Methods and systems for fire detection

Patent metadata

Field	Value
Publication number	US-11532156-B2
Application number	US-201916578228-A
Country	US
Kind code	B2
Filing date	Sep 20, 2019
Priority date	Mar 28, 2017
Publication date	Dec 20, 2022
Grant date	Dec 20, 2022

Abstract

In accordance with various aspects of the present disclosure, methods and systems for fire detection are disclosed. In some embodiments, a method for fire detection includes: acquiring data related to a monitored area, wherein the data comprises image data related to the monitored area; determining, based on the acquired data, whether a first mode or a second mode is to be executed for fire detection, wherein the first mode comprises a first smoke detection, and the second mode comprises a first flame detection; and executing, by a hardware processor, at least one of the first mode or the second mode based on a result of the determination.

First claim

What is claimed is:

1. A method for fire detection, comprising:
acquiring image data related to a monitored area;
determining, based on a time at which the image data is captured or grey values of the acquired image data, whether a first mode corresponding to daytime or a second mode corresponding to nighttime is to be executed for fire detection, wherein the first mode comprises a first smoke detection and a second smoke detection, and the second mode comprises a first flame detection;
wherein performing the first smoke detection comprises:
updating, based on the image data, a first background image at a first update rate;
updating, based on the image data, a second background image at a second update rate that exceeds the first update rate;
generating a difference image between the updated first background image and the updated second background image;
identifying a candidate smoke region from the difference image, pixel values of pixels in the candidate smoke region exceed a threshold; and
determining whether the second smoke detection needs to be executed by determining whether there is a risk of fire in the candidate smoke region by
conducting color analysis on the candidate smoke region.

2. The method of claim 1 , wherein the determining whether the first mode or the second mode is to be executed comprises:
determining, based on the time at which the image data is captured or the grey values of the acquired image data, whether the acquired image data corresponds to daytime or nighttime; and
determining whether the first mode corresponding to daytime or the second mode corresponding to nighttime is to be executed based on the determination of whether the acquired image data corresponds to daytime or nighttime.

3. The method of claim 1 , wherein the performing the second smoke detection comprises:
dividing the candidate smoke region into a plurality of blocks;
conducting at least one of a smoke energy analysis or a smoke diffusion analysis on the plurality of blocks; and
determining a potential smoke region based on a result of the at least one of the smoke energy analysis or the smoke diffusion analysis.

4. The method of claim 3 , wherein the second smoke detection further comprises determining whether the potential smoke region is a smoke region by processing a portion of the image data that relates to the potential smoke region using a classifier.

5. The method of claim 1 , wherein the first flame detection comprises:
generating a background image based on the image data;
generating a highlight background region based on the background image;
producing a foreground image based on the background image; and
identifying a candidate flame region based on at least one of the highlight background region or the foreground image.

6. The method of claim 5 , wherein the second mode further comprises a second flame detection, and wherein the second flame detection comprises:
acquiring a highlight edge region related to the highlight background region;
identifying a foreground accumulation image based on the candidate flame region;
determining a flicker foreground image based on at least one of the highlight edge region or the foreground accumulation image; and
processing image data related to the flicker foreground image using a classifier.

7. The method of claim 1 , further comprising identifying a portion of the image data corresponding to an area for the fire detection by performing a skyline detection on the image data.

8. The method of claim 1 , wherein the identifying a candidate smoke region based on the difference image comprises:
determining whether one or more smoke pixels exist in the difference image by comparing the pixel value of each pixel of the difference image with the threshold;
in response to determining that one or more smoke pixels exist in the difference image, generating one or more foreground images based on the difference image; and
identifying the candidate smoke region based on the one or more foreground images.

9. The method of claim 8 , wherein:
the one or more foreground images comprise a plurality of foreground images, and
the identifying the candidate smoke region based on the one or more foreground images comprises identifying the candidate smoke region by accumulating the plurality of foreground images.

10. A system for fire detection, comprising:
a storage device; and
a hardware processor operatively coupled to the storage device and configured to:
acquire image data related to a monitored area;
determine, based on a time at which the image data is captured or grey values of the acquired image data, whether a first mode corresponding to daytime or a second mode corresponding to nighttime is to be executed for fire detection, wherein the first mode comprises a first smoke detection and a second smoke detection;
wherein performing the first smoke detection comprises:
updating, based on the image data, a first background image at a first update rate;

updating, based on the image data, second background image at a second update rate that exceeds the first update rate;

generating a difference image between the updated first background image and the updated second background image;
identifying a candidate smoke region based on the difference image, pixel values of pixels in the candidate smoke region exceed a threshold; and
determining whether the second smoke detection needs to be executed by determining whether there is a risk of fire in the candidate smoke region by
conducting color analysis on the candidate smoke region.

11. The system of claim 10 , wherein to determine whether the first mode or the second mode is to be executed, the hardware processor is further configured to determine, based on the time at which the image data is captured or the grey values of the acquired image data, whether the acquired image data corresponds to daytime or nighttime; and
determining whether the first mode corresponding to daytime or the second mode corresponding to nighttime is to be executed based on the determination of whether the acquired image data corresponds to daytime or nighttime.

12. The system of claim 10 , wherein the performing the second smoke detection comprises:
dividing the candidate smoke region into a plurality of blocks;
conducting at least one of a smoke energy analysis or a smoke diffusion analysis on the plurality of blocks; and
determining a potential smoke region based on a result of the at least one of the smoke energy analysis or the smoke diffusion analysis.

13. The system of claim 12 , wherein to conduct the second smoke detection, the hardware processor is further configured to determine whether the potential smoke region is a smoke region by processing a portion of the image data that relates to the potential smoke region using a classifier.

14. The system of claim 10 , wherein to conduct the first flame detection, the hardware processor is further configured to:
generate a background image based on the image data;
generate a highlight background region based on the background image;
produce a foreground image based on the background image; and
identify a candidate flame region based on at least one of the highlight background region or the foreground image.

15. The system of claim 14 , wherein to execute the second mode, the hardware processor is further configured to conduct a second flame detection by:
acquiring a highlight edge region related to the highlight background region;
generating a foreground accumulation image based on the candidate flame region;
determining a flicker foregroun

Assignees

• Zhejiang Dahua Technology Co

Inventors

• Zheng Jia
• Tian Jianguo
• Pan Huadong

Classifications

• G08B17/125 Primary

  by using a video camera to detect fire or smoke · CPC title

• G06V10/764

  using classification, e.g. of video objects · CPC title

• G06V20/20 Primary

  in augmented reality scenes · CPC title

• G06F18/2433

  Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection · CPC title

• G08B25/08

  using communication transmission lines {(G08B13/19658, G08B21/0286, G08B25/016 take precedence)} · CPC title