Machine vision for ego-motion, segmenting, and classifying objects

• US 10,157,469 B2
• Filed: 07/30/2018
• Issued: 12/18/2018
• Est. Priority Date: 04/13/2015
• Status: Active Grant

• Current Assignee: Samsung Semiconductor Incorporated (Samsung Electronics Co. Ltd.)
• Original Assignee: Samsung Electronics Co. Ltd.
• Inventors: Smits, Gerard Dirk
• Primary Examiner(s): Wang, Carol
• Application Number:
  US16/049,380
• Publication Number:
  US 20180336686A1
• Time in Patent Office: 141 Days
• CPC Class Codes: G01P 3/36 Devices characterised by the use of optical means, e.g. using infrared, visible, or ultraviolet light G01P3/68 takes precedence; gyrometers using the Sagnac effect, i.e. rotation-induced shifts between counter-rotating electromagnetic beams G01C19/64
• CPC Class Codes: G06T 7/20 Analysis of motion motion estimation for coding, decoding, compressing or decompressing digital video signals H04N19/43, H04N19/51
• CPC Class Codes: H04N 13/254 in combination with electromagnetic radiation sources for illuminating objects
• CPC Class Codes: H04N 23/56 provided with illuminating means
• CPC Class Codes: H04N 23/71 Circuitry for evaluating the brightness variation
• CPC Class Codes: H04N 5/04 Synchronising for television systems using pulse code modulation H04N7/56

Abstract

Systems and methods for machine vision are presented. Such machine vision includes ego-motion, as well as the segmentation and/or classification of image data of one or more targets of interest. The projection and detection of scanning light beams that generate a pattern are employed. Real-time continuous and accurate spatial-temporal 3D sensing is achieved. The relative motion between an observer and a projection surface is determined. A combination of visible and non-visible patterns, as well as a combination of visible and non-visible sensor arrays is employed to sense 3D coordinates of target features, as well as acquire color image data to generate 3D color images of targets. Stereoscopic pairs of cameras are employed to generate 3D image data. Such cameras are dynamically aligned and calibrated. Information may be encoded in the transmitted patterns. The information is decoded upon detection of the pattern and employed to determine features of the reflecting surface.

Claims

1. A system for identifying motion of a target, comprising:

one or more transmitters for transmitting one or more light beams at the target;

one or more receivers for receiving reflection of the one or more light beams from the target;

one or more memory devices for storing instructions; and

one or more processor devices that execute the instructions that enable actions, including;

employing detection of one or more reflected light beams by the one or more receivers to employ the one or more transmitters to modulate the one or more light beams transmitted onto the target;

employing the one or more receivers to generate one or more signals based on detection of the reflection of the one or more light beams;

employing the one or more transmitters to dynamically modify an illumination intensity of the one or more light beams based on a reflection of one or more light patterns by different detected textures of one or more surfaces of the target; and

providing one or more characteristics of a relative velocity between the target and the one or more receivers based on the one or more signals.

2. The system of claim 1, wherein the one or more transmitters transmit a rotating pattern onto the target that is controlled by the modulation of the one or more light beams.

3. The system of claim 1, further comprising employing the one or more signals to determine one or more of a spatial elongation or a spatial contraction of the one or more reflected light beams in one or more directions.

4. The system of claim 1, further comprising:

employing the one or more transmitters to encode information in the one or more light beams, wherein the encoding includes one or more of spatially encoded information or temporally encoded information; and

employing the one or more receivers to decode the encoded information in the one or more reflected light beams to identify one or more properties of the target.

5. The system of claim 1, further comprising dynamically aligning two receivers to detect the one or more reflected light beams to provide a stereoscopic image of the target.

6. The system of claim 1, further comprising employing the one or more receivers to detect a reflected combination of visible and non-visible light beams transmitted by the one or more transmitters to the target, wherein the reflection of these light beams from the target are used to determine three dimensional coordinates of the target.

7. A method for identifying motion of a target, comprising:

employing detection of one or more reflected light beams from the target, by one or more receivers, to employ one or more transmitters to modulate one or more light beams transmitted onto the target;

employing the one or more receivers to generate one or more signals based on the detection of the one or more reflected light beams from the target;

employing the one or more transmitters to dynamically modify an illumination intensity of the one or more light beams based on a reflection of one or more light patterns by different detected textures of one or more surfaces of the target; and

providing one or more characteristics of a relative velocity between the target and the one or more receivers based on the one or more signals.

8. The method of claim 7, wherein the one or more transmitters transmit a rotating pattern onto the target that is controlled by the modulation of the one or more light beams.

9. The method of claim 7, further comprising employing the one or more signals to determine one or more of a spatial elongation or a spatial contraction of the one or more reflected light beams in one or more directions.

10. The method of claim 7, further comprising:

employing the one or more transmitters to encode information in the one or more light beams, wherein the encoding includes one or more of spatially encoded information or temporally encoded information; and

employing the one or more receivers to decode the encoded information in the one or more reflected light beams to identify one or more properties of the target.

11. The method of claim 7, further comprising dynamically aligning two receivers to detect the one or more reflected light beams to provide a stereoscopic image of the target.

12. The method of claim 7, further comprising employing the one or more receivers to detect a reflected combination of visible and non-visible light beams transmitted by the one or more transmitters to the target, wherein the reflection of these light beams from the target are used to determine three dimensional coordinates of the target.

13. A processor readable non-transitory storage media that includes instructions for identifying motion of a target, wherein the execution of the instructions by one or more processors enable actions, comprising:

employing detection of one or more reflected light beams from the target, by one or more receivers, to employ one or more transmitters to modulate one or more light beams transmitted onto the target;

employing the one or more receivers to generate one or more signals based on the detection of the one or more reflected light beams from the target;

employing the one or more transmitters to dynamically modify an illumination intensity of the one or more light beams based on a reflection of one or more light patterns by different detected textures of one or more surfaces of the target; and

providing one or more characteristics of a relative velocity between the target and the one or more receivers based on the one or more signals.

14. The media of claim 13, wherein the one or more transmitters transmit a rotating pattern onto the target that is controlled by the modulation of the one or more light beams.

15. The media of claim 13, further comprising:

employing the one or more transmitters to encode information in the one or more light beams, wherein the encoding includes one or more of spatially encoded information or temporally encoded information; and

employing the one or more receivers to decode the encoded information in the one or more reflected light beams to identify one or more properties of the target.

16. The media of claim 13, further comprising dynamically aligning two receivers to detect the one or more reflected light beams to provide a stereoscopic image of the target.

17. The method of claim 13, further comprising employing the one or more receivers to detect a reflected combination of visible and non-visible light beams transmitted by the one or more transmitters to the target, wherein the reflection of these light beams from the target are used to determine three dimensional coordinates of the target.