METHOD AND DEVICE FOR STEREOSCOPIC VISION

• US 20180227564A1
• Filed: 02/06/2018
• Published: 08/09/2018
• Est. Priority Date: 02/06/2017
• Status: Active Grant

• Current Assignee: Robotemi Limited
• Original Assignee: Roboteam Home Ltd.
• Inventors: WOLF, Yosef Arie, GOREN, Gal, VITZRABIN, Efraim, GILAD, Gil
• Granted Patent: US 10,567,732 B2
• Time in Patent Office: Days
• CPC Class Codes: H04N 13/15 for colour aspects of image signals
• CPC Class Codes: H04N 13/239 using two 2D image sensors having a relative position equal to or related to the interocular distance H04N13/243 takes precedence
• CPC Class Codes: H04N 13/254 in combination with electromagnetic radiation sources for illuminating objects
• CPC Class Codes: H04N 13/271 wherein the generated image signals comprise depth maps or disparity maps
• CPC Class Codes: H04N 2013/0081 Depth or disparity estimation from stereoscopic image signals
• CPC Class Codes: H04N 23/11 for generating image signals from visible and infrared light wavelengths

Abstract

A stereoscopic vision device, a method and a robot, the stereoscopic vision device comprising: a first image capture device and a second image capture device configured to capture a light pattern formed in an environment of the device by a light source configured to emit NIR light at a multiplicity of directions, each of the first image capture device and the second image capture device comprising a filter configured to transmit at least 75 percent of the NIR light, and to transmit at most 50 percent of visible light; and a processor configured to: determine depth information of the environment from a first image captured by the first image capture device and a second image captured by the second image capture device, based on the light pattern as depicted in the first image and the second image; and determine color information of the environment the first image and the second image.

Claims

1. A stereoscopic vision device, comprising:

a first image capture device and a second image capture device each configured to capture a light pattern formed in an environment of the device by a light source configured to emit near infrared (NIR) light at a multiplicity of directions, each of the first image capture device and the second image capture device comprising a filter configured to transmit at least 75 percent of the NIR light in a NIR wavelength of at least 850 nanometers, and to transmit at most 50 percent of visible light; and

a processor configured to;

determine depth information of the environment from a first image captured by the first image capture device and a second image captured by the second image capture device, based on the light pattern depicted in the first image and the second image; and

determine color information of the environment based on the first image and the second image.

2. The device of claim 1, wherein the NIR light has a wavelength of at least 850 nanometers.

3. The device of claim 1, wherein the visible light has a wavelength range of 400-650 nanometers.

4. The device of claim 1, wherein the filter is configured to transmit at least 85 percent of the NIR light at a wavelength of at least 850 nanometers.

5. The device of claim 1, wherein the filter is configured to transmit at most between 10 and 40 percent of the visible light at a wavelength range of 400-650 nanometers.

6. The device of claim 1, further comprising the light source configured to emit near infrared (NIR) light at a multiplicity of directions.

7. The device of claim 1, wherein the multiplicity of directions comprise selectively emitting the NIR light in the light pattern.

8. The device of claim 1, wherein the depth information is determined by (i) identifying the light pattern depicted in the first image and in the second image, and (ii) registration of the light pattern depicted in the first image with the the light pattern depicted in the second image.

9. A method for obtaining stereoscopic data of an environment, comprising:

emitting NIR light at a multiplicity of directions, thus forming a light pattern in the environment;

obtaining a first image captured by a first image capture device and a second image captured by a second capture device, the first image and the second image each depicting at least a part of the light pattern,

wherein each of the first image capture device and the second image capture device capture the first image and the second image, respectively, through a filter configured to transmit at least 75 percent of the NIR light, and to transmit at most 50 percent of visible light;

identifying the light pattern depicted in each of the first image and the second image;

registering the first image and the second image in accordance with the light pattern;

determining depth information for each light point comprised in the light pattern depicted in the registered first and second images; and

determining color information of the environment from the first image and in the second image.

10. The method of claim 9, wherein the NIR light has a wavelength of at least 850 nanometers.

11. The method of claim 9, wherein the visible light has a wavelength range of 400-650 nanometers.

12. The method of claim 9, wherein the filter is configured to transmit at least 85 percent of the NIR light at a wavelength of at least 850 nanometers.

13. The method of claim 9, wherein the filter is configured to transmit at most between 10 and 40 percent of the visible light at a wavelength range of 400-650 nanometers.

14. The method of claim 9, wherein the depth information is determined by comparing locations of points in the light pattern as depicted in the first image and in the second image.

15. The method of claim 9, wherein the multiplicity of directions comprise selectively emitting the NIR light in the light pattern.

16. A robot comprising:

a light source configured to emit a pattern of NIR light at a multiplicity of directions, thus forming a light pattern in an environment of the robot;

a first image capture device and a second image capture device, each comprising a filter configured to transmit at least 75 percent of the NIR light, and to transmit at most 50 percent of visible light, the first image capture device and a second image capture device configured to each capture at least a part of the light pattern; and

a processor configured to;

determine depth information in the environment from a first image captured by the first image capture device and a second image captured by the second image capture device, based on the at least part of the light pattern as depicted in each of the first image and the second image; and

determine color information of at least one object from the first image and the second image;

a steering mechanism or changing a position of the robot in accordance with the at least one object; and

a motor for activating the steering mechanism.

17. The robot of claim 16, wherein the filter is configured to transmit at least 85 percent of the NIR light at a wavelength of at least 850 nanometers.

18. The robot of claim 16, wherein the filter is configured to transmit at most 40 percent of the visible light at a wavelength range of 400-650 nanometers.

19. The robot of claim 16, wherein the depth information is determined by comparing locations of points in the light pattern as depicted in the first image and in the second image.

20. The robot of claim 16, wherein the depth information is determined by (i) identifying the light pattern in each of the first image and the second image, and (ii) registration of the light pattern depicted in the first image with the the light pattern depicted in the second image.