makes pixels at the respective pixel coordinate most in-focus by evaluating a magnitude of an image gradient at the respective pixel coordinate of each image in the plurality of second images.

9. The method according to claim 7 , the determining the plurality of optimal values for the focal length setting of the camera further comprising:
determining, with the processor, the plurality of optimal values for the focal length setting of the camera as the subset of the plurality of test values for the focal length setting of the camera that appear in the focal length setting map.

10. The method according to claim 7 , the determining the three-dimensional profile of the surface of the at least one object further comprising:
for each respective optimal value in the plurality of optimal values for the focal length setting of the camera, determining, with the processor, a respective portion of the three-dimensional profile of the surface of the at least one object, the respective portion corresponding to pixels at pixel coordinate in the focal length setting map indicating the respective optimal value for the focal length setting of the camera.

11. The method according to claim 1 further comprising:
determining, with the processor, an optimal value for a focal length setting of the projector based on the optimal value for the focal length setting of the camera, the projector having an electrically tunable lens.

12. The method according to claim 1 further comprising:
determining, with the processor, the geometric model of the digital fringe projection system including a respective continuous function for each respective geometric parameter in the plurality of geometric parameters, each respective continuous function being a function of the focal length setting of the camera.

13. The method according to claim 12 , the determining the geometric model of the digital fringe projection system further comprising:
capturing, with the camera, a plurality of third images of a reference object using a plurality of test values for the focal length setting of the camera; and
determining, with the processor, the respective continuous function for each respective geometric parameter in the plurality of geometric parameters based on the plurality of third images.

14. The method according to claim 13 , wherein for each respective test value in the plurality of test values, the plurality of third images includes images captured using the respective test value for the focal length setting of the camera with the reference object i

Assignees

• Purdue Research Foundation

Inventors

• Zhang Song
• Hu Xiaowei
• Wang Guijin

Classifications

• G01B11/2504 Primary

  Calibration devices · CPC title

• G01B11/002

  for measuring two or more coordinates · CPC title

• G01B11/254 Primary

  Projection of a pattern, viewing through a pattern, e.g. moiré · CPC title

• G02B7/09

  adapted for automatic focusing or varying magnification · CPC title