CN110874907A - Flame identification method based on spectrum camera - Google Patents
Flame identification method based on spectrum camera Download PDFInfo
- Publication number
- CN110874907A CN110874907A CN201811020141.4A CN201811020141A CN110874907A CN 110874907 A CN110874907 A CN 110874907A CN 201811020141 A CN201811020141 A CN 201811020141A CN 110874907 A CN110874907 A CN 110874907A
- Authority
- CN
- China
- Prior art keywords
- spectrum
- camera
- flame
- method based
- spectral
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B17/00—Fire alarms; Alarms responsive to explosion
- G08B17/12—Actuation by presence of radiation or particles, e.g. of infrared radiation or of ions
- G08B17/125—Actuation by presence of radiation or particles, e.g. of infrared radiation or of ions by using a video camera to detect fire or smoke
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Fire-Detection Mechanisms (AREA)
- Fire Alarms (AREA)
Abstract
The invention relates to a flame identification method based on a spectral camera, which mainly solves the problems of poor interference capability and low sensitivity in the prior art. The invention adopts a flame identification method based on a spectrum camera, which comprises the following steps: (1) placing a spectral camera in an environment requiring real-time monitoring; (2) collecting a spectrum of a target position by using a spectrum camera; (3) measuring a target site spectrum, and evaluating the result when the relative intensity change rate of the spectrum exceeds 30%; (4) and (4) evaluating the result: the technical scheme is that the spectral information of the environment to be detected is monitored in real time, the relative intensity change rate of the spectrum exceeds 30%, a near-infrared false color image is generated once, and the possibility of open fire in the environment can be determined by combining video image analysis.
Description
Technical Field
The invention relates to a flame identification method based on a spectral range of 1200-2500nm spectral camera.
Background
At present, the economy of China is rapidly developed, the production scale of petrochemical products is continuously enlarged, on one hand, the quality of life of people is improved, on the other hand, greater potential safety hazards are brought, various petrochemical products can cause accidents such as fire disasters, explosions, poisoning and the like, and the serious chemical accidents caused by the fire disasters and the explosions account for more than seven. The petrochemical enterprises often use or produce flammable, explosive and toxic hazardous substances in the production and processing processes, great disasters can be caused by carelessness or improper operation, properties can be greatly lost, and more importantly, casualties can be caused, for example, the 8 & 12 special major accident of Tianjin gang Ruihai International Logistics Limited company causes general attention to the chemical fire and explosion hazards in various social circles. In chemical fire accidents, the characteristics of high propagation speed and more difficult fire extinguishing and rescue make the technology of fire identification very important.
CN201610702532.9 discloses a flame identification method for a dual infrared channel flame detector, which includes: the MCU main control unit collects flame data through a flame detector, collects contrast reference data through a reference infrared sensor and collects illuminance data through an illuminance sensor; calculating a discrete coefficient of the flame data, if the discrete coefficient is smaller than a first threshold value, not belonging to a fire alarm, otherwise, performing Fourier transform on the flame data, the comparison reference data and the illumination data, and solving an infrared correlation coefficient after Fourier transform of the flame data and the comparison reference data and a visible light correlation coefficient after Fourier transform of the flame data and the illumination data; and if the infrared correlation coefficient and the visible light correlation coefficient are both smaller than a second threshold value, the fire alarm belongs to a fire alarm, and the fire alarm information is sent through the wireless communication unit. The method carries out algorithm flame identification on the flame data collected by the sensors of the two infrared channels and the illuminance data collected by the visible light channel, and improves the anti-interference capability. Through careful comparison, the patent is found to adopt a pyroelectric infrared sensor with the size of 4.3 microns, in the process of early research, the inventor of the patent adopts an infrared spectrometer with a middle and far infrared band to observe the flame combustion condition, and finds that the high-temperature heat wave energy caused by flame has influence on the data measured by the band and influences the identification sensitivity because the flame combustion temperature is higher and the energy of the middle and far infrared band is gradually lower along with the extension of the wavelength.
At present, most of the methods for flame identification in China are established on the image analysis and processing technology, and no flame identification method based on the spectral camera (1200-2500nm) technology exists at present. Aiming at the characteristics of chemical fire accidents, the method for identifying flames is invented based on the use of a spectrum camera.
Disclosure of Invention
The invention aims to solve the technical problems of poor safety and reaction lag in the prior art, and provides a novel flame identification method based on a spectral camera, which has the advantages of good safety and timely reaction.
In order to solve the problems, the technical scheme adopted by the invention is as follows: a flame identification method based on a spectrum camera comprises the following steps: (1) placing a spectral camera in an environment requiring real-time monitoring; (2) collecting a spectrum of a target position by using a spectrum camera; (3) measuring a target site spectrum, and evaluating the result when the relative intensity change rate of the spectrum exceeds 30%; (4) and (4) evaluating the result: the spectral information of the environment to be detected is monitored in real time, the relative intensity change rate of the spectrum exceeds 30%, a near-infrared false color image is generated, and the possibility of open fire in the environment can be determined by combining video image analysis.
In the above technical solution, preferably, the spectral camera is installed in an environment that needs to be monitored in real time, and the exposure time is set to 0.5-2 seconds.
In the above technical solution, preferably, the background spectrum information of the target position is subtracted every 1-4 minutes.
In the above technical solution, preferably, the spectral camera is installed in an environment that needs to be monitored in real time, and the exposure time is set to 1 second.
In the above technical solution, preferably, the background spectrum information of the target position is subtracted every 2 minutes.
In the above technical solution, preferably, the spectral range of the spectral camera is 1200-2500 nm.
In the above technical solution, preferably, the flame recognition system based on the spectral camera is connected to a fire alarm system or a fire extinguishing system, and when it is confirmed that there is an open fire in the environment, the fire alarm system or the fire extinguishing system is started.
In the above technical solution, preferably, the flame recognition based on the spectral camera is used together with the flame recognition based on the image analysis, the spectral analysis quickly finds whether there is an open fire, and the image analysis immediately obtains the open fire position information.
The method for identifying the flame is invented based on the use of the spectrum camera, the spectrum camera is arranged on the ready-made fire which is easy to happen accidents by utilizing the spectrum detection principle of the spectrum camera, and the fire flame is detected to form the emitted spectrum by deducting the influence of the ambient background light on the detection, so that whether a fire source appears on the scene or not is judged. The invention firstly proposes that a flame self-luminous near infrared section (1200-; the near-infrared light energy of the wave band (1200-.
Drawings
FIG. 1 is a schematic flow diagram of the process of the present invention.
FIG. 2 is a near-infrared pseudo-color image obtained in a spectral range of 1200-2500 nm;
fig. 3 is a detection spectrum.
The present invention will be further illustrated by the following examples, but is not limited to these examples.
Detailed Description
[ example 1 ]
A flame recognition method based on a spectrum camera, as shown in FIG. 1, includes the following steps:
(1) a spectral camera (with a spectral range of 1200-2500nm) is placed in an environment requiring real-time monitoring, and the exposure time is set to 1 second.
(2) Spectral information of the target position was collected as background information every 2 minutes using a spectral camera.
(3) Background spectral information at the target site was subtracted every 2 minutes.
(4) And measuring the target site spectrum, and evaluating the result once the change rate of the spectrum information exceeds 30%.
(5) And (4) evaluating the result: and monitoring the spectral information of the environment to be detected in real time, and determining that open fire is likely to occur in the environment once the spectral information matched with the flame spectrum occurs.
The flame recognition system based on the spectrum camera is connected with the fire alarm system or the fire extinguishing system, and when open fire is confirmed in the environment, the fire alarm system or the fire extinguishing system is started; flame identification based on a spectrum camera and flame identification based on image analysis are used together, spectrum analysis quickly finds whether open fire exists, and image analysis is immediately carried out to obtain or verify open fire position information.
[ example 2 ]
A flame recognition method based on a spectrum camera, as shown in FIG. 1, includes the following steps:
(1) a spectral camera (with a spectral range of 1200-2500nm) is placed in an environment needing real-time monitoring, and the exposure time is set to be 0.5 second.
(2) Spectral information of the target position was collected as background information every 1 minute with a spectral camera.
(3) Background spectral information at the target site was subtracted every 1 minute.
(4) And measuring the target site spectrum, and evaluating the result once the change rate of the spectrum information exceeds 30%.
(5) And (4) evaluating the result: and monitoring the spectral information of the environment to be detected in real time, and determining that open fire is likely to occur in the environment once the spectral information matched with the flame spectrum occurs.
The flame recognition system based on the spectrum camera is connected with the fire alarm system or the fire extinguishing system, and when open fire is confirmed in the environment, the fire alarm system or the fire extinguishing system is started; flame identification based on a spectrum camera and flame identification based on image analysis are used together, spectrum analysis quickly finds whether open fire exists, and image analysis is immediately carried out to obtain or verify open fire position information.
[ example 3 ]
A flame recognition method based on a spectrum camera, as shown in FIG. 1, includes the following steps:
(1) a spectral camera (with a spectral range of 1200-2500nm) is placed in an environment requiring real-time monitoring, and the exposure time is set for 2 seconds.
(2) Spectral information of the target position was collected as background information every 2 minutes using a spectral camera.
(3) Background spectral information at the target site was subtracted every 4 minutes.
(4) And measuring the target site spectrum, and evaluating the result once the change rate of the spectrum information exceeds 30%.
(5) And (4) evaluating the result: and monitoring the spectral information of the environment to be detected in real time, and determining that open fire is likely to occur in the environment once the spectral information matched with the flame spectrum occurs.
The flame recognition system based on the spectrum camera is connected with the fire alarm system or the fire extinguishing system, and when open fire is confirmed in the environment, the fire alarm system or the fire extinguishing system is started; flame identification based on a spectrum camera and flame identification based on image analysis are used together, spectrum analysis quickly finds whether open fire exists, and image analysis is immediately carried out to obtain or verify open fire position information.
[ example 4 ]
A flame recognition method based on a spectrum camera, as shown in FIG. 1, includes the following steps:
(1) a spectral camera (with a spectral range of 1200-2500nm) is placed in an environment requiring real-time monitoring, and the exposure time is set for 2 seconds.
(2) Spectral information of the target position was collected as background information every 2 minutes using a spectral camera.
(3) Background spectral information at the target site was subtracted every 4 minutes.
(4) And measuring the target site spectrum, and evaluating the result once the change rate of the spectrum information exceeds 30%.
(5) And (4) evaluating the result: the spectral information of the environment to be detected is monitored in real time, the relative intensity change rate of the spectrum exceeds 30%, a near-infrared false color image is generated, and the possibility of open fire in the environment can be determined by combining video image analysis.
The flame recognition system based on the spectrum camera is connected with the fire alarm system or the fire extinguishing system, and when open fire is confirmed in the environment, the fire alarm system or the fire extinguishing system is started; flame identification based on a spectrum camera and flame identification based on image analysis are used together, spectrum analysis quickly finds whether open fire exists, and image analysis is immediately carried out to obtain or verify open fire position information.
[ example 5 ]
The obtained near-infrared pseudo-color image is shown in fig. 2 according to the conditions and the steps described in the example 1, and it can be seen that after a fire source exists, a flame can be obviously seen on the near-infrared pseudo-color image, the flame is very intuitive, and the flame cannot be seen before the fire occurs. Fig. 3 shows the obtained detection spectrum, and the spectrum before and after ignition has obvious difference and can be used for flame identification.
Claims (8)
1. A flame identification method based on a spectrum camera comprises the following steps: (1) placing a spectral camera in an environment requiring real-time monitoring; (2) collecting a spectrum of a target position by using a spectrum camera; (3) measuring a target site spectrum, and evaluating the result when the relative intensity change rate of the spectrum exceeds 30%; (4) and (4) evaluating the result: the spectral information of the environment to be detected is monitored in real time, the relative intensity change rate of the spectrum exceeds 30%, a near-infrared false color image is generated, and the possibility of open fire in the environment can be determined by combining video image analysis.
2. The flame recognition method based on spectrum camera as claimed in claim 1, wherein the spectrum camera is installed in the environment requiring real-time monitoring, and the exposure time is set to 0.5-2 seconds.
3. The flame recognition method based on spectral camera of claim 1, wherein the background spectral information of the target location is subtracted every 1-4 minutes.
4. The flame recognition method based on spectrum camera as claimed in claim 2, wherein the spectrum camera is installed in the environment requiring real-time monitoring, and the exposure time is set to 1 second.
5. The flame recognition method based on spectral camera of claim 3, wherein the background spectral information of the target location is subtracted every 2 minutes.
6. The flame recognition method based on spectrum camera as claimed in claim 1, wherein the spectrum range of the spectrum camera is 1200-2500 nm.
7. The flame recognition method based on the spectral camera of claim 1, wherein the flame recognition system based on the spectral camera is connected to a fire alarm system or a fire extinguishing system, and the fire alarm system or the fire extinguishing system is activated when an open fire is confirmed in the environment.
8. The flame recognition method based on the spectrum camera as claimed in claim 1, wherein the flame recognition based on the spectrum camera is used together with the flame recognition based on the image analysis, the spectrum analysis rapidly finds whether there is an open fire, and the image analysis obtains the open fire position information immediately.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811020141.4A CN110874907A (en) | 2018-09-03 | 2018-09-03 | Flame identification method based on spectrum camera |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811020141.4A CN110874907A (en) | 2018-09-03 | 2018-09-03 | Flame identification method based on spectrum camera |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110874907A true CN110874907A (en) | 2020-03-10 |
Family
ID=69716863
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811020141.4A Pending CN110874907A (en) | 2018-09-03 | 2018-09-03 | Flame identification method based on spectrum camera |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110874907A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN119380480A (en) * | 2024-12-30 | 2025-01-28 | 南昌南飞防火设备制造有限公司 | Fire shutter alarm control system and method based on Internet of Things |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| SU1339610A1 (en) * | 1986-05-11 | 1987-09-23 | Предприятие П/Я А-3462 | Light detector |
| US5937077A (en) * | 1996-04-25 | 1999-08-10 | General Monitors, Incorporated | Imaging flame detection system |
| GB0006257D0 (en) * | 2000-03-15 | 2000-05-03 | Graviner Ltd Kidde | Image detection |
| CN101046908A (en) * | 2007-05-08 | 2007-10-03 | 中国科学院上海技术物理研究所 | Forest fire behavior dynamic monitoring alarm system based on infrared camera |
| CN101609589A (en) * | 2008-06-17 | 2009-12-23 | 侯荣琴 | Multi-frequency image fire detection system |
| US20110058037A1 (en) * | 2008-04-25 | 2011-03-10 | Thomas Hanses | Fire detection device and method for fire detection |
| US7956761B2 (en) * | 2007-05-29 | 2011-06-07 | The Aerospace Corporation | Infrared gas detection and spectral analysis method |
| CN102143328A (en) * | 2009-10-07 | 2011-08-03 | 约翰新科公司 | Image sensing system, software, apparatus and method for controlling combustion equipment |
| CN104240429A (en) * | 2014-08-27 | 2014-12-24 | 中国资源卫星应用中心 | Method for monitoring forest fire by means of combination of infrared and multi-spectral wave bands |
| CN206292939U (en) * | 2016-12-16 | 2017-06-30 | 君都智能科技(上海)有限公司 | A kind of spectrum flame detector based on flicker frequency |
| US9759628B2 (en) * | 2009-07-23 | 2017-09-12 | International Electronic Machines Corporation | Area monitoring for detection of leaks and/or flames |
| CN107251118A (en) * | 2014-12-17 | 2017-10-13 | 霍尼韦尔国际公司 | The detecting system and method being characterized with multispectral imaging equipment |
-
2018
- 2018-09-03 CN CN201811020141.4A patent/CN110874907A/en active Pending
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| SU1339610A1 (en) * | 1986-05-11 | 1987-09-23 | Предприятие П/Я А-3462 | Light detector |
| US5937077A (en) * | 1996-04-25 | 1999-08-10 | General Monitors, Incorporated | Imaging flame detection system |
| GB0006257D0 (en) * | 2000-03-15 | 2000-05-03 | Graviner Ltd Kidde | Image detection |
| CN101046908A (en) * | 2007-05-08 | 2007-10-03 | 中国科学院上海技术物理研究所 | Forest fire behavior dynamic monitoring alarm system based on infrared camera |
| US7956761B2 (en) * | 2007-05-29 | 2011-06-07 | The Aerospace Corporation | Infrared gas detection and spectral analysis method |
| US20110058037A1 (en) * | 2008-04-25 | 2011-03-10 | Thomas Hanses | Fire detection device and method for fire detection |
| CN101609589A (en) * | 2008-06-17 | 2009-12-23 | 侯荣琴 | Multi-frequency image fire detection system |
| US9759628B2 (en) * | 2009-07-23 | 2017-09-12 | International Electronic Machines Corporation | Area monitoring for detection of leaks and/or flames |
| CN102143328A (en) * | 2009-10-07 | 2011-08-03 | 约翰新科公司 | Image sensing system, software, apparatus and method for controlling combustion equipment |
| CN104240429A (en) * | 2014-08-27 | 2014-12-24 | 中国资源卫星应用中心 | Method for monitoring forest fire by means of combination of infrared and multi-spectral wave bands |
| CN107251118A (en) * | 2014-12-17 | 2017-10-13 | 霍尼韦尔国际公司 | The detecting system and method being characterized with multispectral imaging equipment |
| CN206292939U (en) * | 2016-12-16 | 2017-06-30 | 君都智能科技(上海)有限公司 | A kind of spectrum flame detector based on flicker frequency |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN119380480A (en) * | 2024-12-30 | 2025-01-28 | 南昌南飞防火设备制造有限公司 | Fire shutter alarm control system and method based on Internet of Things |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111111074B (en) | Fire extinguishing scheduling method and system for power tunnel fire-fighting robot | |
| CN101251942B (en) | Method and device for intelligent detection, early warning and forecasting of fire in underground space | |
| CN108325123B (en) | Automobile fire escape self-rescue device and control method | |
| CN106448023B (en) | Fire smoke alarm with storage function | |
| CN205140120U (en) | Long -range monitoring and forewarning system of thing networking image conflagration | |
| CN105336085A (en) | Remote large-space fire monitoring alarm method based on image processing technology | |
| CN205845241U (en) | A kind of intelligent multifunctional alarm device realizing high in the clouds monitoring | |
| CA3210157A1 (en) | Sensor fusion for fire detection & air quality monitoring | |
| CN110910604B (en) | Monitoring method, system and device for fire alarm | |
| CN113936239A (en) | A method and system for intelligent fire identification based on neural network algorithm | |
| CN110801593B (en) | Extremely early fire early warning system and method fusing multi-mode data | |
| CN104318688A (en) | Multi-sensor fire early-warning method based on data fusion | |
| CN205788636U (en) | Intelligent wireless fire detection and alarm system | |
| CN116524666A (en) | Fire monitoring and early warning system based on data analysis | |
| Sharma et al. | Development of an early detection system for fire using Wireless Sensor Networks and Arduino | |
| CN210271155U (en) | Composite smoke-sensing detection labyrinth | |
| CN202033865U (en) | Ultraviolet and infrared composite flame detector | |
| CA2626753C (en) | A method for detecting a fire condition in a monitored region | |
| CN102496235A (en) | Image fire detector possessing optical integrity and method thereof | |
| CN110874907A (en) | Flame identification method based on spectrum camera | |
| CN208405818U (en) | A kind of automobile fire escape self-help apparatus | |
| CN105225428A (en) | A kind of indoor smoke detection alarm method and system | |
| CN113205562B (en) | Mine thermodynamic disaster identification and positioning method based on binocular vision | |
| CN117292311B (en) | Fire-fighting fire re-combustion prediction system based on AI image recognition | |
| CN116989897A (en) | Low-power-consumption vehicle-mounted remote abnormality monitoring system for dangerous waste warehouse |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: 100010 22 Chaoyangmen North Street, Chaoyang District, Beijing Applicant after: CHINA PETROLEUM & CHEMICAL Corp. Applicant after: SINOPEC Research Institute OF SAFETY ENGINEERING Applicant after: Chemical registration center of emergency management department Address before: 100010 22 Chaoyangmen North Street, Chaoyang District, Beijing Applicant before: China Petroleum & Chemical Corp. Applicant before: SINOPEC Research Institute OF SAFETY ENGINEERING Applicant before: NATIONAL REGISTRATION CENTER FOR CHEMICALS, SAWS |
|
| CB02 | Change of applicant information | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200310 |
|
| RJ01 | Rejection of invention patent application after publication |