SYSTEM AND METHOD FOR COUNTING PEOPLE

• US 20080008360A1
• Filed: 11/02/2006
• Published: 01/10/2008
• Est. Priority Date: 11/05/2005
• Status: Active Grant

• Current Assignee: PH Technical Labs
• Original Assignee: PH Technical Labs
• Inventors: Pattikonda, Ram
• Granted Patent: US 8,228,382 B2
• Time in Patent Office: Days
• US Class Current: 382/118
• CPC Class Codes: G06V 40/161 Detection; Localisation; Normalisation

Abstract

A system for counting people in a specified area. The system includes a camera for capturing an image of the specified area and a computer for receiving the captured image. The computer analyzes the image to detect people by detecting head or face shapes in the image. The computer counts the detected head or face shapes to determine a number of people within the specified area. The computer may confirm that the head or face shapes are human by determining if the shapes have the approximate coloration of a human. The system may detect stationary or moving persons. In addition, the system may detect the presence of video recording devices in a room. The system may also detect if a seat is occupied by determining that a pattern in the seat is blocked or the outline of the seat is blocked.

Claims

1. A system for counting people in a specified area, the system comprising:

a camera for capturing an image of the specified area; and

a computer for receiving the captured image, the computer analyzing the image to detect people;

whereby the computer detects head or face shapes within the image and counts the detected head or face shapes to determine a number of people within the specified area.

2. The system for counting people in a specified area of claim 1 wherein the camera captures an image at a specific time.

3. The system for counting people in a specified area of claim 1 wherein the computer confirms a detected head or face shape is a person by determining if the detected shape matches an approximate color of human skin.

4. The system for counting people in a specified area of claim 1 the computer confirms a detected head or face shape is a person by determining if the detected shape geometrically matches eyes, lips or nose of a human.

5. The system for counting people in a specified area of claim 1 wherein the camera utilizes setting having a high aperture and large exposure with screen illumination to capture the image.

6. The system for counting people in a specified area of claim 1 wherein the camera utilizes infrared light to illuminate the specified area.

7. The system for counting people in a specified area of claim 1 wherein the camera utilizes a wide angle lens.

8. The system for counting people in a specified area of claim 1 wherein the camera is positioned in a corner of the specified area and the computer utilizes image compensation to correct for angular distortion from positioning the camera at the corner of the specified area.

9. The system for counting people in a specified area of claim 1 further comprising means to detect a video recording device positioned within the specified area, the video recording device being detected by detecting a specular refection from a lens of the video recording device.

10. The system for counting people in a specified area of claim 1 wherein the people located in the specified area are standing in a queue and, upon determining that the queue is beyond a specified length limit, the computer triggers an alarm.

11. The system for counting people in a specified area of claim 1 wherein the system further comprises means to detect a shape of a face or head in a specified area when people are moving.

12. The system for counting people in a specified area of claim 11 wherein the camera captures a series of images to detect motion of a moving person, the computer comparing the position of the face or head of each person on at least two images.

13. A method of counting people in a specified area, the method comprising the steps of:

capturing an image by a camera of the specified area;

analyzing the image to detect people by the computer, the computer detecting head or face shapes; and

counting the detected head or face shapes to determine a number of people within the specified area.

14. The method of counting people in a specified area of claim 13 wherein the step of analyzing the image includes the step of confirming by the computer that a detected head or face shape is a person by determining if the detected shape geometrically matches eyes, lips or nose of a human.

15. The method of counting people in a specified area of claim 13 wherein the step of analyzing the image includes the step of confirming by the computer that a detected head or face shape is a person by determining if the detected shape matches an approximate color of human skin.

16. The method of counting people in a specified area of claim 13 further comprising the step of detecting a video recording device positioned within the specified area, the video recording device being detected by detecting a specular refection from a lens of the video recording device.

17. The method of counting people in a specified area of claim 13 wherein the people are standing in a queue and further comprises the step of issuing an alarm upon determining that the queue is beyond a specified length limit.

18. A system for counting people in a specified area, the system comprising:

a camera for capturing an image of the specified area having a plurality of seats; and

a computer for receiving the captured image, the computer analyzing the image to detect people;

whereby the computer determines if each seat is occupied when a seat is blocked from view in the captured image to determine a number of people within the specified area.

19. The system for counting people of claim 18 wherein the computer determines if a seat is blocked if a pattern on a seat is blocked from view.

20. The system for counting people of claim 18 wherein the computer determines if a seat is blocked if an outline of a seat is blocked from view.

21. The system for counting people of claim 18 wherein the computer detects head or face shapes within the image and counts the detected head or face shapes to determine a number of people within the specified area.