CN203204762U - A fire flame video early warning system - Google Patents
A fire flame video early warning system Download PDFInfo
- Publication number
- CN203204762U CN203204762U CN 201320090938 CN201320090938U CN203204762U CN 203204762 U CN203204762 U CN 203204762U CN 201320090938 CN201320090938 CN 201320090938 CN 201320090938 U CN201320090938 U CN 201320090938U CN 203204762 U CN203204762 U CN 203204762U
- Authority
- CN
- China
- Prior art keywords
- fire
- early warning
- video
- fire disaster
- flame
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Fire-Detection Mechanisms (AREA)
- Alarm Systems (AREA)
Abstract
The utility model discloses a fire flame video early warning system comprising a network video CCD camera, a video management module, a fire flame detector, and an early warning device. A fire flame detecting method is to perform information fusion of the static and dynamic characteristics of fire flame and to provide a fire discriminating algorithm based on a neural network and D-S evidence judgment. The fire discriminating algorithm is capable of providing three early warning threshold values according to fire degree. The three early warning threshold values correspond to early warning modes of three grades including onsite acousto-optic alarm, remote short message alarm, and network alarm. The fire flame video early warning system monitors tiny fire source at an initial stage of fire hazard and discriminates fire flame grades on the basis of multiple characteristic information fusion. The fire flame video early warning system has characteristics of a strong anti-interference capability, strong space adaptability, high fire early warning accuracy, and low false alarm rate.
Description
Technical field
The utility model belongs to public safety field, relates to the fire monitoring early warning system, is specifically related to a kind of fire disaster flame video preprocessor alarm system.
Background technology
Fire is the major hidden danger that threatens human survival, and especially in recent years, the fire that takes place frequently has caused great Loss of Life and property to people.According to statistics, China was total to 125402 of breaking out of fires in 2011, caused 1106 people dead, 19.8 hundred million yuan of direct economic losses.The wildfire that in June, 2011, Shanghai occured causes 58 people dead, 1.58 hundred million yuan of direct economic losses.Find in time the fire source and in time take effective fire protection to put out incipient flame, can greatly reduce various losses.
Fire detector has approximately experienced from the development course of heat detector → ion smoke detector → optical detector of fire smoke → air-sensitive detector → infrared eye → video detector as the core component of fire early-warning system.Be mainly used in indoor fire detection among a small circle as the temperature-sensitive of traditional fire detector, sense cigarette, air-sensitive class detector, have that the space applicability is limited, information reliability is poor, be subject to the shortcoming such as external interference.And infrared eye is expensive, has limited its usable range.Video detector is a kind of take the fire detector of image processing techniques as the basis, utilize camera supervised monitoring point sight, by the video information that catches is processed, analyzed, and then realize the detection of fire disaster flame and the early warning of fire, antijamming capability is stronger, and the space applicability is strong.In the video fire hazard early warning technology, the fire disaster flame detection method is core, although domestic expert was carrying out certain research and development to the video fire hazard flame detecting method in recent years, proven technique rarely has application, and corresponding product also seldom emerges.
Summary of the invention
The purpose of this utility model is: a kind of fire disaster flame video preprocessor alarm system is provided, it is the fire disaster flame video frequency monitoring system of the trickle burning things which may cause a fire disaster of monitoring during Initial Stage of Fire, and merge based on multicharacteristic information and to carry out the fire disaster flame rank and differentiate, antijamming capability is strong, the space applicability is strong, the fire alarm accuracy is high, and rate of false alarm is low.
Technical solution of the present utility model is: fire disaster flame video monitor and early warning system adopts the distributed system architecture of Internet of Things to dispose, and connects to form early warning system electricity loop by Internet video CCD camera, video management module, fire disaster flame detecting device and precaution device according to circuit mode; Described Internet video CCD camera is responsible for finishing imaging, supports the WiFi wireless transmission, supports simultaneously TCP/IP network remote image transmitting, and has the detection of dynamic recording function; Described video management module adopts the generic video administration module; Described fire disaster flame detecting device is obtained the realtime graphic of Internet video CCD camera by image pick-up card, process obtaining the judgement conclusion that image provides the condition of a fire with Digital Image Processor; Described precaution device adopts three grades of modes of warnings, i.e. on-the-spot acousto-optic early warning, remote short messages breath early warning and Network Warning; When the one-level condition of a fire occured, the voice guard and the warning light that are installed in the monitoring point in monitored district were responsible for realizing on-the-spot acousto-optic early warning, and alarm command is independently assigned by the monitoring site control module; When finding the secondary condition of a fire by Java Servlet real time access early warning database and in the monitoring point, the Android handset server ceases early warning to specifying the person liable to send the early warning short message to realize remote short messages; Wireless transmission is utilized the WiFi technology, and the early warning data transfer adopts encrypted transmission, guarantees the security of data; When the condition of a fire of passing on when the fire disaster flame detection system reaches Senior Three level, jump out warning window and picture, realize network alarming, and it is live to transmit on-the-spot fire.
The utility model also proposes a kind of fire disaster flame detection method, Static and dynamic feature to fire disaster flame is carried out information fusion, provide a kind of fire based on neural network and the judgement of D-S evidence and differentiate algorithm, this algorithm can provide three threshold value of warning according to condition of a fire size, and three the corresponding on-the-spot sound and light alarms of threshold value of warning, long-range short messages are reported to the police and these three other modes of warnings of level of network alarming; Specific as follows:
Step 1: described Internet video CCD camera is finished field imaging, and the digital picture at the scene that image pick-up card acquisition Internet video CCD camera transmits utilizes digital image processing techniques that this digital picture is carried out pre-service;
Step 2: utilize gray scale thresholding technology by the motion feature of the variation detection of fires flame of the gray scale thresholding of two two field pictures before and after comparing; The area change feature whether the growth detection fire disaster flame is arranged by two two field picture flame area before and after comparing; Flame detection color characteristic in the HIS color space; Realize differentiation to naturally light and daylight lamp monochrome information background interference by the circularity feature of extracting fire disaster flame; Extract the wedge angle number of fire disaster flame, the wedge angle number that utilizes fire disaster flame along with the characteristics that present irregular rule burning time with the variation of the wedge angle number of fire disaster flame as one of Fire Criterion; Fire disaster flame burning initial stage predominant frequency is in 12Hz, and the variation of surrounding environment is not remarkable to frequency influence, therefore extracts the blinking characteristics of fire disaster flame as one of Fire Criterion;
Step 3: gather the fire disaster flame data sample of monitoring point, select in advance weights and the threshold value of the fire disaster flame data sample of monitoring point being trained these four kinds of fire disaster flame features of flicker frequency feature of the wedge angle number of the circularity feature that obtains to comprise fire disaster flame color characteristic, fire disaster flame, fire disaster flame and fire disaster flame;
Step 4: utilize formula (1) to obtain the fire disaster flame threshold value of warning:
In the formula:
Be
iIndividual fire disaster flame feature output,
,
H 0Expression is without fire disaster flame,
H 1Expression has fire disaster flame;
,
m i (H
0) and
m i (
H 1) be
iThe detection probability of individual fire disaster flame feature and probability of miscarriage of justice;
Step 5: utilize above-mentioned detected real-time threshold value and empirical value to realize that relatively fire disaster flame detects and early warning in real time.
The utility model has the advantages that:
1, adopt the distributed system architecture of Internet of Things to dispose, the space applicability is strong, both can be used for warehouse, indoor fire monitoring and the early warning among a small circle of laboratory class, also can be used for the monitoring of forest class large-range fire disaster and early warning.
2, Adoption Network video CCD camera obtains the monitoring point image, utilizes digital image processing techniques to make early warning system that higher antijamming capability and networked Characteristics be arranged, and to compare cost performance higher with infrared fire hazard monitoring system.
3, the Static and dynamic feature of fire disaster flame is carried out information fusion, provide a kind of fire based on neural network and the judgement of D-S evidence and differentiate algorithm, the early warning of energy Quick, False Rate is lower.
4, adopt three grades of fire disaster flame early warning subsystems, can realize that early warning system is intelligent according to the real-time flame in scene, scientific and reasonable and manpower, financial resources when dispatching in an orderly manner fire and occuring.
Description of drawings
Fig. 1 is fire disaster flame video preprocessor alarm system structural representation of the present utility model.
Fig. 2 is fire disaster flame image detection process flow diagram.
Fig. 3 is many features fire disaster flame detection method schematic diagram.
Embodiment
Below in conjunction with description of drawings technical solution of the present utility model.
As shown in Figure 1, fire disaster flame video monitor and early warning system adopts the distributed system architecture of Internet of Things to dispose, and connects to form early warning system electricity loop by Internet video CCD camera, video management module, fire disaster flame detecting device and precaution device according to circuit mode; Described Internet video CCD camera is responsible for finishing imaging, supports the WiFi wireless transmission, supports simultaneously TCP/IP network remote image transmitting, and has the detection of dynamic recording function; Described video management module adopts the generic video administration module; Described fire disaster flame detecting device is obtained the realtime graphic of Internet video CCD camera by image pick-up card, process obtaining the judgement conclusion that image provides the condition of a fire with Digital Image Processor; Described precaution device adopts three grades of modes of warnings, i.e. on-the-spot acousto-optic early warning, remote short messages breath early warning and Network Warning; When the one-level condition of a fire occured, the voice guard and the warning light that are installed in the monitoring point in monitored district were responsible for realizing on-the-spot acousto-optic early warning, and alarm command is independently assigned by the monitoring site control module; When finding the secondary condition of a fire by Java Servlet real time access early warning database and in the monitoring point, the Android handset server ceases early warning to specifying the person liable to send the early warning short message to realize remote short messages; Wireless transmission is utilized the WiFi technology, and the early warning data transfer adopts encrypted transmission, guarantees the security of data; When the condition of a fire of passing on when the fire disaster flame detection system reaches Senior Three level, jump out warning window and picture, realize network alarming, and it is live to transmit on-the-spot fire.
Shown in Fig. 2,3, the fire disaster flame detection method is that the Static and dynamic feature of fire disaster flame is carried out information fusion, provide a kind of fire based on neural network and the judgement of D-S evidence and differentiate algorithm, this algorithm can provide three threshold value of warning according to condition of a fire size, and three the corresponding on-the-spot sound and light alarms of threshold value of warning, long-range short messages are reported to the police and these three other modes of warnings of level of network alarming; Specific as follows:
Step 1: described Internet video CCD camera is finished field imaging, and the digital picture at the scene that image pick-up card acquisition Internet video CCD camera transmits utilizes digital image processing techniques that this digital picture is carried out pre-service;
Step 2: utilize gray scale thresholding technology by the motion feature of the variation detection of fires flame of the gray scale thresholding of two two field pictures before and after comparing; The area change feature whether the growth detection fire disaster flame is arranged by two two field picture flame area before and after comparing; Flame detection color characteristic in the HIS color space; Realize differentiation to naturally light and daylight lamp monochrome information background interference by the circularity feature of extracting fire disaster flame; Extract the wedge angle number of fire disaster flame, the wedge angle number that utilizes fire disaster flame along with the characteristics that present irregular rule burning time with the variation of the wedge angle number of fire disaster flame as one of Fire Criterion; Fire disaster flame burning initial stage predominant frequency is in 12Hz, and the variation of surrounding environment is not remarkable to frequency influence, therefore extracts the blinking characteristics of fire disaster flame as one of Fire Criterion;
Step 3: gather the fire disaster flame data sample of monitoring point, select in advance weights and the threshold value of the fire disaster flame data sample of monitoring point being trained these four kinds of fire disaster flame features of flicker frequency feature of the wedge angle number of the circularity feature that obtains to comprise fire disaster flame color characteristic, fire disaster flame, fire disaster flame and fire disaster flame;
Step 4: utilize formula (1) to obtain the fire disaster flame threshold value of warning:
In the formula:
Be
iIndividual fire disaster flame feature output,
,
H 0Expression is without fire disaster flame,
H 1Expression has fire disaster flame;
,
m i (H
0) and
m i (
H 1) be
iThe detection probability of individual fire disaster flame feature and probability of miscarriage of justice;
Step 5: utilize above-mentioned detected real-time threshold value and empirical value to realize that relatively fire disaster flame detects and early warning in real time.
The fire disaster flame early warning system is divided into perception terminal, data collection and analysis layer and three parts of terminal application platform, perception terminal comprises by The Cloud Terrace and is installed in high resolution CCD video camera, voice guard and warning light in the monitoring point in monitored district, the data collection and analysis layer comprises video frequency collection card, Mysql database and the Android handset server in the computing machine that is installed in each monitoring point, and terminal application platform is to have being connected to the Internet and having the computing machine of public network IP address of data processing function; Ccd video camera is responsible for finishing sight imaging in the monitoring point, image pick-up card gathers image that ccd video camera becomes, by the video management module digital picture that reads is carried out pre-service successively, each feature extraction utilizes multicharacteristic information blending algorithm output early warning information; Adopt three grades of modes of warnings according to condition of a fire size, comprise on-the-spot acousto-optic early warning, long-range short message early warning and Network Warning; During Initial Stage of Fire, alarm command is independently assigned or is assigned by Surveillance center by the monitoring site control module, the voice guard and the warning light that are installed in the monitoring point in monitored district are responsible for realizing on-the-spot acousto-optic early warning, select to start simple and easy fire protection and eliminate the condition of a fire, or near being positioned at the monitoring point, can hear that the person liable of alarm song adopts manual type to eliminate the condition of a fire; If the condition of a fire does not weaken, then start long-range short message modes of warning, to specifying the person liable to send the early warning short message, this moment, on-the-spot sound and light alarm was still effective when the Android handset server was found the condition of a fire by Java Servlet real time access early warning database and in the monitoring point; Specify the person liable to the site disposal condition of a fire, if the intensity of a fire is still uncontrollable, start network alarming this moment, jump out warning window and picture, transmit simultaneously the far-end video information, select to be reported to the police or automatic 110 warnings of system by the staff 110 of Surveillance center, employ the fire extinguishing of public fire protection facility.
Claims (1)
1. fire disaster flame video preprocessor alarm system, it is characterized in that: fire disaster flame video monitor and early warning system adopts the distributed system architecture of Internet of Things to dispose, and connects to form early warning system electricity loop by Internet video CCD camera, video management module, fire disaster flame detecting device and precaution device according to circuit mode; Described Internet video CCD camera is responsible for finishing imaging, supports the WiFi wireless transmission, supports simultaneously TCP/IP network remote image transmitting, and has the detection of dynamic recording function; Described video management module adopts the generic video administration module; Described fire disaster flame detecting device is obtained the realtime graphic of Internet video CCD camera by image pick-up card, process obtaining the judgement conclusion that image provides the condition of a fire with Digital Image Processor; Described precaution device adopts three grades of modes of warnings, i.e. on-the-spot acousto-optic early warning, remote short messages breath early warning and Network Warning.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201320090938 CN203204762U (en) | 2013-02-28 | 2013-02-28 | A fire flame video early warning system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201320090938 CN203204762U (en) | 2013-02-28 | 2013-02-28 | A fire flame video early warning system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN203204762U true CN203204762U (en) | 2013-09-18 |
Family
ID=49148851
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201320090938 Expired - Fee Related CN203204762U (en) | 2013-02-28 | 2013-02-28 | A fire flame video early warning system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN203204762U (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111127815A (en) * | 2019-12-24 | 2020-05-08 | 嘉兴恒创电力设计研究院有限公司 | Fire-fighting system and method with hierarchical early warning control |
| CN111420334A (en) * | 2020-04-03 | 2020-07-17 | 西华大学 | Internet-based household intelligent mobile fire-fighting measurement and control system and use method |
-
2013
- 2013-02-28 CN CN 201320090938 patent/CN203204762U/en not_active Expired - Fee Related
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111127815A (en) * | 2019-12-24 | 2020-05-08 | 嘉兴恒创电力设计研究院有限公司 | Fire-fighting system and method with hierarchical early warning control |
| CN111420334A (en) * | 2020-04-03 | 2020-07-17 | 西华大学 | Internet-based household intelligent mobile fire-fighting measurement and control system and use method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103150856B (en) | Fire flame video monitoring and early warning system | |
| CN204288415U (en) | Intelligent early-warning firefighting robot | |
| CN104299351A (en) | Intelligent early warning and fire extinguishing robot | |
| CN107911653B (en) | Intelligent video monitoring module, system, method and storage medium for residence | |
| CN100504942C (en) | Module set of intelligent video monitoring device, system and monitoring method | |
| CN104079874B (en) | A kind of security protection integral system and method based on technology of Internet of things | |
| CN201936415U (en) | Automatic forest fire identification and alarm system | |
| CN102176270A (en) | Safety monitoring and fire alarming integrated system and method | |
| CN201897856U (en) | Safety monitoring and fire hazard early-warning integrated system | |
| CN205068666U (en) | Initiatively take precautions against linkage of formula video and urge tear security protection system | |
| CN110517441A (en) | Based on the frame-embedded smog of deep learning and flame video alarming system and method | |
| CN201237824Y (en) | Intelligent monitoring alarm system for oil well | |
| CN106454253A (en) | Method and system for detecting area wandering | |
| CN111899452A (en) | Forest fire prevention early warning system based on edge calculation | |
| CN212782246U (en) | Smoke and fire recognition system based on artificial intelligence | |
| CN111754714A (en) | Security monitoring system and monitoring method thereof | |
| CN107958562A (en) | A kind of household safe Active Defending System Against and method | |
| CN110519560B (en) | Intelligent early warning method, device and system | |
| CN111277748A (en) | Construction site fire management system based on video intelligent analysis and implementation method thereof | |
| CN109584494A (en) | A kind of hidden fire-fighting danger inspection system based on Internet of Things | |
| CN110930632B (en) | Early warning system based on artificial intelligence | |
| CN105427519A (en) | Intelligent three-in-one alarm system for tunnel | |
| CN112263803A (en) | Unmanned vehicle intelligent security system based on real-time scene inspection and automatic detection fire extinguishing and control method | |
| CN110853287A (en) | Flame real-time monitoring system and method based on Internet of things distributed architecture | |
| CN104484969A (en) | Method and device for monitoring power transmission line |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20130918 Termination date: 20170228 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |