Video motion detector with global insensitivity

• US 6,130,707 A
• Filed: 04/14/1997
• Issued: 10/10/2000
• Est. Priority Date: 04/14/1997
• Status: Expired due to Term

• Current Assignee: Philips Electronics North America Corporation (Koninklijke Philips N.V.)
• Original Assignee: Philips Electronics North America Corporation (Koninklijke Philips N.V.)
• Inventors: Koller, David P., Preschutti, Joseph P.
• Primary Examiner(s): Kelley, Chris S.
• Assistant Examiner(s): An, Shawn S.
• Application Number:
  US08/834,072
• Time in Patent Office: 1,275 Days
• Field of Search: 348/416, 348/699, 348/155, 348/152, 348/154, 340/324
• US Class Current: 348/155
• CPC Class Codes: G08B 13/19602 Image analysis to detect motion of the intruder, e.g. by frame subtraction
• CPC Class Codes: G08B 13/19604 involving reference image or background adaptation with time to compensate for changing conditions, e.g. reference image update on detection of light level change
• 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

Abstract

A motion detection system compares the differences between regions in a reference frame and regions in a subsequent frame to a threshold value. A difference frame is defined containing regions indicating which of the corresponding regions in the subsequent frame exceed the threshold difference when compared to the reference frame. Target and global parameters are defined so that a cluster of regions in the difference frame which indicate above threshold differences, must be greater than the target parameter but less than the global parameter for motion to be deemed to have occurred.

Claims

1. A method for detecting motion in sequential images, said method comprising the steps of:

defining a target size parameter indicating a minimized area to trigger an alarm,

defining a global change parameter,

creating a reference frame in dependence upon one or more first images,

creating a subsequent frame in dependence upon a second image,

dividing said reference frame and said subsequent frame into a plurality of regions;

comparing corresponding regions in said subsequent frame and said reference frame to a threshold thereby forming a difference measure frame composed of difference measure regions,

identifying difference measure regions with a value substantially different from other of said difference measure regions,

generating an alarm signal when a number of contiguous substantially different difference measure regions is greater than said target size parameter, and said number of contiguous substantially different difference measure regions is less than said global change parameter.

2. A method for detecting motion as claimed in claim 1, wherein the step of generating an alarm signal comprises the steps of:

comparing the number of difference measure regions to said target size parameter to form a detect signal,

comparing the number of difference measure regions to said global change parameter to form an inhibit signal, and

generating an alarm signal in dependence upon said detect signal and said inhibit signal.

3. A method as claimed in claim 1, wherein said comparison is determined in dependence upon the luminance of said regions of said images.

4. A method as claimed in claim 1, wherein the identification of said difference measure regions whose values are substantially different from the value of the other difference measure regions comprises the steps of:

computing an average value of all the difference measure regions of the difference frame, and,

comparing each difference measure region'"'"'s value to said average value.

5. A method as claimed in claim 1, wherein the identification of difference measure regions whose values are substantially different from the value of other difference measure regions comprises a statistical test for significant differences.

6. A method as claimed in claim 1, wherein said step of generating of an alarm signal is further dependent upon a characteristic of said first images.

7. A method as claimed in claim 6, wherein said characteristic of the first images is correlated to a contrast measure.

8. A method as claimed in claim 7, wherein said threshold is determined based upon said contrast measure.

9. A motion detection system comprising:

means for creating a reference frame in dependence upon one or more first images,

means for creating a subsequent frame in dependence upon a second image,

said frames being composed of a plurality of regions;

means for comparing regions in said subsequent frame to corresponding regions in said reference frame with respect to a threshold thereby forming a difference frame composed of difference measure regions,

means for creating a target parameter indicating a minimized area to trigger an alarm,

means for creating a global parameter,

means for identifying difference measure regions with values substantially different from other of said difference measure regions,

means for creating a motion detection signal when a number of contiguous substantially different difference measure regions is greater than said target size parameter and said number of contiguous substantially different difference measure regions is less than said global change parameter.

10. A motion detection system as claimed in claim 9, wherein said means for creating the reference frame comprises

means for computing a weighted average of one or more characteristics of said first images.

11. A motion detection system as claimed in claim 9, wherein said means for generating a motion detection signal is dependent upon one or more characteristics of said first images.

12. A motion detection system as claimed in claim 11, wherein one of the characteristics of the first images is a contrast factor.

13. A motion detection system as claimed in claim 9, wherein

said regions of said first and second images are characterized by a luminance measure, and

said means for comparing is based upon the luminance measure of the corresponding first and second subareas.

14. A motion detection system comprising:

means for creating a reference frame in dependence upon a first image,

means for creating a subsequent frame in dependence upon a second image,

means for comparing said reference frame and said subsequent frame to a threshold factor thereby forming a difference frame, said difference frame comprising a plurality of sub-elements,

means for defining a target size parameter indicating a minimized area to trigger an alarm,

means for defining a global change parameter,

means for identifying sub-elements which have a value substantially different from other of said sub-elements; and

means for producing a motion detection signal when a number of contiguous substantially different sub-elements is greater than said target size parameter and said number of contiguous substantially different sub-elements is less than said global change parameter.

15. A motion detection system as claimed in claim 14, wherein

said sub-elements correspond to portions of said second image,

each of said sub-elements having a value representative of the difference between said corresponding portion of the second image and the reference image, and

said threshold factor is a statistic of said sub-element values.

16. A motion detection system as claimed in claim 14, wherein said means for determining a threshold factor is further dependent upon a characteristic of said first reference frame.

17. A motion detection system as claimed in claim 14, wherein said means for creating a reference frame is further dependent upon one or more prior images.