Determination of a position characteristic for an object

Patent metadata

Field	Value
Publication number	US-9373169-B2
Application number	US-201113519363-A
Country	US
Kind code	B2
Filing date	Jan 10, 2011
Priority date	Jan 12, 2010
Publication date	Jun 21, 2016
Grant date	Jun 21, 2016

Abstract

A system for determining a position characteristic for an object in a room comprises an optically identifiable element ( 103 ) positioned on a background surface for a detection area of the room. A camera ( 101 ) is positioned across the detection area from the optically identifiable element ( 103 ) and captures an image comprising the detection area and the optically identifiable element ( 103 ). A detector ( 403 ) detects the optically identifiable element ( 103 ) in the image based on an optical property. An occlusion processor ( 405 ) determines an occlusion property for an occlusion of the optically identifiable element ( 103 ) in the image in response to the optical property. A position processor ( 407 ) determines a depth position in response to the occlusion property where, the depth position is indicative of a position of the object along the optical axis of the camera ( 101 ). The invention may for example allow presence detection and rough localization of a person in a room.

First claim

The invention claimed is:

1. A system for determining a position characteristic for an object in a detection area of a space, the system comprising:
an optically identifiable element having a width at least 10 times more than a height thereof, said optically identifiable element being positioned on a vertical surface proximal to a ground surface in the detection area by a largest distance from the ground surface of no more than 50 cm, the optically identifiable element having an optical property;
a camera positioned across the detection area from the optically identifiable element, the camera capturing an image that comprises (i) the detection area and (ii) the optically identifiable element, wherein the image further comprises an image plane of pixels (u,v) extending in a horizontal direction and in a vertical direction, where u is a vertical axis and v is a horizontal axis;
a detector for detecting the optically identifiable element in the image in response to the optical property;
an occlusion processor for determining an occlusion property of at least one occlusion of the optically identifiable element in the image in response to the optical property, wherein the occlusion property includes at least one of an image position and an image size of an occluded segment of the optically identifiable element; and
a position processor (i) for determining an image characteristic of an image object in the image associated with the occluded segment of the optically identifiable element, wherein the image characteristic is indicative of an extension of the image object in an image direction away from the optically identifiable element, (ii) for determining a segment border in the image between (ii)(a) an image region where the image object is mapped and (ii)(b) an image region where the image object touches a floor of the space in the image and (iii) for determining a depth position for the object in the detection area in response to (iii)(a) the occlusion property, further in response to (iii)(b) the image characteristic of the image object, and still further in response to (iii)(c) a u-coordinate for the image object that represents a vertical image distance between a line corresponding to the optically identifiable element and the segment border such that only a portion of an image object segment extending below the line is considered, and wherein the depth position is indicative of a position of the object along an optical axis of the camera.

2. The system of claim 1 , wherein the space comprises a room.

3. The system of claim 1 , wherein the optically identifiable element is an elongated element (i) applied to a wall or (ii) used to form an artificial boundary.

4. The system of claim 1 , further comprising:
a further camera positioned across the detection area from the optically identifiable element, the further camera capturing a further image that comprises (i) the detection area and (ii) the optically identifiable element, wherein
the detector further detects the optically identifiable element in the further image in response to the optical property,
the occlusion processor determines a further occlusion property of at least one occlusion of the optically identifiable element in the further image in response to the optical property, and
the position processor determines the depth position further in response to both (i) the occlusion property and (ii) the further occlusion property.

5. The system of claim 4 , wherein the position processor determines the depth position further in response to a difference in relative image positions between occluded segments in (i) the image and (ii) the further image.

6. The system of claim 1 , further comprising:
a size processor for determining a size estimate for the object in response to (i) the depth position and (ii) an image size of at least one occluded image segment of the optically identifiable element.

7. The system of claim 1 , wherein the object comprises a human or non-human living creature, the system further comprising:
a detection processor for detecting a presence of the living creature in the detection area in response to the occlusion property.

8. The system of claim 7 , wherein the detection processor detects the presence further in response to a dynamic variation of an occlusion of the optically identifiable element in the image.

9. The system of claim 8 , wherein the detection processor further determines a first characteristic of at least one of (i) the living creature and (ii) an activity of the living creature in response to the dynamic variation.

10. The system of claim 1 , wherein the detector further estimates a geometric area of the image corresponding to the optically identifiable element in response to detections of image areas having an image property matching the optical property, and the occlusion processor further identifies occluded segments of the geometric area having an image property not matching the optical property.

11. The system of claim 1 , further comprising:
a processor device for determining a set of nominal occluded segments of the optically identifiable element in the space, wherein the position processor further compensates the occlusion property for the nominal occluded segments.

12. The system of claim 11 , wherein the processor device further designates an occluded segment of the optically identifiable element as a nominal occluded segment if a duration of the occlusion exceeds a threshold.

13. The system of claim 1 , further comprising:
an occlusion analysis processor for analyzing optical properties of at least a spatial region in at least one captured image of all pixels corresponding to at least one of (i) the optically identifiable element and (ii) an occluding object.

14. The system of claim 1 , further comprising:
a camera containing unit with containing regions that allow incorporation of (i) at least one portable camera or (ii) at least one portable device with a camera.

15. The system of claim 1 , further comprising:
at least two portable devices with a camera; and
a processor for executing software for coordinating between the at least two portable devices with a camera, the coordinating at least involving comparing captured images from the at least two portable devices with a camera.

16. A method of determining a position characteristic for an object in a space, the method comprising:
positioning an optically identifiable element having a width at least 10 times more than a height thereof on a vertical surface proximal to a ground surface in a detection area of the space by a largest distance from the ground surface of no more than 50 cm, the optically identifiable element having an optical property;
positioning a camera across the detection area from the optically identifiable element, the camera capturing an image that comprises (i) the detection area and (ii) the optically identifiable element, wherein the image further comprises an image plane of pixels (u,v) extending in a horizontal direction and in a vertical direction, where u is a vertical axis and v is a horizontal axis;
detecting the optically identifiable element in the image in response to the optical property;
determining an occlusion property for an occlusion of the optically identifiable element in the image in response to the optical property, wherein the occlusion property includes at least one of an image position and an image size of an occluded segment of the optically identifiable element; and
determining (i) an image characteristic of an image object in the image associated with the occluded segment of the

Assignees

• Varekamp Christiaan
• Vandewalle Patrick Luc E
• Mertens Mark Jozef Willem
• Koninkl Philips Nv

Inventors

• Varekamp Christiaan
• Vandewalle Patrick Luc E
• Mertens Mark Jozef Willem

Classifications

• G06T7/004 Primary

  Physics · mapped topic

• G06T7/50

  Depth or shape recovery · CPC title

• G06T7/70 Primary

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

• G06V40/10

  Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands · CPC title

• G06T2207/10

  Image acquisition modality · CPC title