Video-based system and method for counting persons traversing areas being monitored

• US 20040145658A1
• Filed: 12/29/2003
• Published: 07/29/2004
• Est. Priority Date: 01/13/2000
• Status: Active Grant

• Current Assignee: CountWise LLC
• Original Assignee: CountWise LLC
• Inventors: Topaz, Dov, Lev-Ran, Ilan
• Granted Patent: US 7,612,796 B2
• Time in Patent Office: Days
• US Class Current: 348/143
• CPC Class Codes: G06V 20/53 Recognition of crowd images, e.g. recognition of crowd congestion
• CPC Class Codes: G06V 40/103 Static body considered as a whole, e.g. static pedestrian or occupant recognition
• CPC Class Codes: G07C 9/00 Individual registration on entry or exit
• CPC Class Codes: G08B 13/19602 Image analysis to detect motion of the intruder, e.g. by frame subtraction
• CPC Class Codes: G08B 13/19606 Discriminating between target movement or movement in an area of interest and other non-signicative movements, e.g. target movements induced by camera shake or movements of pets, falling leaves, rotating fan
• CPC Class Codes: G08B 13/19619 Details of casing
• CPC Class Codes: G08B 13/19632 Camera support structures, e.g. attachment means, poles
• CPC Class Codes: G08B 13/19656 Network used to communicate with a camera, e.g. WAN, LAN, Internet

Abstract

A video based system and method for detecting and counting persons traversing at least two areas being monitored. The method includes the steps of initialization of at least one end unit, having at least a camera, the camera producing images of at least a first area being monitored and a second area being monitored, digitizing the images and storing the digitized images in a non-volatile memory unit and a working memory unit; detecting potential persons in the images; comparing the digitized images of objects detected in the first area and second area being monitored with digitized images stored in the working memory unit to determine whether the detected object is a new figure in such area or whether the detected object is a known figure that has remained within such area and to determine that a figure which was not detected has left such area; incrementing a counter for such area with an indication of the number of persons that have passed through such area, and comparing the number of persons passing through a first area with the number of persons passing through a second area.

Claims

1. A method of counting persons traversing areas being monitored comprising:

initialization of at least an end unit, said end unit having installed therein, at least a camera producing images of a first area being monitored and a second area being monitored, said end unit comprising at least a non-volatile memory unit and a working memory unit, the non-volatile memory unit comprising a plurality of counters;

digitizing said images and storing said digitized images in a non-volatile memory unit and a working memory unit;

detecting objects being potential persons from said digitized images;

comparing the digitized images of objects detected in said first area being monitored with digitized images of said first area being monitored stored in the working memory unit to determine whether the detected object is a new figure that has entered said first area being monitored or whether the detected object is a known figure, that has remained within said first area being monitored and to determine that a figure which was not detected has left said first area being monitored; and

incrementing at least a first of said plurality of counters with an indication of the number of persons that have passed through said first area being monitored,

comparing the digitized images of objects detected in said second area being monitored with digitized images of said second area being monitored stored in the working memory unit to determine whether the detected object is a new figure that has entered said second area being monitored or whether the detected object is a known figure, that has remained within said second area being monitored and to determine that a figure which was not detected has left said second area being monitored; and

incrementing at least a second of said plurality of counters with an indication of the number of persons that have passed through said second area being monitored, and

comparing said indication of the number of persons that have passed through said first area being monitored with said indication of the number of persons that have passed through said second area being monitored.

2. The method according to claim 1, and further comprising the step of:

remotely retrieving data from said non-volatile memory unit and a working memory unit via a communications network.

3. The method according to claim 2, wherein said communications network is one of a group including a GSM cellular network and the Internet.

4. The method according to claim 1, and further comprising the step of:

processing data to produce a displayable report;

encrypting said report; and

accessing said encrypted report from a remote site via a communications network, said encrypted report being accessible by authorized users.

5. The method according to claim 4, wherein said communications network is one of a group including a GSM cellular network and the Internet.

6. The method according to claim 1 wherein said step of initialization further comprises the step of inputting the network ID data of the end unit if said unit is connected as part of a network of end units.

7. The method according to claim 1, wherein said step of initialization further comprises the step of determining the direction of motion of persons being monitored in said first area being monitored.

8. The method according to claim 7, wherein said step of determining the direction of motion of the figures comprises the step of initializing the video processor with directional definitions for bi-directional motion through the monitored area and the height of the camera above the surface.

9. The method according to claim 1, wherein said step of incrementing comprises the step of incrementing said counter for each object identified as a person passing through said first area being monitored, said person being identified by assigning a direction of motion to an object determined to be a potential person and verifying the direction of the potential person over a period of time to determine the potential person as a person to be counted.

10. The method according to claim 1, wherein said step of digitizing comprises the steps of:

in a time T1, receiving a first array of binary data containing the grayness level existing over a first area being monitored and storing said first array in the program working memory;

in a time T2, receiving a second array of binary data containing grayness levels existing over said first area being monitored;

calculating the differences between the first and second arrays at the single pixel level to obtain an array representing color differences;

creating a bitmap at the single pixel level comprising a binary array indicating a substantial change of color at the single pixel level and changes that have not occurred or are less than a pre-determined threshold level;

processing the bitmap to prepare an “

Y”

histogram to locate coordinates of concentrations of changes in the direction perpendicular to the direction of expected motion and an “

X”

histogram to find overlapping coordinates of change concentrations in the direction of expected motion; and

analyzing said “

Y” and

“

X”

histogram at the single pixel level.

11. The method according to claim 10, wherein said step of detection further comprises the step of filtering said histogram by rejecting those coordinates where the space between them is too narrow to contain a human figure as determined by pre-determined thresholds.

12. The method according to claim 11, wherein said pre-determined thresholds representing the width of a man'"'"'s shoulders and the distance between his chest and back in the direction of the expected bi-directional motion.

13. The method according to claim 11, wherein said step of detection further comprises the step of further filtering said histogram by rejecting objects not having a sufficiently high density change representing a person in motion.

14. The method according to claim 1, wherein said step of detection further comprises the step of applying a spatial frequency threshold.

15. The method according to claim 13, wherein said spatial frequency threshold is performed alternately from both sides of the potential figure.

16. The method according to claim 1, further comprising the steps of:

identifying a person as a moving figure by comparing the potential person with previously verified figures stored in memory and comparing the potential person with previously detected but non-verified figures stored in memory.

17. The method according to claim 16, further comprising the steps of:

comparing the potential person with previously detected figures stored in memory; and

searching for previously detected figures that were not detected during the present detection period.

18. The method according to claim 1, further comprising the step of determining that a person has a first area to be monitored if said person was not detected during a plurality of detection periods.

19. The method according to claim 1, wherein said initialization of at least one end unit comprises initializing said at least one unit including at least two cameras installed therein, a first of said cameras producing images of said first area being monitored and a second of said cameras producing images of said second area being monitored.

20. An end unit for use with a video based image processing system, the end unit configured to obtain information about the motion of people in at least a first area and a second area, the end unit comprising:

at least a CCD camera for capturing images of said first area and said second area;

an A/D converter for converting analog information from said camera to digital information;

a digital switching unit connected to the A/D converter;

a video processor connected to the digital switching unit, for receiving the digital video information from said digital switching unit and to process said digital information at a single pixel level; and

a non-volatile memory unit and a program working memory unit.

21. The end unit according to claim 20, and further comprising:

at least one digital video data input connected to the digital switching unit for receiving data from a second CCD camera.

22. The end unit according to claim 20, wherein said non-volatile memory unit is remotely accessible and stores data on the number of people in said first area and said second area.

23. The end unit according to claim 20, wherein said non-volatile memory unit includes discrete data input stored therein.

24. The end unit according to claim 20, and further configured for installation in an acoustic ceiling, having the body of the end unit above the ceiling and the CCD camera below the ceiling.

25. The end unit according to claim B5, wherein said camera is fitted within a dome unit connectable to said body by snap connectors fitted to extendable flanges attached.

26. A system for counting persons traversing a first area and a second area comprising:

at least an end unit, said end unit comprising;

at least a CCD camera for capturing images of said first area and said second area;

an A/D converter for converting analog information from the camera to digital information;

a digital switching unit connected to the A/D converter;

a video processor connected to the digital switching unit, for receiving the digital video information from said digital switching unit; and

a non-volatile memory unit and a program working memory unit.

27. The system as in claim 26, comprising

a device for remote retrieval of data stored in said non-volatile memory unit;

a processor for processing said data and generating reports comparing a number of persons traversing said first area with a number of persons traversing said second area; and

a device for displaying said reports.

28. The system according to claim 26 wherein said report is encrypted and wherein said encrypted report is accessible from a remote site via a communications network by authorized users.

29. The method according to claim C3, wherein said communications network is one of a group including a GSM cellular network and the Internet.