SMOKE DETECTION METHOD BASED ON VIDEO PROCESSING

• US 20090123074A1
• Filed: 11/13/2007
• Published: 05/14/2009
• Est. Priority Date: 11/13/2007
• Status: Active Grant

• Current Assignee: Huper Laboratories Co., Ltd.
• Original Assignee: Huper Laboratories Co., Ltd.
• Inventors: Yin, Yen-Hui, Kao, Cheng-Liang, Chen, Chao-Ho
• Granted Patent: US 7,609,856 B2
• Time in Patent Office: Days
• US Class Current: 382/190
• 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 17/12 Actuation by presence of radiation or particles, e.g. of infrared radiation or of ions

Abstract

The static and dynamic characteristics of smoke are developed for extracting smoke pixels of an image and effective smoke detection. In static characteristic, a chromaticity-based analysis is employed to extract grayish pixels in the moving region from a monitored area. Then, a circumference-based disorder measurement in dynamic characteristic is developed to obtain the smoke pixels from the grayish pixels. To enhance the reliability of alarming, a fuzzy system is introduced to judge if the smoke will spread out by checking its growth-rate.

Claims

1. A smoke detection method based on video processing comprising:

capturing images in a monitored area;

segmenting moving regions from the captured images, each moving region comprising a plurality of pixels;

extracting a plurality of smoke pixels from the plurality of pixels in the moving regions according to a first rule; and

determining whether the plurality of smoke pixels forms real smoke according to a second rule.

2. The smoke detection method of claim 1, wherein capturing images in a monitored area comprises:

capturing images in the monitored area at a first time slice and a second time slice, wherein the first time slice and the second time slice are successive time slices with fixed interval.

3. The smoke detection method of claim 1, wherein segmenting moving regions from the captured images comprises segmenting a moving region in a captured image according to the successive images'"'"' frame difference in the captured images.

4. The smoke detection method of claim 1, wherein extracting a plurality of smoke pixels according to a first rule comprises:

updating the number of smoke pixels according to the color components'"'"' gray levels and the intensity (I) of HSI color model of the plurality of pixels.

5. The smoke detection method of claim 4, wherein updating the number of the plurality of smoke pixels according to the color components'"'"' gray levels and the intensity (I) of HSI color model of the plurality of pixels comprises:

determining a pixel as a smoke pixel when red, green and blue gray levels of the pixel are approximately equal and when the intensity of the pixel is within a predetermined range.

6. The smoke detection method of claim 1, wherein determining whether the plurality of smoke pixels forms real smoke according to a second rule comprises:

determining the plurality of smoke pixels as real smoke according to the disorderness and the growth-rate of the plurality of smoke pixels.

7. The smoke detection method of claim 6, wherein determining the plurality of smoke pixels as real smoke according to the disorderness and the growth-rate of the plurality of smoke pixels comprises:

determining the plurality of smoke pixels as real smoke when the ratio of circumference of the region formed by the smoke pixels and the number of smoke pixels is greater than a threshold.

8. The smoke detection method of claim 6, wherein determining the plurality of smoke pixels as real smoke according to the disorderness and the growth-rate of the plurality of smoke pixels comprises:

determining the plurality of smoke pixels as real smoke when the number of times which the average growth-rate of the smoke pixels is greater than a threshold of growth-rate exceeds a predetermined value within a time duration;

wherein the average growth-rate of the smoke pixels is an average of the growth-rates of the smoke pixels;

each growth-rate of the smoke pixels denotes the difference between the numbers of smoke pixels at a first time slice and a second time slice divided by the time difference between the first time slice and the second time slice.

9. The smoke detection method of claim 1, further comprising:

giving an alarm when the plurality of smoke pixels is real smoke.

10. A smoke detection method based on video processing comprising:

capturing images in a monitored area and segmenting moving regions from the captured images, each moving region comprising a plurality of pixels;

determining real smoke from the moving regions wherein the real smoke comprising a plurality smoke pixels; and

giving an alarm according to the number of smoke pixels of the real smoke and the time slices for the real smoke spanning across by a fuzzy system.

11. The smoke detection method of claim 10, wherein capturing images in a monitored area and segmenting moving regions from the captured images comprises:

capturing images in the monitored area at a first time slice and a second time slice, wherein the first time slice and the second time slice are successive time slices with fixed interval; and

segmenting a moving region in a captured image according to the successive images'"'"' frame difference in the captured images.

12. The smoke detection method of claim 10, wherein determining real smoke from the moving regions comprises:

updating the number of the smoke pixels according to the color components'"'"' gray levels and the intensity (I) of HSI color model of the plurality of pixels; and

determining the plurality of smoke pixels as real smoke according to the disorderness and the growth-rate of the plurality of smoke pixels.

13. The smoke detection method of claim 12, wherein updating the number of the smoke pixels according to the color components'"'"' gray levels and the intensity (I) of HSI color model of the plurality of pixels comprises:

determining a pixel as a smoke pixel when red, green and blue gray levels of the pixel are approximately equal and when the intensity of the pixel is within a predetermined range.

14. The smoke detection method of claim 12, wherein determining the plurality of smoke pixels as real smoke according to the disorderness and the growth-rate of the plurality of smoke pixels comprises:

determining the plurality of smoke pixels as real smoke when the ratio of circumference of the region formed by the smoke pixels and the number of smoke pixels is greater than a threshold.

15. The smoke detection method of claim 12, wherein determining the plurality of smoke pixels as real smoke according to the disorderness and the growth-rate of the plurality of smoke pixels comprises:

determining the plurality of smoke pixels as real smoke when the number of times which the average growth-rate of the smoke pixels is greater than a threshold of growth-rate exceeds a predetermined value within a time duration;

wherein the average growth-rate of the smoke pixels is an average of the growth-rates of the smoke pixels;

each growth-rate of the smoke pixels denotes the difference between the numbers of smoke pixels at a first time slice and a second time slice divided by the time difference between the first time slice and the second time slice.

16. The smoke detection method of claim 10, wherein giving an alarm according to the number of smoke pixels of the real smoke and the time slices for the real smoke spanning across comprises:

generating a fuzzy result by the fuzzy system according to the number of smoke pixels of the real smoke and the time slices for the real smoke spanning across;

defuzzifying the fuzzy result for generating the alarm strength; and

giving the alarm according to the alarm strength.

17. The smoke detection method of claim 16, wherein generating a fuzzy result by a fuzzy system according to the number of smoke pixels of the real smoke and the time slices for the real smoke spanning across comprises:

performing the maximum operation on the number of smoke pixels of the real smoke and the time slices for the real smoke spanning across for generating the fuzzy result.

18. The smoke detection method of claim 16, wherein defuzzifying the fuzzy result for generating the alarm strength comprises:

defuzzifying the fuzzy result by weighted average defuzzification for generating the alarm strength.