Method of Detecting Presence and Motion for Door Control Devices and Door Control Devices Implementing Such a Demand

• US 20080117020A1
• Filed: 12/22/2005
• Published: 05/22/2008
• Est. Priority Date: 12/23/2004
• Status: Active Grant

• Current Assignee: Celec Conception Electronique En Abrege Celec
• Original Assignee: Celec Conception Electronique En Abrege Celec
• Inventors: Martin, Christian
• Granted Patent: US 8,068,006 B2
• Time in Patent Office: Days
• US Class Current: 340/5.7
• CPC Class Codes: B66B 13/26 between closing doors
• CPC Class Codes: G06T 7/269 using gradient-based methods
• CPC Class Codes: H04N 7/18 Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
• CPC Class Codes: H04N 7/186 Video door telephones

Abstract

A method of detecting presence and motion for door control devices, and a door control device implementing the method. On receipt of a door opening command (24), at least a portion of at least one image captured by a digital camera of the floor area is compressed (28) and the compressed image is compared (32) with at least one active presence criterion to inhibit or not inhibit the door closure command (34).

Claims

1. Method of detecting presence and motion for door control devices of the type implementing a camera (4) for capturing, in at least one spectral zone (visible light, infrared), digital images including at least one image plane and in analysing the information in at least one image or image plane, characterized in that it consists in:

capturing (26;

60, 66;

80) at least one digital image, of at least one floor area (56) in which the door installation is disposed, in a digital memory in the form of at least one two-dimensional matrix of pixel values which is representative of at least one image signal component (Y;

R, G, B;

Y, Cr, Cb);

executing a compression (28;

62;

86) of at least one captured image or image plane so as to adapt it to a given memory capacity and to a given resolution in the floor area;

executing a comparison (32;

72;

92;

116) of at least one portion of the compressed image or image plane with at least one comparison value established on the basis of an active presence criterion to produce (34;

74;

94;

118, 120) an inhibition signal controlling at least one door control device; and

deducing from this a door control device control decision.

2. Method according to claim 1, characterized in that the compression step includes a step for selecting determined dimensions of an individual block based on the requested resolution at least in a floor area, the aperture of the image sensor and the dimensions of the image field, and the available memory size.

3. Method according to claim 2, characterized in that the compression step consists in calculating the average value of a characteristic of an image signal, such as the brightness, in each individual block.

4. Method according to claim 1, characterized in that the comparison step includes a step for calculating the inter-block difference of the reference image and the second image.

5. Method according to claim 4, characterized in that the comparison step also includes a step for comparing at least one inter-block difference value with a predetermined threshold value, Threshold:

Δ

Y>

Threshold, such that if the inter-block difference value exceeds the predetermined threshold “

Threshold”

, a signal detecting motion in the second image is produced.

6. Method according to claim 5, characterized in that said threshold value is adjusted according to the relation:

Threshold=Threshold 0+tl*K

in which the brightness ratio tl is determined by the relation;

$\mathrm{tl}=\frac{\mathrm{Lm}}{\mathrm{Lm}-\mathrm{Ym}}$

Threshold_—

0 is a threshold offset value and K is a coefficient, both of which are determined on installation on a particular door control device by successive tests in order to cancel out the error ratio of the active detection test (72).

7. Method according to claim 1, characterized in that the image capture is sequential and in that the reference image is reloaded periodically.

8. Method according to claim 1, characterized in that the floor area determination step (50) includes:

a step for determining door pillars by producing at least two lines (52, 54) representing the pillars;

a step for determining a line (56) joining the ends of the reproducing lines (52, 54) representing the pillars;

a step for dimensioning in the orthogonal direction (58) a rectangle determining floor area (50).

9. Method according to claim 8, characterized in that it includes a step for calculating a gradient (88) in the compressed image of the floor area (50).

10. Method according to claim 9, characterized in that the active presence detection step includes a step (92) for comparing at least one gradient value in the compressed image of the floor area with a predetermined threshold.

11. Method according to claim 10, characterized in that the step for comparing at least one gradient value is preceded by a threshold calculation and/or adaptation step (90).

12. Method according to claim 9, characterized in that the gradient calculation step includes a step for calculating two gradients (100, 102) in orthogonal directions to detect forward and/or sideways motions.

13. Method according to claim 12, characterized in that the detection of forward and/or sideways motions is executed by searching (108, 110) for the positions of the maxima of the two gradients (104, 106), then by calculating the variations (dx, dy) of the positions (i, j) of the maxima at predetermined reference values (i0, j0) and by comparing the two position variations (dy>

dx) to deduce from this whether the motion in the image is forward or sideways to deduce from this a decision to inhibit a closure command.

14. Method according to claim 13, characterized in that it includes a step for calculating reference values (I0, j0) by assigning, in a sequence of processed images, at least one value captured previously for each of the positions (i, j) of the maxima of the gradients.

15. Method according to claim 1, characterized in that it consists in determining a combination of at least one of the active presence detection criteria:

detection of presence (60-74);

detection of motion (80-94);

detection of forward or sideways motion (100-120).

16. Method according to claim 1, characterized in that it is executed for each component of the image signal (R, G, B;

Y, Cr, Cb) and in that it consists in determining a decision by majority vote on each parameter drawn from each component of the image signal.

17. Method according to claim 16, characterized in that it consists in associating, in one and the same plane of a given resolution, at least two components of resolutions at most equal to the given resolution then in executing the steps on the basis of said image planes.

18. Door control device of the type including at least one digital image matrix sensor coupled with a two-dimensional memory of pixel values, characterized in that it includes:

at least one processor for executing the compression of the captured image in memory;

at least one processor for producing a decision signal to execute the door control device command on the basis of a predetermined active presence decision criterion; and

a circuit controlling the closure of the door control device sensitive to an active presence decision signal that is produced.

19. Control device according to claim 18, characterized in that it includes a memory for each image plane, the image signal components of lower resolutions being associated in one and the same image plane and in that it includes a means for executing decisions on majority vote for all of the image planes processed.