CN101458865A - Fire disaster probe system and method - Google Patents
Fire disaster probe system and method Download PDFInfo
- Publication number
- CN101458865A CN101458865A CNA2008101062620A CN200810106262A CN101458865A CN 101458865 A CN101458865 A CN 101458865A CN A2008101062620 A CNA2008101062620 A CN A2008101062620A CN 200810106262 A CN200810106262 A CN 200810106262A CN 101458865 A CN101458865 A CN 101458865A
- Authority
- CN
- China
- Prior art keywords
- image
- fire
- flame
- module
- camera
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 31
- 239000000523 sample Substances 0.000 title description 4
- 238000001514 detection method Methods 0.000 claims abstract description 33
- 239000000779 smoke Substances 0.000 claims abstract description 32
- 238000001914 filtration Methods 0.000 claims abstract description 9
- 230000004927 fusion Effects 0.000 claims description 33
- 238000012545 processing Methods 0.000 claims description 30
- 238000005286 illumination Methods 0.000 claims description 27
- 238000012544 monitoring process Methods 0.000 claims description 26
- 238000004364 calculation method Methods 0.000 claims description 22
- 230000008859 change Effects 0.000 claims description 18
- 230000033001 locomotion Effects 0.000 claims description 11
- 230000004044 response Effects 0.000 claims description 11
- 230000003595 spectral effect Effects 0.000 claims description 11
- 230000000052 comparative effect Effects 0.000 claims description 10
- 238000004458 analytical method Methods 0.000 claims description 7
- 238000010835 comparative analysis Methods 0.000 claims description 6
- 230000007246 mechanism Effects 0.000 claims description 6
- 238000006243 chemical reaction Methods 0.000 claims description 5
- 238000009792 diffusion process Methods 0.000 claims description 4
- 230000003287 optical effect Effects 0.000 claims description 4
- 238000004891 communication Methods 0.000 description 6
- 230000000007 visual effect Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 238000001228 spectrum Methods 0.000 description 5
- LFQSCWFLJHTTHZ-UHFFFAOYSA-N Ethanol Chemical compound CCO LFQSCWFLJHTTHZ-UHFFFAOYSA-N 0.000 description 4
- 238000002485 combustion reaction Methods 0.000 description 4
- 238000013459 approach Methods 0.000 description 3
- 230000007257 malfunction Effects 0.000 description 3
- 239000000126 substance Substances 0.000 description 3
- 238000001931 thermography Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 230000000295 complement effect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 229910044991 metal oxide Inorganic materials 0.000 description 2
- 150000004706 metal oxides Chemical class 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000035945 sensitivity Effects 0.000 description 2
- 235000009781 Myrtillocactus geometrizans Nutrition 0.000 description 1
- 240000009125 Myrtillocactus geometrizans Species 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 239000000443 aerosol Substances 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 235000019504 cigarettes Nutrition 0.000 description 1
- 238000007621 cluster analysis Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000012467 final product Substances 0.000 description 1
- 238000007499 fusion processing Methods 0.000 description 1
- 229910010272 inorganic material Inorganic materials 0.000 description 1
- 239000011147 inorganic material Substances 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000012538 light obscuration Methods 0.000 description 1
- 239000011368 organic material Substances 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000010223 real-time analysis Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Chemical compound O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Landscapes
- Fire-Detection Mechanisms (AREA)
- Closed-Circuit Television Systems (AREA)
- Fire Alarms (AREA)
Abstract
The invention discloses a fire disaster detection system and a method thereof. The fire disaster detection system at least comprises two changeable cameras as a first camera of colorful or near infrared mode and a second camera of colorful or white-black mode, for obtaining the video images of detected region; an image collection module connected with two changeable cameras for converting and filtering the video images collected by the two changeable cameras into digital images, to obtain time sequence images; a fire disaster recognition module connected with the image collection module for analyzing the time sequence images, to obtain flame, smoke and fire disaster occurrence probability; and an alarm module connected with the fire disaster recognition module for comparing the flame, smoke and fire disaster occurrence probability with a preset threshold value, and emitting alarm message according to the comparison result. The invention can effectively obtain the fire disaster video images under various backlight conditions, to realize early and reliable fire disaster alarm.
Description
Technical field
The present invention relates to the security against fire monitoring technology, relate in particular to a kind of fire detecting system and method.
Background technology
Along with the continuous development of whole world industry, commerce, covil construction facility, fire is also increasing by on a year-on-year basis.The climatic variation that globalized in the last few years, the easier generation of fire, another wildfire takes place frequently together, has brought tremendous loss and puzzlement to the mankind, especially to the forest in the whole world, has brought the huge disaster that is difficult to save especially.
Though present fire detecting system can be used for the Detection And Warning to fire, but there is the problem of operating lag in the conventional fire detector, for example, heat that detector can detect and smog arrive position of detector from fire location needs bigger delay, especially to the detection in large-scale places such as forest difficult problem especially.For large-scale places such as forests, even adopt satellite monitoring or air sampling smoke detector etc. to survey, the problem of operating lag also is difficult to avoid.
Fire detecting system based on image then can be surveyed to eliminate operating lag simultaneously from flame and smog two aspects, realizes quick early warning.Fire detecting system based on image does not rely on some physical parameters, for example temperature or rate of temperature change, light extinction rate, UV (Ultraviolet Ray, ultraviolet) or IR (infraredRay, infrared ray) etc., but the visible features by one or more fire in the recognition image, for example color, flicker, texture, dim light etc., and each characteristic parameter combined, by a decision-making mechanism, determine whether breaking out of fire.In general, based on the fire detecting system of image large space fire early detection, fire physical features not according to the detection of the place of common rule development or zone (for example tunnel, forest), visual, can have suitable advantage with aspect such as CCTV (Closed-Circuit Television, closed-circuit TV monitoring system) supervisory system compatibility.
But, owing to the visible features that must depend on fire based on the fire detecting system of image, for example size, motion, transparency, continuation etc., need be based upon under the visible environmental baseline, to having relatively high expectations of background environment, therefore there is certain limitation, that is:
(1) possibly can't detect background color, the intensity scene fire similar, produce and fail to report the police to fire.
For example, can not survey the flame of absolute alcohol perspective, scrappy sheet flame, the blue flame under the blue background or the flame on the moving vehicle etc. after being blown by air-flow.
(2) may when being very similar to fire, abiogenous situation produce false alarm.
For example sunshine or moonlight by ripples reflections, wear the leaf that waves in people that orange gym suit moving, the wind, water vapor, controlled fire, the cloud etc. that seems to be smog all may cause system's false alarm.
In addition, for the camera head in the detection system, normally adopt the camera of fixed spectrum characteristic.And when surveying flame, wish that generally aperture is smaller, and compensate smallerly, can get access to flame contours comparatively clearly like this; But the detection to smog is then different, because the general gray of cigarette, black or other darker tone wish that generally aperture is big, the light compensation is more, especially when taking smog night.Therefore adopt a kind of camera of fixed spectrum characteristic will be difficult to satisfy the requirement that flame and smog are surveyed simultaneously.But, if system increases to camera parameter control function the complexity that increase system greatly realizes.
Also can adopt thermal imaging system to carry out detection, thermal imaging system can be surveyed flame, hot spot preferably, but the flame that can't survey smog and be blocked, thermal imaging system often adopts long wave or MID INFRARED in addition, and the sensor cost is higher.
Summary of the invention
In view of this, the embodiment of the invention provides a kind of fire detecting system and method, can effectively obtain the fire video image under the diversity of settings light condition, thereby realizes early stage, reliable fire alarm.
For reaching above-mentioned purpose, on the one hand, the embodiment of the invention provides a kind of fire detecting system, comprising:
At least comprise first camera that is operated in colour or near infrared pattern or be operated in colour or the two variable video camera of second camera of white-black pattern, be used to obtain the video image of search coverage;
Image capture module is connected with described two variable video cameras, be used for from described pair of variable camera acquisitions to video image be converted to digital picture, and carry out Filtering Processing, obtain time-series image;
The fire identification module is connected with described image capture module, is used for described time-series image is carried out analyzing and processing, obtains flame, smog and fire probability of happening;
Alarm module is connected with described fire identification module, is used for more described flame, smog and fire probability of happening and predetermined threshold value, and sends corresponding warning message according to comparative result.
Described two variable video camera comprises described first camera and described second camera that is operated in colour or white-black pattern that is operated in colour or near infrared pattern;
Described time-series image comprises:
From described first camera collection to first video image handle the very first time sequence image obtain;
From described second camera collection to second video image handle second time-series image obtain.
Described first camera adopts the spectral response scope at the CCD of 400nm to 1200nm or cmos image sensor and the cutoff frequency high pass infrared fileter at the above wave band of 850nm;
Described second camera adopts CCD or the cmos image sensor of spectral response scope at 400nm to 1200nm.
Also comprise:
The switching controls module, be connected with described two variable video cameras, described image capture module and described fire identification module, be used for to carry out background illumination intensity and light Distribution calculation from the time-series image that described image capture module gets access to, control described second camera according to the processing requirements of result of calculation or described fire identification module and carry out the switching of mode of operation, also be used for controlling described first camera and carry out the switching of mode of operation according to the processing requirements of described fire identification module.
Described fire identification module further comprises:
Background modeling and update module are connected with described image capture module, are used for the described very first time sequence image and described second time-series image that get access to are analyzed and self study, obtain corresponding long period background image and short period background image;
The flame identification module, be connected with update module with described background modeling, be used for according to described long period background image and short period background image, calculate the flame characteristic parameter of described very first time sequence image and described second time-series image respectively, and described flame characteristic parameter carried out data fusion, obtain the flame probability of happening;
The smog identification module, be connected with update module with described background modeling, be used for according to described long period background image and short period background image, calculate the smoke characteristics parameter of described very first time sequence image and described second time-series image respectively, and described smoke characteristics parameter carried out data fusion, obtain the smog probability of happening;
The fire probability Fusion Module is connected with described smog identification module with described flame identification module, is used for described flame and smog probability of happening are carried out data fusion, determines the fire probability of happening.
Also comprise:
The background light source module is used for when the Luminance Distribution of search coverage and intensity of illumination grade are lower than detection criterion, for described two variable video cameras provide background light source.
Also comprise:
Light source control module, be connected with described background light source module with described image capture module, be used for the time-series image that basis gets access to from described image capture module, the Luminance Distribution of analytical calculation search coverage and intensity of illumination grade, and when the Luminance Distribution of search coverage and intensity of illumination grade are lower than detection criterion, control the startup of described background light source module.
Also comprise:
Set debugging and self-learning module, is connected, be used for the parameter setting of system, and at the generation false alarm or fail to report when alert, utilize the parameter of self-study mechanism update system with described fire identification module.
Also comprise:
Information record display module, be connected with described fire identification module, be used to show, preserve video image and various warning message in the search coverage that collects, and be connected with self-learning module with described setting debugging, false alarm taking place or is failing to report when alert provides the accident video record to described setting debugging and self-learning module.
Also comprise:
The Cloud Terrace and the cradle head control module of controlling described The Cloud Terrace rotation;
Described cradle head control module is connected with described The Cloud Terrace, described fire recognition device and described two variable video cameras, the The Cloud Terrace fixed cycle or the variable period rotation condition that are used for being provided with according to described fire identification module are calculated the presetting bit of The Cloud Terrace or the rotational angle of each position, or calculate the The Cloud Terrace rotational angle according to the field angle of described two variable video cameras, to control the rotation of described The Cloud Terrace.
Also comprise:
The monitored area is provided with module, be connected with described two variable shootings with described fire identification module, be used for recognition result according to described fire identification module, the search coverage of described two variable video cameras is set to a plurality of monitor areas of different nature, described monitor area comprises: highly sensitive smog and flame monitoring district, insensitive smog and flame monitoring district, fault monitoring district and non-surveillance zone, or fire alarm subregion.
On the other hand, the embodiment of the invention also provides a kind of fire detecting method, comprising:
Will from comprise first camera that is operated in colour or near infrared pattern at least be operated in colour or the two variable camera acquisition of second camera of white-black pattern to video image be converted to digital picture, and carry out Filtering Processing, obtain time-series image;
Described time-series image is carried out analyzing and processing, obtain flame, smog and fire probability of happening;
More described flame, smog and fire probability of happening and predetermined threshold value, and send corresponding warning message according to comparative result.
Described two variable video camera comprises described first camera and described second camera that is operated in colour or white-black pattern that is operated in colour or near infrared pattern;
Described time-series image comprises:
From described first camera collection to first video image handle the very first time sequence image obtain;
From described second camera collection to second video image handle second time-series image obtain.
Described first camera adopts the spectral response scope at the CCD of 400nm to 1200nm or cmos image sensor and the cutoff frequency high pass infrared fileter at the above wave band of 850nm;
Described second camera adopts CCD or the cmos image sensor of spectral response scope at 400nm to 1200nm.
Also comprise:
Described time-series image is carried out background illumination intensity and light Distribution calculation, control the step that described second camera carries out the switching of mode of operation according to the processing requirements of result of calculation or fire identification; And
Processing requirements according to fire identification is controlled the step that described first camera carries out the switching of mode of operation.
Described very first time sequence image and described second time-series image are carried out analyzing and processing, and the method that obtains flame, smog and fire probability of happening is specially:
Described very first time sequence image and described second time-series image are analyzed and self study, obtained corresponding long period background and short period background image;
According to described long period background and short period background image, calculate the flame characteristic parameter of described very first time sequence image and described second time-series image respectively, and described flame characteristic parameter is carried out data fusion, obtain the flame probability of happening;
According to described long period background and short period background image, calculate the smoke characteristics parameter of described very first time sequence image and described second time-series image respectively, and described smoke characteristics parameter is carried out data fusion, obtain the smog probability of happening;
Described flame and smog probability of happening are carried out data fusion, determine the fire probability of happening.
The described very first time sequence image that is used to calculate described flame probability of happening comprises coloured image and near-infrared image, and described second time-series image is coloured image or black white image;
The described very first time sequence image that is used to calculate described smog probability of happening is a coloured image, and described second time-series image is coloured image or black white image.
The method of determining the flame probability of happening is specially:
Coloured image in the described very first time sequence image is analyzed, obtained the first flame characteristic parameter I
Chara1=f
3{ F
Qi, A
i, PR
i... };
If the described second camera work at present is at color mode, then described second time-series image is a coloured image, and described second time-series image is analyzed, and obtains the second flame characteristic parameter I
Chara2=f
3{ F
Qi, A
i, PR
i... };
Near-infrared image in the described very first time sequence image is analyzed, obtained the 3rd flame characteristic parameter I
Chara3=f
3{ F
Qi, A
i, PR
i... };
Wherein, F
QiFor flashing frequency, A
iBe flame area rate of change, PR
iBe the long period and the short period rate of spread;
Described very first time sequence image or described second time-series image are analyzed, obtained reacting the parameter I of flame flicking feature
Freq
The described flame characteristic parameter that aforementioned calculation obtains is carried out data fusion, obtains described flame probability of happening:
P
Flame(t)＝F
2{I
Chara1，I
Chara2，I
Chara3，I
Freq}。
The method of determining described smog probability of happening is specially:
Respectively with current frame image and former frame or preceding some two field picture comparative analyses of described very first time sequence image and described second time-series image, and, obtain reacting the smoke characteristics parameter I of the fast slow characteristic of smog movement with described current frame image and described long period background image or short period background image comparative analysis
Speed, I
SpeedComputing formula be:
I
Speed=(current frame image and former frame or preceding some two field pictures change area)/(current frame image and short period or long period background image change area);
With described very first time sequence image and the described second time-series image five equilibrium or non-ly be divided into a plurality of monitored areas, each monitored area is analyzed respectively, obtained reacting the characteristic parameter I of smog disperse characteristic
Disp=f
1{ R
i, G
i, S
i, T
i, F
i... };
Calculate the characteristic parameter I of the reaction smog movement diffusion property of described very first time sequence image and described second time-series image respectively
Move=f
2{ R
i, G
i, S
i, T
i, F
i, I
Speed...;
Wherein, R
iBe related coefficient, G
iBe graded, S
iFor saturation degree changes, T
iBe texture variations, F
iBe optical flow field;
Get access to described smoke characteristics parameter and carry out data fusion above-mentioned, obtain described smog probability of happening, that is:
P
Smoke(t)＝F
1{I
Disp1，I
Move1，I
Disp2，I
Move2}
Wherein, I
Disp1, I
Move1The smoke characteristics parameter of representing described very first time sequence image, I
Disp2, I
Move2The smoke characteristics parameter of representing described second time-series image.
The method of determining described fire probability of happening is specially:
P
f1(t)＝P
Smoke(t)[1+(P
Flame(t)-K)]
P
f2(t)＝P
Flame(t)[1+(P
Smoke(t)-K)]
P
Fire(t)＝max{P
f1(t)，P
f2(t)}
Wherein, P
Flame(t) and P
Smoke(t) be respectively t flame probability of happening and smog probability of happening constantly, P
Fire(t) be t fire probability of happening constantly, K is a dead band value.
Also comprise:
The Luminance Distribution of analytical calculation search coverage and intensity of illumination grade, and when the Luminance Distribution of search coverage and intensity of illumination grade are lower than detection criterion, use background light source.
Also comprise:
Detection system is carried out parameter setting, and, utilize self-study mechanism update system parameter false alarm taking place or failing to report when alert.
Also comprise:
Calculate the presetting bit of The Cloud Terrace or the rotational angle of each position according to The Cloud Terrace fixed cycle that is provided with or variable period rotation condition, or calculate the The Cloud Terrace rotational angle, control the rotation of described The Cloud Terrace according to the field angle of described two variable video cameras.
Also comprise:
According to the fire recognition result, the search coverage of described two variable video cameras is set to a plurality of monitor areas of different nature, described monitor area comprises: highly sensitive smog and flame monitoring district, insensitive smog and flame monitoring district, fault monitoring district and non-surveillance zone, or fire alarm subregion.
The embodiment of the invention has following beneficial effect:
System analyzes the image of the multiple spectral characteristic that gets access to simultaneously, can detect the fire under the diversity of settings light condition, and can effectively avoid false alarm and fail to report the police;
System carries out the identification of flame and smog simultaneously to the image that gets access to, and makes that the result of fire identification is more accurate;
System has self-learning function, can constantly adapt to residing environment, improves the anti-mistaking warning greatly and fails to report alert ability;
System can realize the detection of variable field of view.
Description of drawings
Fig. 1 is the fire detecting system structural representation of the embodiment of the invention one;
Fig. 2 is the fire detecting method schematic flow sheet of the embodiment of the invention one;
Fig. 3 is the fire identification process synoptic diagram of the embodiment of the invention one;
Fig. 4 is the distributed intelligence fire detecting system frame diagram based on DSP of the embodiment of the invention one;
Fig. 5 is the fire detecting system frame diagram based on outer computer of the embodiment of the invention one;
The black white image and the near-infrared image of the flame that Fig. 6 arrives for the two variable camera acquisition of the embodiment of the invention one;
Fig. 7 is the fire detecting system structural representation of the embodiment of the invention two;
Fig. 8 is the fire detecting system structural representation of the embodiment of the invention three.
Embodiment
Be illustrated in figure 1 as the fire detecting system structural representation of the embodiment of the invention one, this fire detecting system comprises:
Two variable video cameras 10 are used to obtain the video image of search coverage.
Two variable video cameras 10 comprise first camera 101 that is operated in colour or near infrared pattern and second camera 102 that is operated in colour or white-black pattern.
Concrete, when search coverage is in illumination condition following time, arranged, switching controls module 14 controls second camera 102 switches to color mode; When search coverage is in unglazed photograph or illumination more weak following time of condition, control second camera 102 and switch to white-black pattern.
Generally, system with the Working mode set of first camera 101 at color mode, when above-mentioned fire identification module 12 when carrying out flame identification, also need first camera 101 is switched to the near infrared pattern to obtain near-infrared image, carry out the secondary checking of flame characteristic; In addition, when when blocking fire or flame flicking and survey, also need to control second camera 102 and switch to white-black pattern.
Therefore, switching controls module 14 also is connected with fire identification module 12, is used for controlling first camera 101 and second camera 102 carries out the switching of mode of operation according to the processing requirements of fire identification module 12.
Certainly, each above-mentioned module all needs to power, so system entails also comprises supply module, is used for each module for power supply, guarantees the normal operation of system.
Wherein, the standard of above-mentioned two variable video camera 10 can be selected PAL (PhaseAlternating Line, line-by-line inversion) or NTSC (National Television System Committee, National Television System Committee (NTSC)) as required.
Preferably, first camera 101 adopts little aperture and low compensation, and to guarantee adapting to high bias light condition, second camera 102 adopts large aperture and high compensation, to catch smog movement and the smog disperse feature under the various illumination conditions to greatest extent.
In the embodiment of the invention, first camera 101 adopts the spectral response scope at the CCD of 400nm (nanometer)～1200nm (nanometer) (Charge Coupled Device, Charge Coupled Device (CCD)) or CMOS (Complementary Metal Oxide Semiconducto, CMOS (Complementary Metal Oxide Semiconductor)) imageing sensor and cutoff frequency at the high pass infrared fileter of 850nm (or 950nm and more than) wave band.
Second camera, 102 same CCD or the cmos image sensors that adopt the spectral response scope at 400nm～1200nm.
The near infrared pattern of the white-black pattern of second camera 102 and first camera 101 can also effectively be caught the flame flicking feature, promptly when flame is blocked, reflection or refraction by peripheral background, visible light that sends when obtaining flame combustion and infrared light, and then detect the fire that flame is blocked.
Be illustrated in figure 6 as the black white image and the near-infrared image of the flame that the two variable camera acquisition of the embodiment of the invention arrives.
Above-mentioned fire identification module 12 is when carrying out the identification of flame and smog, can at first carry out modeling to background by self study, obtain the background image of specific period, with the time-series image and the contrast of specific period background image that get access to, can quick identification go out flame or smoke characteristics then.
Therefore, fire identification module 12 further comprises:
Background modeling and update module 121, be connected with image capture module 11, be used for the very first time sequence image and second time-series image that get access to from image capture module 11 are analyzed and self study, obtain corresponding long period background image and short period background image.
A reference background image is different with only adopting in the traditional approach, equal self study obtains a long period background and a short period background at each spectrum picture that gets access to (colour, near infrared and black and white) for background modeling and update module 121, the time span of long period background be some minutes to some hrs, the time span of short period background is some seconds to some minutes.
Fire probability Fusion Module 124 is connected with smog identification module 123 with flame identification module 122, is used for described flame probability of happening and smog probability of happening are carried out data fusion, obtains the fire probability of happening of search coverage.
Above-mentioned background modeling and update module 124 are by analyzing and self study the time-series image that gets access to, and the cycle background image that obtains specifically comprises:
The image that gets access to from first camera 101 can obtain with self study by analysis:
The first colored long period background image and the first colored short period background image; And
Near infrared long period background image and near infrared short period background image.
The image that gets access to from second camera 102 can obtain with self study by analysis:
The second colored long period background image and the second colored short period background image; And
Black and white long period background image and black and white short period background image.
In addition, the above-mentioned time-series image that is used for flame and smog identification is not that each two field picture is all valuable to flame and smog identification.
Therefore, preferably, system can at first analyze and self study the very first time sequence image and second time-series image that get access to, from the very first time sequence image and second time-series image, select the particular series two field picture that is applicable to that flame and smoke characteristics are analyzed respectively, and then the particular series two field picture that this is selected carried out the identification of flame and smog, will improve the efficient of flame and smog identification greatly.
According to above-mentioned long period background image that obtains and short period background image, the concrete steps that flame identification module 122 is carried out flame identification are as follows:
At first, need to prove that the very first time sequence image that is used to carry out flame analysis comprises coloured image and near-infrared image, second time-series image is coloured image or black white image.
Generally, system at color mode, according to the requirement of flame identification, also needs the Working mode set of first camera 101 first camera 101 is switched to the near infrared pattern to obtain near-infrared image, carries out the secondary checking of flame characteristic.
(1) at first the coloured image in the very first time sequence image is analyzed, obtained the first flame characteristic parameter I
Chara1=f
3{ F
Qi, A
i, PR
i... }, F wherein
QiFor flashing frequency, A
iBe flame area rate of change, PR
iBe the long period and the short period rate of spread;
(2) if 102 works at present of second camera at color mode, at this moment, second time-series image is a coloured image, and second time-series image is analyzed, and obtains the second flame characteristic parameter I
Chara2=f
3{ F
Qi, A
i, PR
i... }, F wherein
QiFor flashing frequency, A
iBe flame area rate of change, PR
iBe the long period and the short period rate of spread;
(3) the above-mentioned flame characteristic that analyzes is carried out the secondary checking:
Near-infrared image in the very first time sequence image is analyzed, obtained the 3rd flame characteristic parameter I
Chara3=f
3{ F
Qi, A
i, PR
i... }, F wherein
QiFor flashing frequency, A
iBe flame area rate of change, PR
iFor the long period and the short period rate of spread, wherein flash frequency F
QiBe fire predominant frequency 2～12Hz;
(4) very first time sequence image or second time-series image are analyzed, obtained reacting the parameter I of flame flicking feature
Freq, wherein calculating flicker frequency is 2～6Hz;
(5) the flame characteristic parameter that aforementioned calculation is obtained is carried out data fusion, obtains the flame probability of happening of search coverage, that is:
P
Flame(t)＝F
2{I
Chara1，I
Chara2，I
Chara3，I
Freq}
Above-mentioned flame probability of happening can be learnt to calculate and obtain by neural network, fuzzy algorithm etc.
According to above-mentioned long period background image that obtains and short period background image, smog identification module 123 carries out the specific as follows of smog identifying:
Generally, the very first time sequence image that is used to carry out smog identification is a coloured image, and second time-series image is coloured image or black white image.
(1) respectively with the current frame image in the very first time sequence image and second time-series image and former frame or preceding some two field picture comparative analysis, and, determine the smoke characteristics parameter I of the fast slow characteristic of reaction smog movement with current frame image and above-mentioned long period background image that obtains or short period background image comparative analysis
Speed, I
SpeedComputing formula be:
I
Speed=(current frame image and former frame or preceding some two field pictures change area)/(current frame image and short period or long period background image change area);
(2) respectively with very first time sequence image and the second time-series image five equilibrium or non-ly be divided into a plurality of monitored areas, calculate coefficient R at each monitored area
i, graded G
i, saturation degree changes S
i, texture variations T
iWith optical flow field F
iDeng characteristic parameter, obtain reacting the characteristic parameter I of smog disperse characteristic
Disp=f
1{ R
i, G
i, S
i, T
i, F
i... };
(3) calculate the characteristic parameter I that reacts the smog movement diffusion property
Move=f
2{ R
i, G
i, S
i, T
i, F
i, I
Speed...;
(4) the above-mentioned smoke characteristics parameter that obtains is carried out data fusion, obtain the smog probability of happening of search coverage, that is:
P
Smoke(t)＝F
1{I
Disp1，I
Move1，I
Disp2，I
Move2}
Wherein, I
Disp1, I
Move1The smoke characteristics parameter of expression very first time sequence image, I
Disp2, I
Move2The smoke characteristics parameter of representing second time-series image.
Above-mentioned smog probability of happening can be learnt to calculate and obtain by neural network or fuzzy algorithm etc.
Above-mentioned fire probability Fusion Module 124 is determined the fire probability of happening according to the flame probability of happening and the smog probability of happening that get access to from flame identification module 122 and smog identification module 123, and fire probability of happening fusion method is:
P
f1(t)＝P
Smoke(t)[1+(P
Flame(t)-K)]
P
f2(t)＝P
Flame(t)[1+(P
Smoke(t)-K)]
P
Fire(t)＝max{P
f1(t)，P
f2(t)}
Wherein, P
Flame(t) and P
Smoke(t) be respectively t flame probability of happening and smog probability of happening constantly, P
Fire(t) be t fire probability of happening constantly, K is a dead band value, general desirable 25% or other numerical value, and dead band value K is big more, and single fire is characterized, and is big more as fire probability of happening inhibiting effect, and the ability that reduces wrong report is strong more, but sensitivity also will reduce.
The purpose of flame probability of happening and the fusion of smog probability of happening is that at the image that has flame and smoke characteristics simultaneously, the easier fire of confirming as is with reaction capacity and the reliability of raising system to this class fire.When flame or smog probability of happening during less than dead band value K, the fire probability of happening is suppressed, and when flame or smog probability of happening during all greater than K, the fire probability of happening promptly can accelerated growth.
Flame, smog and fire probability of happening that above-mentioned alarm module 13 sends according to fire identification module 12, according to following rule warn, early warning and warning:
If P
Smoke(t) 〉=Atten1 or P
Flame(t) 〉=Atten2 or P
Fire(t) 〉=Atten3 then warns;
If P
Smoke(t) 〉=Warn1 or P
Flame(t) 〉=Warn2 or P
Fire(t) 〉=Warn3 then carries out early warning;
If P
Smoke(t) 〉=Alarm1 or P
Flame(t) 〉=Alarm2 or P
Fire(t) 〉=Alarm3 then reports to the police;
Wherein, Atten1, Atten2, Atten3, Warn1, Warn2, Warn3, Alarm1, Alarm2, Alarm3 are judgment threshold.
Simultaneously, consider that Luminance Distribution and light intensity in the search coverage may change along with the variation of available light and artificial light source, and then can not satisfy the requirement (as this special time in evening) of detection, therefore, embodiment of the invention fire detecting system also comprises:
Light source control module, be connected with image capture module 11, be used for the time-series image that basis gets access to from image capture module 11, judge the Luminance Distribution and the background illumination intensity of search coverage, for the situation that is lower than detection criterion, to the order of background light source module output start-up control, and the visual field situation of change after the startup of monitoring light source.
The background light source module is connected with light source control module, is used to receive the start-up control order that light source control module is sent, for search coverage provides background light source.
This background light source can be the infrared background light source, also can be the background light source of other types.
When background light source started, even on-the-spot natural conditions cause light darker, the background light source that two variable video cameras 10 also can utilize the background light source module to provide collected qualified video image.
Simultaneously; because above-mentioned part of module need be assembled the back and is installed in the search coverage scene as detector; therefore consider being installed in the protection of each on-the-spot module; also a shell can be set; be used to hold modules such as two variable video cameras 10; described shell is provided with the window eyeglass; be positioned at the place ahead of two variable video cameras 10; the main pair variable video cameras 10 that cooperate obtain distinct image; this window eyeglass can be made by the cutting of organic or inorganic material, guarantees that the visible light and the infrared ray of search coverage can be received by two variable video camera 10.
The fire detecting system of the embodiment of the invention can adopt the distributed structure/architecture based on DSP (Digital SignalProcessing, digital signal processor), also can adopt the framework based on outer computer.When system adopted distributed structure/architecture based on DSP, being installed in on-the-spot detector can comprise: two variable video cameras 10, image capture module 11, fire identification module 12, input/output module and background light source module; When system adopts framework based on outer computer, be installed in on-the-spot detector and also can only comprise: two variable video cameras 10, input/output module and background light source module.
No matter be based on the fire detecting system of DSP or outer computer, being installed in on-the-spot detector all needs to comprise an input/output module, and the hardware and software that is connected with external module interface is provided, and corresponding information is carried out the input and output operation.
Input/output module comprises input/output signal modulate circuit, short range and long-range (RS232/485, TCP/IP etc.) telecommunication circuit interface etc.
Be illustrated in figure 2 as the schematic flow sheet of fire detecting method of the fire detecting system of the embodiment of the invention one, may further comprise the steps:
Simultaneously, said method also comprises:
Carry out background illumination intensity and light Distribution calculation to getting access to time-series image, control the step that second camera carries out the switching of mode of operation according to the processing requirements of result of calculation or fire identification;
Processing requirements according to fire identification is controlled the step that first camera carries out the switching of mode of operation; And
When the Luminance Distribution of search coverage and intensity of illumination grade are lower than detection criterion, use the step of background light source.
Below fire identifying in the step 23 is described in detail, as shown in Figure 3, step 23 specifically comprises:
Simultaneously, the fire detecting system of the embodiment of the invention further comprises:
The fire locating module, be connected with fire identification module 12 with two variable video cameras 10, when fire identification module 12 confirms that fire takes place, the flame, the smoke characteristics parameter that occur in the image are carried out cluster analysis, different parts or the regional fire that occurs are simultaneously indicated respectively, and the accurate coordinate position of notifying two variable video cameras 10 to obtain corresponding fire.
The simulated operation module is connected with fire identification module 12, is used for simulated failure or simulation fire alarm, installs, debugs and safeguard so that carry out system.
The system that considers is easy to use and false alarm may take place in some special place and fail to report alert situation, and the fire detecting system of the embodiment of the invention also comprises:
Set debugging and self-learning module, is connected, be used to carry out the parameter setting of detection system, and, utilize the parameter of self-study mechanism renewal detection system false alarm taking place or failing to report when alert with fire identification module 12.
Information record display module, be connected with fire identification module 12, be used to show, preserve video image and various warning message in the search coverage that collects, and with set debugging and be connected with self-learning module, false alarm is taking place or fail to report when alert to setting and debug and self-learning module provides the accident video record.
The setting debugging specifically comprises with the process that self-learning module is carried out self study:
(1) at first to false alarm or fail to report alert video image and carry out self study, calculates the weight parameter of flame, smog and fire data fusion process;
(2) at the weight parameter of new acquisition, use the sample in the standard fire video image storehouse, whether the ability of checking system response reaches the expection regulation, if depart from expection, then utilize the sample in the standard fire video image storehouse to carry out self study, the weight of adjustment data fusion;
(3) utilize the data fusion weight relearn to carry out the trial run of specific period, promptly can be used as the reference parameter of this particular place after errorless, provide from now on and use.
The fire detecting system of the embodiment of the invention also comprises:
Communication module is connected with fire identification module 12, is used for being connected with regional or long-range alarm monitoring system, and long-range control information is uploaded or accepted to the image, fire probability of happening, fire location and the warning message that get access to.
Communication module can be with digital picture according to H.264 waiting international standard to compress transmission.
The Zone Alerts monitoring module is connected with communication module, is used for each area detection device that distributes is carried out telemonitoring and control.
The Zone Alerts monitoring module also can be used as the central alarm supervisory system, mainly finishes following function:
(1) monitors at the multiplexed detection device, and in the monitoring video image superposition fire location, fire probability of happening information, adopt Geographic Information System GIS to carry out the supervision in big zone, information such as probe address coding, detector installation region, fire probability, the doubtful regional location of image, alert levels are shown;
(2) setting up database manages, stores various information and fire alarm zone fire image;
(3) detector of search coverage, The Cloud Terrace etc. are controlled;
(4) parameter detectors such as sensitivity, monitor area and type are carried out long-range setting;
Because the complicacy of environment in actual the use, search coverage that can two variable video cameras 10 is set to a plurality of monitored areas of different nature, therefore, system comprises that also the monitored area is provided with module, the monitored area is provided with the module search coverage and is set to highly sensitive smog and flame monitoring district, insensitive smog and flame monitoring district, malfunction monitoring district and non-monitored zone, or the fire alarm subregion etc.Three above malfunction monitoring zones can guarantee to judge fast malfunctions such as visual field offset in the image; Zone for controlled fire is clearly arranged can be changed to non-monitored zone; For the big zone of obvious interference, can reduce some system sensitivities, to improve reliability; For disturbing very important seldom again zone can be made as highly sensitive monitoring section.
(5) can increase fire secondary identification module as required, the unusual video image of fire probability of happening is carried out secondary identification and confirms the reliability of elevator system.
In addition, the camera head in traditional fire detecting system is fixing visual field substantially, because variable field of view can bring the delay of the complicated and detection alarm that system realizes, yet in the detection in large-scale places such as forest, variable field of view becomes a kind of essential.
Therefore, consider the demand of variable field of view, the fire detecting system of the embodiment of the invention also comprises the cradle head control module that the The Cloud Terrace that can vertical-horizontal rotates and control The Cloud Terrace rotate.
The cradle head control module, be connected with two variable video cameras 10 with fire identification module 12, the The Cloud Terrace fixed cycle or the variable period that are provided with according to fire identification module 10 rotate condition, or the presetting bit of the field angle of two variable video cameras 10 calculating The Cloud Terrace or the rotational angle of each position, rotate with the control The Cloud Terrace.
The change-over period of each position of The Cloud Terrace is fixed cycle or variable cycle, the image that fixed cycle length will guarantee to gather the sufficient length frame is used for the judgement of fire, the situation of each angle or presetting bit real-time analysis is then depended in variable cycle control, when finding doubtful fire, or when confirming fire alarm, the control The Cloud Terrace keeps the visual field stable, till getting rid of doubtful or system reset; For in the visual field during without any doubtful fire or trend, the new rotation period value that obtains of output, the control The Cloud Terrace rotates, and rotation period is minimum to be not less than a certain threshold level.
The cradle head control module can be divided fixing a plurality of presetting bits according to the field angle of two variable video cameras 10, only need control the The Cloud Terrace rotation by presetting bit and get final product.But also can adopt the The Cloud Terrace of angle feedback, realize continuous, complicated control requirement.
Consider that above-mentioned fire detecting system also may be embedded in the UAV pilotless helicopter, at this moment, the fire detecting system of the embodiment of the invention also comprises:
The aircraft control module is used for hang and attitude adjustment are controlled.
For the UAV image fire detection system, the most difficult is that aircraft is bigger in motion and shake, and therefore the image that is obtained is necessary can rapid reaction fire essential characteristic.The aircraft control module then can be controlled, and then make the associated picture that can obtain more fire hang, flight attitude such as fall back.
Fig. 4 is the distributed fire detecting system frame diagram based on DSP of the embodiment of the invention.
Processor among Fig. 4 and controller are promptly finished the fire identification module in the fire detecting system of the embodiment of the invention and the function of switching controls module.
In the distributed fire detecting system based on DSP, adopt the DSP parallel processor to form complete and independent distributed intelligence fire detector, the distributed intelligence fire detector both can with conventional fire detection alarm system compatibility, also can with conventional CCTV frequency image monitoring system compatibility, have broad application prospects.
Fig. 5 is the fire detecting system frame diagram based on outer computer of the embodiment of the invention.
Same, processor among Fig. 5 and controller are promptly finished the fire identification module in the fire detecting system of the embodiment of the invention and the function of switching controls module.
Different with distributed fire detecting system based on DSP, can connect the two variable video cameras of many covers based on the fire detecting system of outer computer.
Two variable video camera in the above-mentioned fire detecting system also can include only first camera that is operated in colour or near infrared pattern, or includes only second camera that is operated in colour or white-black pattern.The two variable video camera that includes only first camera that is operated in colour or near infrared pattern is mainly used in the detection system at flame identification, and the two variable video camera that includes only second camera that is operated in colour or white-black pattern is mainly used in the detection system at smog identification.
Be illustrated in figure 7 as the fire detecting system structural representation of the embodiment of the invention two, this fire detecting system comprises:
The variable video camera 20 that comprises first camera 101 that is operated in colour or near infrared pattern is used to obtain the video image of search coverage.
Generally, system with the Working mode set of first camera 101 at color mode, when above-mentioned fire identification module 12 when carrying out flame identification, also need first camera 101 is switched to the near infrared pattern to obtain near-infrared image, carry out the secondary checking of flame characteristic.
Therefore, system also comprises:
Certainly, each above-mentioned module all needs to power, so system entails also comprises supply module, is used for each module for power supply, guarantees the normal operation of system.
Above-mentioned fire identification module 12 is when carrying out flame identification, can at first carry out modeling to background by self study, obtain the background image of specific period, with the time-series image and the contrast of specific period background image that get access to, can quick identification go out flame characteristic then.
Therefore, fire identification module 12 further comprises:
Background modeling and update module 121 are connected with image capture module 11, are used for the time-series image that gets access to from image capture module 11 is analyzed and self study, obtain corresponding long period background image and short period background image.
A reference background image is different with only adopting in the traditional approach, equal self study obtains a long period background and a short period background at each spectrum picture that gets access to (colour and near infrared) for background modeling and update module 121, the time span of long period background be some minutes to some hrs, the time span of short period background is some seconds to some minutes.
Above-mentioned background modeling and update module 124 are by analyzing and self study the time-series image that gets access to, and the cycle background image that obtains specifically comprises:
Colored long period background image and colored short period background image; And
Near infrared long period background image and near infrared short period background image.
In addition, the above-mentioned time-series image that is used for flame identification is not that each two field picture is all valuable to flame identification.
Therefore, preferably, system can at first analyze and self study the time-series image that gets access to, from described time-series image, select the particular series two field picture that is applicable to that flame characteristic is analyzed, and then the particular series two field picture that this is selected carried out flame identification, will improve the efficient of flame identification greatly.
According to above-mentioned long period background image that obtains and short period background image, the concrete steps that flame identification module 122 is carried out flame identification are as follows:
At first, need to prove that the time-series image that is used to carry out flame analysis comprises coloured image and near-infrared image.
Generally, system at color mode, according to the requirement of flame identification, also needs the Working mode set of first camera 101 first camera 101 is switched to the near infrared pattern to obtain near-infrared image, carries out the secondary checking of flame characteristic.
(1) at first the coloured image in the time-series image is analyzed, obtained the first flame characteristic parameter I
Chara1=f
3{ F
Qi, A
i, PR
i... }, F wherein
QiFor flashing frequency, A
iBe flame area rate of change, PR
iBe the long period and the short period rate of spread;
(2) the above-mentioned flame characteristic that analyzes is carried out the secondary checking:
Near-infrared image in the time-series image is analyzed, obtained the 3rd flame characteristic parameter I
Chara3=f
3{ F
Qi, A
i, PR
i... }, F wherein
QiFor flashing frequency, A
iBe flame area rate of change, PR
iFor the long period and the short period rate of spread, wherein flash frequency F
QiBe fire predominant frequency 2～12Hz;
(3) time-series image is analyzed, obtained reacting the parameter I of flame flicking feature
Freq, wherein calculating flicker frequency is 2～6Hz;
(4) the flame characteristic parameter that aforementioned calculation is obtained is carried out data fusion, obtains the flame probability of happening of search coverage, that is:
P
Flame(t)＝F
2{I
Chara1，I
Chara2，I
Chara3，I
Freq}
Above-mentioned flame probability of happening can be learnt to calculate and obtain by neural network, fuzzy algorithm etc.
The flame probability of happening that above-mentioned alarm module 13 sends according to fire identification module 12, according to following rule warn, early warning and warning:
If P
Flame(t) 〉=Atten2 then warns;
If P
Flame(t) 〉=Warn2 then carries out early warning;
If P
Flame(t) 〉=Alarm2 then reports to the police;
Wherein, Atten2, Warn2, Alarm2 are judgment threshold.
Certainly, the fire detecting system of the embodiment of the invention two also comprises background light source module, background light source control module, input/output module etc., with other module class in the foregoing description one seemingly, be not described in detail in this.
The fire detecting system of the two variable video cameras that include only first camera that is operated in colour or near infrared pattern of foregoing description only is used for flame identification, obtain the flame probability of happening, certainly, it also can be used for smog identification, obtain the smog probability of happening, and flame and smog probability of happening can be merged, obtain the fire probability of happening, just the effect of its smog identification is relatively poor.
Be illustrated in figure 8 as the fire detecting system structural representation of the embodiment of the invention three, this fire detecting system comprises:
The two variable video camera 30 that comprises second camera 102 that is operated in colour or white-black pattern is used to obtain the video image of search coverage.
Concrete, when search coverage is in illumination condition following time, arranged, switching controls module 14 controls second camera 102 switches to color mode; When search coverage is in unglazed photograph or illumination more weak following time of condition, control second camera 102 and switch to white-black pattern.
Certainly, each above-mentioned module all needs to power, so system entails also comprises supply module, is used for each module for power supply, guarantees the normal operation of system.
Above-mentioned fire identification module 12 is when carrying out smog identification, can at first carry out modeling to background by self study, obtain the background image of specific period, with the time-series image and the contrast of specific period background image that get access to, can quick identification go out smoke characteristics then.
Therefore, fire identification module 12 further comprises:
Background modeling and update module 121 are connected with image capture module 11, are used for the time-series image that gets access to from image capture module 11 is analyzed and self study, obtain corresponding long period background image and short period background image.
A reference background image is different with only adopting in the traditional approach, equal self study obtains a long period background and a short period background at each spectrum picture that gets access to (colour and black and white) for background modeling and update module 121, the time span of long period background be some minutes to some hrs, the time span of short period background is some seconds to some minutes.
Above-mentioned background modeling and update module 124 are by analyzing and self study the time-series image that gets access to, and the cycle background image that obtains specifically comprises:
Colored long period background image and colored short period background image; And
Black and white long period background image and black and white short period background image.
In addition, the above-mentioned time-series image that is used for smog identification is not that each two field picture is all valuable to smog identification.
Therefore, preferably, system can at first analyze and self study the time-series image that gets access to, from described time-series image, select the particular series two field picture that is applicable to that smoke characteristics is analyzed, and then the particular series two field picture that this is selected is carried out smog discern, will improve the efficient of smog identification greatly.
According to above-mentioned long period background image that obtains and short period background image, smog identification module 123 carries out the specific as follows of smog identifying:
(1) with the current frame image in the time-series image and former frame or preceding some two field picture comparative analysis, and, obtain reacting the smoke characteristics parameter I of the fast slow characteristic of smog movement with current frame image and above-mentioned long period background image that obtains or short period background image comparative analysis
Speed, I
SpeedComputing formula be:
I
Speed=(current frame image and former frame or preceding some two field pictures change area)/(current frame image and short period or long period background image change area);
(2), calculate coefficient R at each monitored area with the time-series image five equilibrium or non-ly be divided into a plurality of monitored areas
i, graded G
i, saturation degree changes S
i, texture variations T
iWith optical flow field F
iDeng characteristic parameter, obtain reacting the characteristic parameter I of smog disperse characteristic
Disp=f
1{ R
i, G
i, S
i, T
i, F
i... };
(3) calculate the characteristic parameter I that reacts the smog movement diffusion property
Move=f
2{ R
i, G
i, S
i, T
i, F
i, I
Speed...;
(4) the above-mentioned smoke characteristics parameter that obtains is carried out data fusion, obtain the smog probability of happening of search coverage, that is:
P
Smoke(t)＝F
1{I
Disp，I
Move}
Above-mentioned smog probability of happening can be learnt to calculate and obtain by neural network or fuzzy algorithm etc.
The smog probability of happening that above-mentioned alarm module 13 sends according to fire identification module 12, according to following rule warn, early warning and warning:
If P
Smoke(t) 〉=Atten1 then warns;
If P
Smoke(t) 〉=Warn1 then carries out early warning;
If P
Smoke(t) 〉=Alarm1 then reports to the police;
Wherein, Atten1, Warn1, Alarm1 are judgment threshold.
Certainly, the fire detecting system of the embodiment of the invention three also comprises background light source module, background light source control module, input/output module etc., with other module class in the foregoing description one seemingly, be not described in detail in this.
The fire detecting system of the two variable video cameras that include only second camera that is operated in colour or white-black pattern of foregoing description only is used for smog identification, obtain the smog probability of happening, certainly, it also can be used for flame identification, obtain the flame probability of happening, and flame and smog probability of happening can be merged, obtain the fire probability of happening, just the effect of its flame identification is relatively poor.
Certainly, the fire detecting system (include only the two variable video camera of first camera that is operated in colour or near infrared pattern or include only the two variable video camera of second camera that is operated in colour or white-black pattern) that utilizes the embodiment of the invention two and embodiment three is when carrying out detection, also can solve most of fire false alarm and fail to report alert problem, but compare with the fire detecting system (comprising the two variable video camera that is operated in colored or near infrared first camera and second camera that is operated in colour or white-black pattern) of the embodiment of the invention one, its accuracy to detection is low relatively.
The above only is a preferred implementation of the present invention; should be pointed out that for those skilled in the art, under the prerequisite that does not break away from the principle of the invention; can also make some improvements and modifications, these improvements and modifications also should be considered as protection scope of the present invention.
Claims (24)
1. a fire detecting system is characterized in that, comprising:
At least comprise first camera that is operated in colour or near infrared pattern or be operated in colour or the two variable video camera of second camera of white-black pattern, be used to obtain the video image of search coverage;
Image capture module is connected with described two variable video cameras, be used for from described pair of variable camera acquisitions to video image be converted to digital picture, and carry out Filtering Processing, obtain time-series image;
The fire identification module is connected with described image capture module, is used for described time-series image is carried out analyzing and processing, obtains flame, smog and fire probability of happening;
Alarm module is connected with described fire identification module, is used for more described flame, smog and fire probability of happening and predetermined threshold value, and sends corresponding warning message according to comparative result.
2. fire detecting system according to claim 1 is characterized in that:
Described two variable video camera comprises described first camera and described second camera that is operated in colour or white-black pattern that is operated in colour or near infrared pattern;
Described time-series image comprises:
From described first camera collection to first video image handle the very first time sequence image obtain;
From described second camera collection to second video image handle second time-series image obtain.
3. fire detecting system according to claim 2 is characterized in that:
Described first camera adopts the spectral response scope at the CCD of 400nm to 1200nm or cmos image sensor and the cutoff frequency high pass infrared fileter at the above wave band of 850nm;
Described second camera adopts CCD or the cmos image sensor of spectral response scope at 400nm to 1200nm.
4. fire detecting system according to claim 2 is characterized in that, also comprises:
The switching controls module, be connected with described two variable video cameras, described image capture module and described fire identification module, be used for to carry out background illumination intensity and light Distribution calculation from the time-series image that described image capture module gets access to, control described second camera according to the processing requirements of result of calculation or described fire identification module and carry out the switching of mode of operation, also be used for controlling described first camera and carry out the switching of mode of operation according to the processing requirements of described fire identification module.
5. fire detecting system according to claim 2 is characterized in that, described fire identification module further comprises:
Background modeling and update module are connected with described image capture module, are used for the described very first time sequence image and described second time-series image that get access to are analyzed and self study, obtain corresponding long period background image and short period background image;
The flame identification module, be connected with update module with described background modeling, be used for according to described long period background image and short period background image, calculate the flame characteristic parameter of described very first time sequence image and described second time-series image respectively, and described flame characteristic parameter carried out data fusion, obtain the flame probability of happening;
The smog identification module, be connected with update module with described background modeling, be used for according to described long period background image and short period background image, calculate the smoke characteristics parameter of described very first time sequence image and described second time-series image respectively, and described smoke characteristics parameter carried out data fusion, obtain the smog probability of happening;
The fire probability Fusion Module is connected with described smog identification module with described flame identification module, is used for described flame and smog probability of happening are carried out data fusion, determines the fire probability of happening.
6. fire detecting system according to claim 2 is characterized in that, also comprises:
The background light source module is used for when the Luminance Distribution of search coverage and intensity of illumination grade are lower than detection criterion, for described two variable video cameras provide background light source.
7. fire detecting system according to claim 6 is characterized in that, also comprises:
Light source control module, be connected with described background light source module with described image capture module, be used for the time-series image that basis gets access to from described image capture module, the Luminance Distribution of analytical calculation search coverage and intensity of illumination grade, and when the Luminance Distribution of search coverage and intensity of illumination grade are lower than detection criterion, control the startup of described background light source module.
8. fire detecting system according to claim 2 is characterized in that, also comprises:
Set debugging and self-learning module, is connected, be used for the parameter setting of system, and at the generation false alarm or fail to report when alert, utilize the parameter of self-study mechanism update system with described fire identification module.
9. fire detecting system according to claim 8 is characterized in that, also comprises:
Information record display module, be connected with described fire identification module, be used to show, preserve video image and various warning message in the search coverage that collects, and be connected with self-learning module with described setting debugging, false alarm taking place or is failing to report when alert provides the accident video record to described setting debugging and self-learning module.
10. fire detecting system according to claim 2 is characterized in that, also comprises:
The Cloud Terrace and the cradle head control module of controlling described The Cloud Terrace rotation;
Described cradle head control module is connected with described The Cloud Terrace, described fire recognition device and described two variable video cameras, the The Cloud Terrace fixed cycle or the variable period rotation condition that are used for being provided with according to described fire identification module are calculated the presetting bit of The Cloud Terrace or the rotational angle of each position, or calculate the The Cloud Terrace rotational angle according to the field angle of described two variable video cameras, to control the rotation of described The Cloud Terrace.
11. fire detecting system according to claim 2 is characterized in that, also comprises:
The monitored area is provided with module, be connected with described two variable shootings with described fire identification module, be used for recognition result according to described fire identification module, the search coverage of described two variable video cameras is set to a plurality of monitor areas of different nature, described monitor area comprises: highly sensitive smog and flame monitoring district, insensitive smog and flame monitoring district, fault monitoring district and non-surveillance zone, or fire alarm subregion.
12. a fire detecting method is characterized in that, comprising:
Will from comprise first camera that is operated in colour or near infrared pattern at least be operated in colour or the two variable camera acquisition of second camera of white-black pattern to video image be converted to digital picture, and carry out Filtering Processing, obtain time-series image;
Described time-series image is carried out analyzing and processing, obtain flame, smog and fire probability of happening;
More described flame, smog and fire probability of happening and predetermined threshold value, and send corresponding warning message according to comparative result.
13. fire detecting method according to claim 12 is characterized in that:
Described two variable video camera comprises described first camera and described second camera that is operated in colour or white-black pattern that is operated in colour or near infrared pattern;
Described time-series image comprises:
From described first camera collection to first video image handle the very first time sequence image obtain;
From described second camera collection to second video image handle second time-series image obtain.
14. fire detecting method according to claim 13 is characterized in that:
Described first camera adopts the spectral response scope at the CCD of 400nm to 1200nm or cmos image sensor and the cutoff frequency high pass infrared fileter at the above wave band of 850nm;
Described second camera adopts CCD or the cmos image sensor of spectral response scope at 400nm to 1200nm.
15. fire detecting method according to claim 13 is characterized in that, also comprises:
Described time-series image is carried out background illumination intensity and light Distribution calculation, control the step that described second camera carries out the switching of mode of operation according to the processing requirements of result of calculation or fire identification; And
Processing requirements according to fire identification is controlled the step that described first camera carries out the switching of mode of operation.
16. fire detecting method according to claim 13 is characterized in that, described very first time sequence image and described second time-series image are carried out analyzing and processing, the method that obtains flame, smog and fire probability of happening is specially:
Described very first time sequence image and described second time-series image are analyzed and self study, obtained corresponding long period background and short period background image;
According to described long period background and short period background image, calculate the flame characteristic parameter of described very first time sequence image and described second time-series image respectively, and described flame characteristic parameter is carried out data fusion, obtain the flame probability of happening;
According to described long period background and short period background image, calculate the smoke characteristics parameter of described very first time sequence image and described second time-series image respectively, and described smoke characteristics parameter is carried out data fusion, obtain the smog probability of happening;
Described flame and smog probability of happening are carried out data fusion, determine the fire probability of happening.
17. fire detecting method according to claim 16 is characterized in that:
The described very first time sequence image that is used to calculate described flame probability of happening comprises coloured image and near-infrared image, and described second time-series image is coloured image or black white image;
The described very first time sequence image that is used to calculate described smog probability of happening is a coloured image, and described second time-series image is coloured image or black white image.
18. fire detecting method according to claim 17 is characterized in that, determines that the method for flame probability of happening is specially:
Coloured image in the described very first time sequence image is analyzed, obtained the first flame characteristic parameter I
Chara1=f
3{ F
Qi, A
i, PR
i... };
If the described second camera work at present is at color mode, then described second time-series image is a coloured image, and described second time-series image is analyzed, and obtains the second flame characteristic parameter I
Chara2=f
3{ F
Qi, A
i, PR
i... };
Near-infrared image in the described very first time sequence image is analyzed, obtained the 3rd flame characteristic parameter I
Chara3=f
3{ F
Qi, A
i, PR
i... };
Wherein, F
QiFor flashing frequency, A
iBe flame area rate of change, PR
iBe the long period and the short period rate of spread;
Described very first time sequence image or described second time-series image are analyzed, obtained reacting the parameter I of flame flicking feature
Freq
The described flame characteristic parameter that aforementioned calculation obtains is carried out data fusion, obtains described flame probability of happening:
P
Flame(t)＝F
2{I
Chara1，I
Chara2，I
Chara3，I
Freq}。
19. fire detecting method according to claim 17 is characterized in that, determines that the method for described smog probability of happening is specially:
Respectively with current frame image and former frame or preceding some two field picture comparative analyses of described very first time sequence image and described second time-series image, and, obtain reacting the smoke characteristics parameter I of the fast slow characteristic of smog movement with described current frame image and described long period background image or short period background image comparative analysis
Speed, I
SpeedComputing formula be:
I
Speed=(current frame image and former frame or preceding some two field pictures change area)/(current frame image and short period or long period background image change area);
With described very first time sequence image and the described second time-series image five equilibrium or non-ly be divided into a plurality of monitored areas, each monitored area is analyzed respectively, obtained reacting the characteristic parameter I of smog disperse characteristic
Disp=f
1{ R
i, G
i, S
i, T
i, F
i... };
Calculate the characteristic parameter I of the reaction smog movement diffusion property of described very first time sequence image and described second time-series image respectively
Move=f
2{ R
i, G
i, S
i, T
i, F
i, I
Speed...;
Wherein, R
iBe related coefficient, G
iBe graded, S
iFor saturation degree changes, T
iBe texture variations, F
iBe optical flow field;
Get access to described smoke characteristics parameter and carry out data fusion above-mentioned, obtain described smog probability of happening, that is:
P
Smoke(t)＝F
1{I
Disp1，I
Move1，I
Disp2，I
Move2}
Wherein, I
Disp1, I
Move1The smoke characteristics parameter of representing described very first time sequence image, I
Disp2, I
Move2The smoke characteristics parameter of representing described second time-series image.
20. according to claim 18 or 19 described fire detecting methods, it is characterized in that, determine that the method for described fire probability of happening is specially:
P
f1(t)＝P
Smoke(t)[1+(P
Flame(t)-K)]
P
f2(t)＝P
Flame(t)[1+(P
Smoke(t)-K)]
P
Fire(t)＝max{P
f1(t)，P
f2(t)}
Wherein, P
Flame(t) and P
Smoke(t) be respectively t flame probability of happening and smog probability of happening constantly, P
Fire(t) be t fire probability of happening constantly, K is a dead band value.
21. fire detecting method according to claim 13 is characterized in that, also comprises:
The Luminance Distribution of analytical calculation search coverage and intensity of illumination grade, and when the Luminance Distribution of search coverage and intensity of illumination grade are lower than detection criterion, use background light source.
22. fire detecting method according to claim 13 is characterized in that, also comprises:
Detection system is carried out parameter setting, and, utilize self-study mechanism update system parameter false alarm taking place or failing to report when alert.
23. fire detecting method according to claim 13 is characterized in that, also comprises:
Calculate the presetting bit of The Cloud Terrace or the rotational angle of each position according to The Cloud Terrace fixed cycle that is provided with or variable period rotation condition, or calculate the The Cloud Terrace rotational angle, control the rotation of described The Cloud Terrace according to the field angle of described two variable video cameras.
24. fire detecting method according to claim 13 is characterized in that, also comprises:
According to the fire recognition result, the search coverage of described two variable video cameras is set to a plurality of monitor areas of different nature, described monitor area comprises: highly sensitive smog and flame monitoring district, insensitive smog and flame monitoring district, fault monitoring district and non-surveillance zone, or fire alarm subregion.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008101062620A CN101458865B (en) | 2008-05-09 | 2008-05-09 | Fire disaster probe system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008101062620A CN101458865B (en) | 2008-05-09 | 2008-05-09 | Fire disaster probe system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101458865A true CN101458865A (en) | 2009-06-17 |
| CN101458865B CN101458865B (en) | 2012-06-27 |
Family
ID=40769705
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2008101062620A Active CN101458865B (en) | 2008-05-09 | 2008-05-09 | Fire disaster probe system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101458865B (en) |
Cited By (40)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101753995A (en) * | 2009-12-31 | 2010-06-23 | 上海杰远环保科技有限公司 | Digital range pole transmitting system |
| CN102117529A (en) * | 2011-02-22 | 2011-07-06 | 黄剑锋 | Automatic intelligent visible fire alarm system for detecting fire information |
| 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 |
| CN102739963A (en) * | 2012-06-06 | 2012-10-17 | 山东神戎电子股份有限公司 | Day and night two-functional high-temperature mobile target detection alarming system and detection method |
| CN102737466A (en) * | 2012-06-07 | 2012-10-17 | 清华大学 | Method and system for estimating position and intensity of ignition source of fire |
| CN102784451A (en) * | 2012-08-06 | 2012-11-21 | 西安硅光电子科技有限公司 | Automatic positioning flame detection system for three-dimensional space |
| CN102915613A (en) * | 2012-11-19 | 2013-02-06 | 镇江石鼓文智能化系统开发有限公司 | Flame detection device based on video image |
| CN103824409A (en) * | 2012-11-19 | 2014-05-28 | 镇江石鼓文智能化系统开发有限公司 | Detection system for analyzing flame on the basis of video image |
| CN103824410A (en) * | 2012-11-19 | 2014-05-28 | 镇江石鼓文智能化系统开发有限公司 | Detection system for analyzing flame on the basis of video image |
| CN104463112A (en) * | 2014-11-27 | 2015-03-25 | 深圳市科葩信息技术有限公司 | Method and system for carrying out biological recognition through RGB+IR image sensor |
| CN104469312A (en) * | 2014-12-12 | 2015-03-25 | 成都栖林测控科技有限责任公司 | Vision-based fire detecting device and method |
| CN104505919A (en) * | 2015-01-16 | 2015-04-08 | 蔡乌力吉 | Parking lot |
| CN104867265A (en) * | 2015-04-22 | 2015-08-26 | 深圳市佳信捷技术股份有限公司 | Camera apparatus, and fire detection alarm system and method |
| CN104966372A (en) * | 2015-06-09 | 2015-10-07 | 四川汇源光通信有限公司 | Multi-data fusion forest fire intelligent recognition system and method |
| CN105311783A (en) * | 2014-07-29 | 2016-02-10 | 北京市劳动保护科学研究所 | Fire disaster early warning method and system |
| CN106097346A (en) * | 2016-06-13 | 2016-11-09 | 中国科学技术大学 | A kind of video fire hazard detection method of self study |
| CN106384102A (en) * | 2016-09-30 | 2017-02-08 | 深圳火星人智慧科技有限公司 | IR-card-equipped day-night digital network camera flame detection system and method |
| CN106468717A (en) * | 2016-10-18 | 2017-03-01 | 秦艳霞 | Biochip for HUMAN HEAT SHOCK PROTEINS Hhsp HSP70 detection |
| CN106652298A (en) * | 2016-11-16 | 2017-05-10 | 扬州翠佛堂珠宝有限公司 | Firefighting monitoring and alarming system for chemical plant |
| CN106898110A (en) * | 2015-12-14 | 2017-06-27 | 南京三宝弘正视觉科技有限公司 | Method and device is monitored using the fire of CCTV |
| CN107123227A (en) * | 2017-07-06 | 2017-09-01 | 合肥科大立安安全技术股份有限公司 | A kind of embedded image flame detector and its recognition methods based on two waveband |
| US20180102034A1 (en) * | 2015-09-28 | 2018-04-12 | Dongguan Frontier Technology Institute | Fire disaster monitoring method and apparatus |
| CN108520615A (en) * | 2018-04-20 | 2018-09-11 | 芜湖岭上信息科技有限公司 | A kind of fire identification system and method based on image |
| CN109074714A (en) * | 2016-05-04 | 2018-12-21 | 罗伯特·博世有限公司 | For detecting detection device, the method and computer program of event |
| CN109145796A (en) * | 2018-08-13 | 2019-01-04 | 福建和盛高科技产业有限公司 | A kind of identification of electric power piping lane fire source and fire point distance measuring method based on video image convergence analysis algorithm |
| CN109920199A (en) * | 2018-06-06 | 2019-06-21 | 周超强 | Radiation device alarm system based on parameter extraction |
| CN110585625A (en) * | 2019-10-21 | 2019-12-20 | 中车大连机车研究所有限公司 | Self-adjustable locomotive fire prevention and control system |
| CN111219934A (en) * | 2018-11-27 | 2020-06-02 | 宁波泽锦电器科技有限公司 | Ventilation guarantee platform for refrigerating chamber |
| CN111652184A (en) * | 2020-06-19 | 2020-09-11 | 成都通甲优博科技有限责任公司 | Smoke identification method and device, storage medium and data processing equipment |
| CN112235539A (en) * | 2020-10-11 | 2021-01-15 | 任晓娇 | Security monitoring system, monitoring method and medium based on Internet of things |
| CN112242036A (en) * | 2019-07-01 | 2021-01-19 | 国电南瑞科技股份有限公司 | Fire alarm method and system for transformer substation |
| CN112257554A (en) * | 2020-10-20 | 2021-01-22 | 南京恩博科技有限公司 | Forest fire recognition method, system, program and storage medium |
| CN112312081A (en) * | 2020-09-08 | 2021-02-02 | 深圳中核普达测量科技有限公司 | Fire scene intelligent monitoring method and system |
| CN113066251A (en) * | 2021-03-17 | 2021-07-02 | 国核电力规划设计研究院有限公司 | Fire detection and video monitoring linkage system and method |
| CN113506419A (en) * | 2021-06-30 | 2021-10-15 | 中标慧安信息技术股份有限公司 | Indoor safety state analysis method and system based on video data |
| CN113793470A (en) * | 2021-08-09 | 2021-12-14 | 上海腾盛智能安全科技股份有限公司 | Detection device based on dynamic image detection analysis |
| CN114446002A (en) * | 2022-01-17 | 2022-05-06 | 厦门理工学院 | Fire on-line monitoring method, device, medium and system |
| CN115206051A (en) * | 2021-04-12 | 2022-10-18 | 天地融科技股份有限公司 | Smoke detection method and system based on space-time internet of things |
| CN115294717A (en) * | 2022-08-03 | 2022-11-04 | 中国民用航空飞行学院 | Unmanned aerial vehicle fire detection method, device and medium |
| CN115439996A (en) * | 2022-08-31 | 2022-12-06 | 杭州数尔安防科技股份有限公司 | Fire recognition method and device based on spatial point energy analysis |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107180507A (en) * | 2017-05-22 | 2017-09-19 | 肇庆市智高电机有限公司 | Smokescope monitoring system for Internet bar |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1234002C (en) * | 2001-12-21 | 2005-12-28 | 中国科学技术大学 | Aspiration type laser image smoke sensing fire hazard detecting method and device |
| EP1762995B1 (en) * | 2005-09-09 | 2008-05-07 | Siemens Aktiengesellschaft | Detection of smoke with a video camera |
| CN100538757C (en) * | 2005-12-07 | 2009-09-09 | 浙江工业大学 | Fire-disaster monitoring device based on omnibearing vision sensor |
| CN100511305C (en) * | 2007-08-10 | 2009-07-08 | 山东省科学院自动化研究所 | Double wave band image identification fire hazard detection alarming system and monitoring method |
| CN101175202A (en) * | 2007-10-31 | 2008-05-07 | 天津市亚安科技电子有限公司 | Video monitoring apparatus based on double optical band |
| CN201259717Y (en) * | 2008-05-09 | 2009-06-17 | 丁国锋 | Fire disaster probe system |
-
2008
- 2008-05-09 CN CN2008101062620A patent/CN101458865B/en active Active
Cited By (54)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN101753995A (en) * | 2009-12-31 | 2010-06-23 | 上海杰远环保科技有限公司 | Digital range pole transmitting system |
| CN102117529A (en) * | 2011-02-22 | 2011-07-06 | 黄剑锋 | Automatic intelligent visible fire alarm system for detecting fire information |
| CN102739963B (en) * | 2012-06-06 | 2016-01-27 | 山东神戎电子股份有限公司 | One is dual-purpose type high-temperature mobile target detection warning system and detection method round the clock |
| CN102739963A (en) * | 2012-06-06 | 2012-10-17 | 山东神戎电子股份有限公司 | Day and night two-functional high-temperature mobile target detection alarming system and detection method |
| CN102737466A (en) * | 2012-06-07 | 2012-10-17 | 清华大学 | Method and system for estimating position and intensity of ignition source of fire |
| CN102784451A (en) * | 2012-08-06 | 2012-11-21 | 西安硅光电子科技有限公司 | Automatic positioning flame detection system for three-dimensional space |
| CN102784451B (en) * | 2012-08-06 | 2014-12-10 | 西安硅光电子科技有限公司 | Automatic positioning flame detection system for three-dimensional space |
| CN102915613A (en) * | 2012-11-19 | 2013-02-06 | 镇江石鼓文智能化系统开发有限公司 | Flame detection device based on video image |
| CN103824410A (en) * | 2012-11-19 | 2014-05-28 | 镇江石鼓文智能化系统开发有限公司 | Detection system for analyzing flame on the basis of video image |
| CN103824409A (en) * | 2012-11-19 | 2014-05-28 | 镇江石鼓文智能化系统开发有限公司 | Detection system for analyzing flame on the basis of video image |
| CN105311783B (en) * | 2014-07-29 | 2018-09-04 | 北京市劳动保护科学研究所 | Fire alarm method and system |
| CN105311783A (en) * | 2014-07-29 | 2016-02-10 | 北京市劳动保护科学研究所 | Fire disaster early warning method and system |
| CN104463112A (en) * | 2014-11-27 | 2015-03-25 | 深圳市科葩信息技术有限公司 | Method and system for carrying out biological recognition through RGB+IR image sensor |
| CN104469312A (en) * | 2014-12-12 | 2015-03-25 | 成都栖林测控科技有限责任公司 | Vision-based fire detecting device and method |
| CN104505919A (en) * | 2015-01-16 | 2015-04-08 | 蔡乌力吉 | Parking lot |
| CN104867265A (en) * | 2015-04-22 | 2015-08-26 | 深圳市佳信捷技术股份有限公司 | Camera apparatus, and fire detection alarm system and method |
| CN104867265B (en) * | 2015-04-22 | 2018-05-01 | 深圳市佳信捷技术股份有限公司 | Camera device, fire detection alarm system and method |
| CN104966372A (en) * | 2015-06-09 | 2015-10-07 | 四川汇源光通信有限公司 | Multi-data fusion forest fire intelligent recognition system and method |
| US20180102034A1 (en) * | 2015-09-28 | 2018-04-12 | Dongguan Frontier Technology Institute | Fire disaster monitoring method and apparatus |
| CN106898110B (en) * | 2015-12-14 | 2019-04-12 | 南京三宝弘正视觉科技有限公司 | Method and device is monitored using the fire of CCTV |
| CN106898110A (en) * | 2015-12-14 | 2017-06-27 | 南京三宝弘正视觉科技有限公司 | Method and device is monitored using the fire of CCTV |
| CN109074714A (en) * | 2016-05-04 | 2018-12-21 | 罗伯特·博世有限公司 | For detecting detection device, the method and computer program of event |
| CN106097346A (en) * | 2016-06-13 | 2016-11-09 | 中国科学技术大学 | A kind of video fire hazard detection method of self study |
| CN106384102A (en) * | 2016-09-30 | 2017-02-08 | 深圳火星人智慧科技有限公司 | IR-card-equipped day-night digital network camera flame detection system and method |
| CN106468717A (en) * | 2016-10-18 | 2017-03-01 | 秦艳霞 | Biochip for HUMAN HEAT SHOCK PROTEINS Hhsp HSP70 detection |
| CN106652298A (en) * | 2016-11-16 | 2017-05-10 | 扬州翠佛堂珠宝有限公司 | Firefighting monitoring and alarming system for chemical plant |
| CN107123227A (en) * | 2017-07-06 | 2017-09-01 | 合肥科大立安安全技术股份有限公司 | A kind of embedded image flame detector and its recognition methods based on two waveband |
| CN108520615A (en) * | 2018-04-20 | 2018-09-11 | 芜湖岭上信息科技有限公司 | A kind of fire identification system and method based on image |
| CN108520615B (en) * | 2018-04-20 | 2020-08-25 | 吉林省林业科学研究院 | Fire identification system and method based on image |
| CN109920199A (en) * | 2018-06-06 | 2019-06-21 | 周超强 | Radiation device alarm system based on parameter extraction |
| CN109145796A (en) * | 2018-08-13 | 2019-01-04 | 福建和盛高科技产业有限公司 | A kind of identification of electric power piping lane fire source and fire point distance measuring method based on video image convergence analysis algorithm |
| CN111219934A (en) * | 2018-11-27 | 2020-06-02 | 宁波泽锦电器科技有限公司 | Ventilation guarantee platform for refrigerating chamber |
| CN112242036A (en) * | 2019-07-01 | 2021-01-19 | 国电南瑞科技股份有限公司 | Fire alarm method and system for transformer substation |
| CN110585625A (en) * | 2019-10-21 | 2019-12-20 | 中车大连机车研究所有限公司 | Self-adjustable locomotive fire prevention and control system |
| CN111652184A (en) * | 2020-06-19 | 2020-09-11 | 成都通甲优博科技有限责任公司 | Smoke identification method and device, storage medium and data processing equipment |
| CN111652184B (en) * | 2020-06-19 | 2023-06-20 | 成都通甲优博科技有限责任公司 | Smoke identification method, device, storage medium and data processing equipment |
| CN112312081B (en) * | 2020-09-08 | 2023-04-07 | 深圳中核普达测量科技有限公司 | Fire scene intelligent monitoring method and system |
| CN112312081A (en) * | 2020-09-08 | 2021-02-02 | 深圳中核普达测量科技有限公司 | Fire scene intelligent monitoring method and system |
| CN112235539A (en) * | 2020-10-11 | 2021-01-15 | 任晓娇 | Security monitoring system, monitoring method and medium based on Internet of things |
| CN112235539B (en) * | 2020-10-11 | 2023-04-21 | 合肥岭雁科技有限公司 | Security monitoring system, monitoring method and medium based on Internet of things |
| CN112257554A (en) * | 2020-10-20 | 2021-01-22 | 南京恩博科技有限公司 | Forest fire recognition method, system, program and storage medium |
| CN112257554B (en) * | 2020-10-20 | 2021-11-05 | 南京恩博科技有限公司 | Forest fire recognition method, system, program and storage medium based on multiple spectra |
| CN113066251A (en) * | 2021-03-17 | 2021-07-02 | 国核电力规划设计研究院有限公司 | Fire detection and video monitoring linkage system and method |
| CN115206051A (en) * | 2021-04-12 | 2022-10-18 | 天地融科技股份有限公司 | Smoke detection method and system based on space-time internet of things |
| CN113506419B (en) * | 2021-06-30 | 2022-08-19 | 中标慧安信息技术股份有限公司 | Indoor safety state analysis method and system based on video data |
| CN113506419A (en) * | 2021-06-30 | 2021-10-15 | 中标慧安信息技术股份有限公司 | Indoor safety state analysis method and system based on video data |
| CN113793470A (en) * | 2021-08-09 | 2021-12-14 | 上海腾盛智能安全科技股份有限公司 | Detection device based on dynamic image detection analysis |
| CN114446002A (en) * | 2022-01-17 | 2022-05-06 | 厦门理工学院 | Fire on-line monitoring method, device, medium and system |
| CN114446002B (en) * | 2022-01-17 | 2023-10-31 | 厦门理工学院 | Fire on-line monitoring method, device, medium and system |
| CN115294717A (en) * | 2022-08-03 | 2022-11-04 | 中国民用航空飞行学院 | Unmanned aerial vehicle fire detection method, device and medium |
| CN115294717B (en) * | 2022-08-03 | 2023-12-22 | 中国民用航空飞行学院 | Unmanned aerial vehicle fire detection method, unmanned aerial vehicle fire detection device and unmanned aerial vehicle fire detection medium |
| CN115439996A (en) * | 2022-08-31 | 2022-12-06 | 杭州数尔安防科技股份有限公司 | Fire recognition method and device based on spatial point energy analysis |
| CN115439996B (en) * | 2022-08-31 | 2024-02-13 | 杭州数尔安防科技股份有限公司 | Fire identification method and device based on space point energy analysis |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101458865B (en) | 2012-06-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101458865B (en) | Fire disaster probe system and method | |
| CN201259717Y (en) | Fire disaster probe system | |
| US7786877B2 (en) | Multi-wavelength video image fire detecting system | |
| CN101609589A (en) | Multi-frequency image fire detection system | |
| CN101334924B (en) | Fire detection system and fire detection method thereof | |
| US7991187B2 (en) | Intelligent image smoke/flame sensor and detection system | |
| CN100511305C (en) | Double wave band image identification fire hazard detection alarming system and monitoring method | |
| CN101527073B (en) | Fire detection system and method thereof | |
| US6937743B2 (en) | Process and device for detecting fires based on image analysis | |
| CN104143248B (en) | Forest fire detection based on unmanned plane and preventing control method | |
| CN100538757C (en) | Fire-disaster monitoring device based on omnibearing vision sensor | |
| CN1943824B (en) | An automatic fire fighting unit based on omnibearing visual sensor | |
| CN201936415U (en) | Automatic forest fire identification and alarm system | |
| KR20140127574A (en) | Fire detecting system using unmanned aerial vehicle for reducing of fire misinformation | |
| KR102407327B1 (en) | Apparatus for Monitoring Fire And System having the same | |
| CN111488802B (en) | Temperature curve synthesis algorithm utilizing thermal imaging and fire disaster early warning system | |
| CN102176270A (en) | Safety monitoring and fire alarming integrated system and method | |
| CN201091014Y (en) | Fire detecting device | |
| US9648284B2 (en) | Occupancy sensor and associated methods | |
| CN108389359A (en) | A kind of Urban Fires alarm method based on deep learning | |
| CN202523169U (en) | Infrared thermal image type fire detection alarm system | |
| CN203630902U (en) | Improved image fire-hazard detector | |
| CN109040711A (en) | A kind of image-pickup method, monitoring method and imaging sensor | |
| KR102174606B1 (en) | Sense integrated IP network camera, and monitoring system for the same | |
| CN205121786U (en) | Tunnel video fire alarm system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20201103 Address after: 102, 1 / F, building 8, Shangdi 4th Street, Haidian District, Beijing 100089 Patentee after: BEIJING YINGTE WEISHI TECHNOLOGY Co.,Ltd. Address before: 100107 Beijing Chaoyang District Anli road Wankexingyuan 2 Room No. 2007 Patentee before: Ding Guofeng |
|
| TR01 | Transfer of patent right |