Rejection of light intrusion false alarms in a video security system

• US 5,937,092 A
• Filed: 12/23/1996
• Issued: 08/10/1999
• Est. Priority Date: 12/23/1996
• Status: Expired due to Term

• Current Assignee: ADT Security Services SA (Apollo Global Management, Inc.)
• Original Assignee: ESCO Electronics Corporation
• Inventors: Wootton, John R., Hobson, Gregory L., Waldman, Gary S.
• Primary Examiner(s): Moore, David K.
• Assistant Examiner(s): Chen, Wenpeng
• Application Number:
  US08/772,595
• Time in Patent Office: 960 Days
• Field of Search: 382/283, 382/100, 382/192, 382/153, 348/155, 348/154, 348/145, 348/152, 348/399, 348/700, 340/441, 340/555, 250/342
• US Class Current: 382/192
• 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: H04N 7/18 Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast

Abstract

This invention is a video security system (10) and a method for visually monitoring a scene and detecting motion of an intruder within the scene. A camera (C) continually views the scene and produces a representative signal. A processor (12) processes the signal and produces an image (f2) represented by the signal. This image is compared with a similar image (f1) of the scene from a previous point in time. Segments of the later image which differ from segments of the earlier image are identified. A discriminator (14) evaluates these segments to determine if the differences result simply from lighting changes, or the movement of an intruder within the scene as indicated by surface differences between segments of the respective differences. If caused by an intruder, an appropriate indication is provided. An algorithm is employed by which differences from one image to another caused by global or local lighting changes effects are identified so as not to produce false alarms.

Claims

1. A video security system visually monitoring a scene and detecting motion of an object within the scene comprising:

imaging means continually viewing the scene and producing a signal representative of the scene;

processor means processing said signal, comparing the signal representing the scene at one point in time with a similar signal representing the scene at a previous point in time, and identifying those segments of the scene at said point in time which differs from segments of the scene at the earlier point in time;

discriminator means evaluating those segments of the scene identified as being different to determine if the differences are caused by surface differences which are indicative of the presence of an intruder within the scene, or lighting changes which occur within the scene and do not indicate the presence of an intruder, if the difference is caused by the presence of an intruder providing an indication thereof;

said discriminator means including means comparing pixel elements contained in each segment of the scene at one point in time and corresponding pixel elements contained in a corresponding segment from the scene at an earlier point in time; and

said discriminator means determining a ratio of light intensity between each pixel in a segment with each pixel adjacent thereto, and means comparing the ratio values for the pixels in the segment of the scene at one point in time with the ratio values for the pixels in the corresponding segment of the scene at the earlier point in time.

2. The video security system of claim 1 wherein said imaging means comprises a camera capable of producing an image of a scene.

3. The video security system of claim 2 wherein said camera is a television camera operating in the visible portion of the light spectrum.

4. The video security system of claim 2 wherein said camera is an infrared camera operating in the infrared portion of the light spectrum.

5. The video security system of claim 1 wherein the output from said imaging means is an analog signal and said processing means is an analog processor.

6. The video security system of claim 1 wherein the output from said imaging means is a digital signal and said processing means is a digital processor.

7. The video security system of claim 1 wherein said processor means comprises a hybrid processor.

8. The video security system of claim 1 wherein said discriminator means further includes threshold means determining if a computed ratio for one pixel and an adjacent pixel in the segment of the scene at the one point in time varies by a predetermined threshold value from the computed ratio for the corresponding pixels in the scene segment at the earlier point in time.

9. The video security system of claim 8 wherein said discriminator means further includes a second threshold means determining the number of computed ratios for said pixel elements in the scene at the one point in time which exceed the predetermined value, the number of computed ratios, if exceeding a second threshold value, indicating that the difference between the scenes at the separate points in time are the result of the presence of an intruder in the scene.

10. A method of determining the presence of an object in a scene comprising:

continuously viewing the scene and developing a signal representative of the scene at different points in time;

processing each signal representing the scene, processing of said signal including comparing the signal representing the scene at one point in time with a signal representing the scene at a previous point in time, and identifying those segments of the scene at the later time which differ from segments of the scene at the earlier time, by determining the ratio of the gray level of adjacent pixels representing an area of interest within the scene, and counting the number of occurrences in which the ratio exceeds a predetermined value; and

evaluating those scene segments identified as being different to determine if the differences are caused by surface differences which are indicative of the presence of an intruder within the scene, or lighting changes which occur within the scene and do not indicate the presence of an intruder.

11. The method of claim 10 wherein if a difference in the scene is caused by the movement of an object into the scene rather than a change in lighting, an indication thereof is provided.

12. The method of claim 11 wherein evaluating the causes of differences in segments of the scene includes identifying differences resulting from light intrusions into the scene.

13. The method of claim 12 wherein evaluating the differences resulting from light intrusions includes identifying global lighting changes within the scene.

14. The method of claim 13 wherein evaluating the differences resulting from light intrusions includes identifying local lighting changes within the scene.

15. The method of claim 10 further including determining if the count value for the number of occurrences exceeds a threshold value and performing a further evaluation if the threshold value is exceeded.

16. The method of claim 10 wherein identifying those segments of the scene which differ includes constructing an edge map for both the earlier and the later viewing of the scene and determining if the resulting edges are substantially the same in each viewing, whereby if the differences are substantially the same, the differences between the segments is the result of lighting changes and further evaluation is not required, but if the differences are not substantially the same, further evaluation is performed.

17. A method of determining the presence of an object in a scene comprising:

continuously viewing the scene and developing a signal representative of the scene at different points in time;

processing each signal representing the scene, processing of said signal including comparing the signal representing the scene at one point in time with a signal representing the scene at a previous point in time, and identifying those segments of the scene at the later time which differ from segments of the scene at the earlier time;

evaluating those scene segments identified as being different to determine if the differences are caused by surface difference which are indicative of the presence of an intruder within the scene, or lighting changes which occur within the scene and do not indicate the presence of an intruder; and

wherein identifying segments of the scene which differ from one time to another includes

(a) calculating the ratios of adjacent pairs of pixels in a first frame representing an image of the scene at one point in time to determine if a ratio for a pixel pair differs by an amount exceeding a first threshold level, and incrementing a first count value each time the first threshold level is exceeded;

(b) calculating the ratios for corresponding pixel pairs in a second frame representing an image of the scene at another point in time for those pixel pairs in the first frame whose ratio exceeds the first threshold value;

(c) determining if a calculated ratio for a corresponding pixel pair in the second frame differs from the calculated ratio for the pixel pair in the first frame by an amount exceeding a second threshold value, and incrementing a second count value each time the second threshold level is exceeded;

calculating the ratio of adjacent pixels for any edge of a segment which appears in the second frame but not the first frame to determine if a ratio for a pixel pair differs by an amount exceeding the first threshold level, and incrementing the first count value each time the first threshold level is exceeded;

(d) calculating the ratios for corresponding pixel pairs in the first frame if the ratios for the pixel pairs in the second frame at said edge of a segment which exceeds the first threshold value;

(e) determining if a calculated ratio for a corresponding pixel pair in the first frame differs from the calculated ratio for the pixel pair in the second frame at said edge of a segment by an amount exceeding the second threshold value, and incrementing the second count value each time the second threshold level is exceeded; and

(f) identifying the segment as an intrusion if the ratio of the second count value to the first count value exceeds a third predetermined threshold value.

18. A method for discriminating an intrusion in a scene from changes in lighting of said scene comprising:

viewing said scene at a first point in time and developing an image signal representative of said scene at said first point in time;

viewing said scene at a second, later point in time, and developing a second image signal representative of said scene at said second point in time;

comparing said first and second image signals and obtaining a set of differences, if any, between the two image signals;

responding to said set of differences exceeding a threshold by viewing said scene at a third, and still later point in time, and developing a third image signal representative of said scene at said third point in time;

comparing said first and third image signals to produce a second set of differences;

comparing said second and third image signals to produce a third set of differences; and

responding to said second and third sets of differences exceeding said threshold, evaluation of said first and third image signals being performed to determine if any differences between the respective image signals result from changes in the irradiation of surfaces in the scene caused by changes in global lighting of said scene, said differences, if not a result of changes in the irradiation of surfaces, being due to the introduction of a new reflecting surface into said scene.

19. The method of claim 18 for discriminating an intrusion in a scene from changes in lighting of said scene wherein evaluating said first and third image signals to reject local illumination changes in said scene includes:

identifying pixels representing edges in said first image signal;

counting said identified edge pixels;

comparing ratios of reflectance of adjacent pixels representing edges in said first image signal with ratios of reflectance for corresponding pixels in said third image signal;

counting said compared ratios of reflectance exceeding a threshold value;

identifying pixels in said third image signal representing new edges in said third image signal not present in said first image signal;

incrementing said count of identified edge pixels by the number of edge pixels identified in said third image signal;

comparing ratios of reflectance of adjacent pixels representing new edges in said third image signal with ratios of reflectance for corresponding pixels in said first image signal;

incrementing said count of compared ratios of reflectance exceeding said threshold value by the number of compared ratios of reflectance between said third and first image signals exceeding said threshold value; and

responding to a ratio of said count of compared ratios of reflectance to said count of identified edge pixels exceeding a second threshold value, to provide an indication of an intrusion into said scene.

20. The method of claim 19 wherein the steps of identifying edge pixels and comparing ratios of reflectance are performed only on segmented portions of said first and third image signals corresponding to said second set of differences between said first and third signals.

21. The method of claim 19 wherein the steps of identifying edge pixels in said first image signal and identifying new edge pixels in said third image signal further include:

comparing the reflectance of horizontally and vertically adjacent pixels to determine if the ratio of reflectance differs from unity by a predetermined amount; and

classifying said adjacent pixels as a edge pixels if said ratio of reflectance exceeds said predetermined amount.