Smoke detection system and method using a camera

• US 10,304,306 B2
• Filed: 06/28/2016
• Issued: 05/28/2019
• Est. Priority Date: 02/19/2015
• Status: Active Grant

• Current Assignee: Smoke Detective LLC
• Original Assignee: Smoke Detective LLC
• Inventors: Mills, Russell P., Rohde, Gustavo K., Lanni, Frederick, Davis, Jr., Stephen C
• Primary Examiner(s): Messmore, Jonathan R
• Application Number:
  US15/194,878
• Publication Number:
  US 20160307424A1
• Time in Patent Office: 1,064 Days
• CPC Class Codes: G06V 10/751 Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching
• CPC Class Codes: G06V 20/52 Surveillance or monitoring of activities, e.g. for recognising suspicious objects recognising microscopic objects G06V20/69
• CPC Class Codes: G08B 17/125 by using a video camera to detect fire or smoke
• CPC Class Codes: H04N 7/183 for receiving images from a single remote source

Abstract

Image-based fire detection methods are provided which include analyzing a field of view based on a plurality of images, determining if changes occur, and analyzing whether any such changes are due to smoke-like material. Changes due to large objects are ruled out, and drift in the scene or shaking of the image collection device is accounted for by the method. The methods employ representing images of the field of view with quantifiable values and comparing to thresholds established for each event, and activating an alarm if changes due to smoke-like material are detected. The fire detection methods may also include monitoring a field of view for changes, and only performing the analysis on the images if changes are detected. The monitoring and validating stages of the methods include collecting images at different sampling rates for different amounts of information.

Claims

1. A method of detecting smoke in a field of view, comprising:

collecting a plurality of electronic images of said field of view with an image collection device, wherein each of said electronic images includes a plurality of pixels each having a quantitative pixel value;

comparing each of said pixel values in each of said plurality of electronic images to a predetermined pixel value threshold;

calculating a quantitative image pixel value representative of the number of said pixels having pixel values exceeding said predetermined pixel value threshold if any of said individual pixel values exceeds said predetermined pixel value threshold;

comparing said image pixel value to a predetermined image pixel value threshold;

parsing at least two of said plurality of electronic images into portions based on the resolution of the image collection device and the level of detail desired for the monitoring of the field of view;

selecting corresponding portions of the at least two of said plurality of electronic images if said image pixel value is less than said predetermined image pixel value threshold;

calculating a quantitative image portion value representative of each of said corresponding portions of said at least two electronic images;

comparing each of said image portion values to a predetermined image portion value threshold; and

activating an alarm if said image portion value for at least one of said corresponding portions is less than said predetermined image portion value threshold.

2. The method as recited in claim 1, further comprising converting said plurality of electronic images to a corresponding plurality of derivative images of said field of view.

3. The method as recited in claim 2, wherein said quantitative pixel value is defined as an absolute value.

4. The method as recited in claim 1, wherein each of said portions of said at least two electronic images is a subset of said electronic image.

5. The method as recited in claim 4, wherein each of said portions of said at least two electronic images comprises up to 10% of said electronic image.

6. The method as recited in claim 5, wherein each of said portions of said at least two electronic images comprises between 2%-4% of said electronic image.

7. The method as recited in claim 5, wherein each of said portions of said at least two electronic images comprises less than 1% of said electronic image.

8. The method as recited in claim 4, wherein the ratio of said portions of said at least two electronic images to the total pixels in each of said electronic images is in the range of 1:

500 to 1;

2000.

9. The method as recited in claim 8, wherein the ratio of said portions of said at least two electronic images to the total pixels in each of said electronic images is 1:

1024.

10. The method as recited in claim 8, wherein the ratio of said portions of said at least two electronic images to the total pixels in each of said electronic images is 1:

1156.

11. The method as recited in claim 1, wherein said predetermined pixel value threshold is dependent on a dynamic range of said image collecting device.

12. The method as recited in claim 11, wherein said predetermined pixel value threshold is at least one-fifth of the dynamic range of said image collecting device.

13. The method as recited in claim 1, wherein said predetermined pixel value threshold is defined as a digitized value proportionate to the pixel intensity and the threshold is defined as a value between 10 and 80.

14. The method as recited in claim 13, wherein said predetermined pixel value threshold is defined as 50.

15. The method as recited in claim 1, wherein said predetermined image pixel value threshold is defined as a certain percentage of the total number of pixels in an image, and the threshold is defined as a value in the range of 0.001%-0.1% of the total pixels in any one of said electronic images.

16. The method as recited in claim 15, wherein said predetermined image pixel value threshold is defined as a digitized value proportionate to the pixel intensity and the threshold is defined as a value of at least 400.

17. The method as recited in claim 1, wherein the image portion value is the result of an error function which is performed on at least one set of corresponding portions and said predetermined image portion value threshold is defined as a value between 0.75 and 0.99.

18. The method as recited in claim 17, wherein said predetermined image portion value threshold is defined as 0.95.

19. The method as recited in claim 1, wherein calculating a quantitative image portion value includes performing an error function calculation between corresponding portions of said at least two electronic images.

20. The method as recited in claim 19, wherein said error function is selected from the group consisting of a normalized scalar product, scalar product, correlation coefficient, sum of squared differences, dot product, vector dot product, non-standard dot product, and root mean squared.

21. The method as recited in claim 1, wherein said plurality of electronic images are at least one of RGB and grayscale.

22. The method as recited in claim 21, further comprising converting said plurality of electronic images between RGB and grayscale.

23. A method of detecting smoke in field of view, comprising:

collecting a first plurality of electronic images of said field of view with an image collection device at a first sampling rate;

selecting corresponding portions of at least two of said first plurality of electronic images;

calculating a first quantitative image portion value representative of each of said corresponding portions of said at least two electronic images from said first plurality of images;

comparing said first image portion values representative of each of said corresponding portions of said at least two electronic images from said first plurality of images to a first predetermined image portion value threshold;

collecting a second plurality of electronic images of said field of view with an image collection device at a second sampling rate greater than said first sampling rate if any of said first image portion values is less than said predetermined first image portion value threshold, wherein each of said second plurality of electronic images includes a plurality of pixels each defined by a quantitative pixel value;

comparing each of said pixel values in each of said second plurality of electronic images to a predetermined pixel value threshold;

calculating a quantitative image pixel value representative of the number of said pixels having pixel values exceeding said predetermined pixel value threshold if any of said individual pixel values exceeds said predetermined pixel value threshold;

comparing said image pixel value to a predetermined image pixel value threshold;

parsing at least two of said second plurality of electronic images into portions based on the resolution of the image collection device and the level of detail desired for the monitoring of the field of view;

selecting corresponding portions of the at least two of said second plurality of electronic images if said image pixel value is less than said predetermined image pixel value threshold;

calculating a second quantitative image portion value representative of each of said corresponding portions of said at least two electronic images of said second plurality of images;

comparing each of said second image portion values to a second predetermined image portion value threshold; and

activating an alarm if said second image portion value for at least one of said corresponding portions is less than said second predetermined image portion value threshold.

24. The method as recited in claim 23, wherein said second sampling rate is at least one order of magnitude greater than said first sampling rate.

25. The method as recited in claim 23, wherein said first sampling rate is one image every multiple seconds.

26. The method as recited in claim 25, wherein said first sampling rate is one image every 2-5 seconds.

27. The method as recited in claim 23, wherein said second sampling rate is multiple images per second.

28. The method as recited in claim 27, wherein said second sampling rate is 10-50 images per second.

29. The method as recited in claim 23, wherein the first image portion value is the result of an error function which is performed on at least one set of corresponding portions and said first predetermined image portion value threshold is defined to be a value between 0.70 and 0.99.

30. The method as recited in claim 29, wherein said first predetermined image portion value threshold is defined as 0.95.

31. The method as recited in claim 23, wherein calculating said first image portion value and said second image portion value includes performing an error function calculation between corresponding portions of said at least two electronic images.

32. The method as recited in claim 31, wherein said error function is selected from the group consisting of a normalized scalar product, scalar product, correlation coefficient, sum of squared differences, dot product, vector dot product, non-standard dot product, and root mean squared.

33. The method as recited in claim 23, further comprising:

determining the lowest value of said first image portion values of said corresponding portions of said at least two electronic images from said first plurality of images;

comparing said lowest value of said first image portion values to a predetermined first image portion value threshold; and

collecting a second plurality of electronic images of said field of view with an image collection device at a second sampling rate greater than said first sampling rate if said lowest value of said first image portion values is less than said predetermined first image portion value threshold.

34. The method as recited in claim 23, further comprising the steps of:

defining a reference image by collecting a first electronic image and defining said first electronic image as the reference image, and

at a predetermined interval, updating the reference image by collecting at least one subsequent electronic image and redefining at least one of said subsequent electronic images as the reference image.

35. The method as recited in claim 34, wherein said predetermined interval is a time interval.

36. The method as recited in claim 35, wherein said predetermined interval is 2-10 seconds.

37. The method as recited in claim 36, wherein said predetermined interval is 9 seconds.

38. The method as recited in claim 34, wherein said predetermined interval is a number of subsequent electronic images.

39. The method as recited in claim 38, wherein said predetermined interval is 2-4 subsequent electronic images.

40. The method as recited in claim 39, wherein said predetermined interval is 3 subsequent electronic images.

41. The method as recited in claim 23, wherein said first and second plurality of electronic images are at least one of RGB and grayscale.

42. The method as recited in claim 41, further comprising converting said first and second plurality of electronic images between RGB and grayscale.