Enhanced contrast for object detection and characterization by optical imaging

Patent metadata

Field	Value
Publication number	US-9626591-B2
Application number	US-201314106148-A
Country	US
Kind code	B2
Filing date	Dec 13, 2013
Priority date	Jan 17, 2012
Publication date	Apr 18, 2017
Grant date	Apr 18, 2017

Abstract

Enhanced contrast between an object of interest and background surfaces visible in an image is provided using controlled lighting directed at the object. Exploiting the falloff of light intensity with distance, a light source (or multiple light sources), such as an infrared light source, can be positioned near one or more cameras to shine light onto the object while the camera(s) capture images. The captured images can be analyzed to distinguish object pixels from background pixels.

First claim

What is claimed is:

1. A method for analyzing captured images, the method comprising: capturing a first plurality of digital images of a field of view using a camera and at least one light source to illuminate the field of view at one or more illumination powers, the capturing of the first plurality of digital images including: capturing a first digital image when the at least one light source is at a first illumination power; capturing a second digital image when the at least one light source is at a second illumination power; and capturing a third digital image when the at least one light source is at the first illumination power; identifying pixels of the first plurality of digital images that correspond to an object rather than to a background, wherein the pixels corresponding to the object are identified based on a difference between the second digital image and the first digital image and a difference between the second digital image and the third digital image; and wherein the identifying of the pixels of the first plurality of digital images further comprises: forming a first difference image based upon the difference between the second digital image and the first digital image; forming a second difference image based upon the difference between the second digital image and the third digital image; and determining a third difference image, wherein each pixel of the third difference image comprises a pixel of a smaller pixel value between a corresponding pixel in the first difference image and a corresponding pixel in the second difference image and constructing a model of the object based on the identified pixels, the model including a position and a shape of the object.

2. The method of claim 1 , wherein the capturing of the first plurality of digital images comprises:
pulsing the at least one light source from the first illumination power to the second illumination power.

3. The method of claim 2 , wherein the pulsing comprises:
pulsing the at least one light source at regular intervals while the camera captures the first plurality of digital images, such that a light intensity is stronger for object pixels than for background pixels.

4. The method of claim 3 , wherein the identifying of the pixels of the first plurality of digital images comprises:
comparing pixels in successive digital images to distinguish the object pixels from the background pixels.

5. The method of claim 1 , wherein the pixel value corresponds to a brightness level, such that each pixel of the third difference image comprises one pixel from the corresponding pixels of the first difference image and the second difference image having a smaller brightness level.

6. The method of claim 5 , further comprising:
using a threshold to distinguish foreground pixels and background pixels.

7. The method of claim 1 , wherein the identifying of the pixels of the first plurality of digital images comprises:
identifying pixels that correspond to a hand rather than to a ceiling.

8. The method of claim 1 , wherein the identifying of the pixels of the first plurality of digital images comprises:
sending a reduced representation of each digital image of the first plurality of digital images which identified background pixels are zeroed out.

9. The method of claim 1 , wherein the constructing of the model comprises:
constructing a 3D model.

10. The method of claim 1 , further comprising:
capturing a second plurality of digital images;
identifying pixels of the second plurality of digital images that correspond to the object rather than to the background; and
detecting a motion of the object based at least in part upon the identified pixels of the first plurality of digital images and the identified pixels of the second plurality of digital images.

11. The method of claim 10 , wherein the detecting of the motion comprises:
comparing a detected motion to a library; and
enabling a particular gesture associated with the detected motion.

12. The method of claim 1 , wherein the capturing of the first plurality of digital images further comprises:
adjusting an intensity of the at least one light source, such that the pixels corresponding to the object are saturated due to a brightness level of the object.

13. The method of claim 12 , wherein the identifying comprises:
identifying the saturated pixels as pixels that might correspond to the object.

14. The method of claim 1 , wherein the capturing of the first plurality of digital images further comprises:
filtering light outside a band around a peak frequency of the at least one light source.

15. The method of claim 1 , wherein the identifying comprises:
detecting in at least one digital image, of the first plurality of digital images, a Gaussian brightness falloff pattern indicative of a rounded object.

16. The method of claim 1 , wherein the constructing of the model further comprises:
reconstructing a position of the object from an outline of shape of the object, as seen from a particular vantage point, and defined by the identified pixels of the first plurality of digital images, the reconstruction of the position including:
defining a plurality of lines tangent to the object in the first plurality of digital images from the particular vantage point in planes, referred to herein as “slices”, from the particular vantage point to the object in a given slice; and
determining the position and an approximate cross-section of the object in the slice using one or more simple closed curves.

17. A wearable goggle, comprising:
at least one camera oriented toward a field of view;
at least one light source disposed on a same side of the field of view as the at least one camera and oriented to illuminate the field of view; and
a processor coupled to the at least one camera and the at least one light source and configured to:
capture a first plurality of digital images of the field of view using the at least one camera and the at least one light source to illuminate the field of view at one or more illumination powers, the capturing of the first plurality of digital images including:
capturing a first digital image when the at least one light source is at a first illumination power;
capturing a second digital image when the at least one light source is at a second illumination power; and
capturing a third digital image when the at least one light source is at the first illumination power;

identify pixels of the first plurality of digital images that correspond to an object rather than to a background, wherein the pixels corresponding to the object are identified based on a difference between the second digital image and the first digital image and a difference between the second digital image and the third digital image;
construct a model of the object based on the identified pixels, the model including a position and a shape of the object;
capture a second plurality of digital images;
identify pixels of the second plurality of digital images that correspond to the object rather than to the background;
detect a motion of the object based at least in part upon the identified pixels of the first plurality of digital images and the identified pixels of the second plurality of digital images; and
interpret the detected motion as gestures indicating user interaction with a virtual or augmented environment.

18. The wearable goggle of claim 17 , wherein the virtual or augmented environment comprises:
a representation of a user's hand based at least in part upon the captured first plurality of digital images and the

Assignees

• Leap Motion Inc

Inventors

• Holz David
• Yang Hua

Classifications

• G06F3/017 Primary

  Gesture based interaction, e.g. based on a set of recognized hand gestures (interaction based on gestures traced on a digitiser G06F3/04883) · CPC title

• H04N23/661

  Transmitting camera control signals through networks, e.g. control via the Internet · CPC title

• G06T7/70

  Determining position or orientation of objects or cameras (camera calibration G06T7/80) · CPC title

• G06T7/11

  Region-based segmentation · CPC title

• G06T7/194

  involving foreground-background segmentation · CPC title