CN106530569A - Method and device for fire monitoring - Google Patents
Method and device for fire monitoring Download PDFInfo
- Publication number
- CN106530569A CN106530569A CN201610902204.3A CN201610902204A CN106530569A CN 106530569 A CN106530569 A CN 106530569A CN 201610902204 A CN201610902204 A CN 201610902204A CN 106530569 A CN106530569 A CN 106530569A
- Authority
- CN
- China
- Prior art keywords
- image
- fire
- target image
- early warning
- target
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B17/00—Fire alarms; Alarms responsive to explosion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/07—Target detection
Abstract
The disclosure is about a method and a device for fire monitoring. The method comprises steps of acquiring a current target image in a target area; obtaining a risk level of a fire hazard in the target area based on the target image; and performing corresponding warning operation according to the risk level. In the technical scheme, that whether there is a fire hazard in the target area can be determined based on the target image in the target area, so that the fire early warning can be automatically conducted by monitoring the target image in the target area, the intelligence of the fire alarm system can be improved, and personnel and economic losses can be reduced.
Description
Technical field
It relates to field of artificial intelligence, more particularly to the method and device of fire trouble monitoring.
Background technology
Everyone security of the lives and property of security against fire and we is closely bound up, and fire fighting device construction also more comes by people
Attention, the loss that fire causes can be reduced by fire alarm system.
At present, it is common to use fire alarm system may include：The sensing of the compositions such as Smoke Sensor, temperature sensor sets
It is standby, control device and alert device is may also include, after sensing equipment collects environmental information, control device is when the ring according to collection
Environment information, it is determined that when being possible to fire occur or fire occur, that is, controlling alert device and being reported to the police, certainly, if fire
When warning system includes fire-extinguishing apparatus, after control alert device is reported to the police, can also start fire-extinguishing apparatus is carried out control device
Fire extinguishing.It can be seen that, fire alarm system can reduce fire probability to a certain extent, but, for some target areas, example
Such as：Gas station, dust operation room, fireworks and firecrackers operation room etc., these regions when there is the hidden danger of fire hazardous, example
Such as：Medicated cigarette, just need to be reported to the police otherwise, it is possible to cause huge personnel and economic loss.
The content of the invention
The embodiment of the present disclosure provides the method and device that fire suffers from monitoring.The technical scheme is as follows：
According to the first aspect of the embodiment of the present disclosure, there is provided a kind of method that fire suffers from monitoring, it may include：
Obtain the current target image in target area；
The target area is obtained according to the target image and there is the danger classes that fire is suffered from；
Corresponding early warning operation is performed according to the danger classes.
It can be seen that, according to the target image of target area, it may be determined that target area so, can pass through prison with the presence or absence of fire trouble
Control target area target image, carry out automatically fire and suffer from early warning, improve the intelligent of fire alarm system, can reduce personnel and
Economic loss.
In one embodiment, described danger for obtaining the target area presence fire trouble according to the target image etc.
Level, it may include：
By being identified to the target image, the first figure for carrying that fire suffers from feature is obtained from the target image
Picture；
By carrying out image Time-Series analyses to the target image, obtain and the analysis result；
The danger classes is obtained according to the analysis result.
It can be seen that, identify in the target image in monitoring after carrying the image that fire suffers from feature, image sequential can be passed through
Analysis model, obtain correspondence analysis result, so that it is determined that it is corresponding fire suffer from danger classes, so, it is any can fire hazard
Feature can all be identified, and be analyzed and obtain corresponding analysis result, further increase the accuracy that fire suffers from monitoring, subtract
Few probability for bringing fire because of fire trouble, reduces personnel and economic loss.
In one embodiment, it is described by being identified to the target image, obtain from the target image and take
Suffer from the first image of feature with fire, it may include：
By depth convolutional neural networks algorithm, the target image is identified；
To identify in the target image that carrying the fiery image for suffering from feature is defined as described first image.
So, image recognition is carried out by depth convolutional neural networks, the image for so identifying is more accurate, further
Improve the accuracy rate that fire suffers from monitoring.
In one embodiment, it is described by carrying out image Time-Series analyses to the target image, obtain and described first
The related analysis result of image, it may include：
The target image is input in object module in chronological order, the object module includes following any one：Pass
Return neural network model, conditional random field models；
The related analysis result of described first image is obtained according to the output result of the object module.
It can be seen that, image Time-Series analyses module can be calculated according to input target image, it is determined that suffering from feature with fire is carried
The related analysis result of image, so that it is determined that target area has the danger classes that fire is suffered from, so, image Time-Series analyses can be adopted
Various models, so, the mode that monitoring fire is suffered from is more flexible, also relatively more accurate.
In one embodiment, it is described that corresponding early warning operation is performed according to the danger classes, including：
Target database is obtained, the target database indicates the corresponding relation between danger classes and early warning operation；
Early warning operation corresponding with the danger classes is obtained according to the target database.
It can be seen that, various danger classes can be preset, the default different early warning operation of correspondence, so as to by current target
After image determines that target area has the danger classes that fire is suffered from, different danger classes can carry out different early warning operations so that
It is more careful that fire suffers from monitoring process, more with level, more with practicality.
In one embodiment, early warning operation include it is following at least one：
Behavior early warning, fire reply operation；
Wherein, the behavior early warning indicate it is following at least one：Behavioral guidance, Behavior adjustment, key monitoring region,
The key monitoring time；
It is described reply operation include it is following at least one：Report to the police, start extinguishing device.
It can be seen that, early warning operation has various, different danger classes, can carry out different early warning operations so that fire suffers from monitoring
Process more has level, more with practicality.
According to the second aspect of the embodiment of the present disclosure, there is provided a kind of fire suffers from the device of monitoring, including：
Acquisition module, for obtaining the current target image of target area；
Analysis module, is connected with the acquisition module, is existed for obtaining the target area according to the target image
The danger classes that fire is suffered from；
Warning module, is connected with the analysis module, for performing corresponding early warning operation according to the danger classes.
It can be seen that, according to the target image of target area, it may be determined that target area so, can pass through prison with the presence or absence of fire trouble
Control target area target image, carry out automatically fire and suffer from early warning, improve the intelligent of fire alarm system, can reduce personnel and
Economic loss.
In one embodiment, the analysis module may include：
Identification submodule, for by being identified to the target image, obtaining and carrying fire from the target image
Suffer from the first image of feature；
Analysis submodule, for by carrying out image Time-Series analyses to the target image, obtaining and the analysis result；
Determination sub-module, for obtaining the danger classes according to the analysis result.
It can be seen that, identify in the target image in monitoring after carrying the image that fire suffers from feature, image sequential can be passed through
Analysis model, obtain correspondence analysis result, so that it is determined that it is corresponding fire suffer from danger classes, so, it is any can fire hazard
Feature can all be identified, and be analyzed and obtain corresponding analysis result, further increase the accuracy that fire suffers from monitoring, subtract
Few probability for bringing fire because of fire trouble, reduces personnel and economic loss.
In one embodiment, the identification submodule may include：
Recognition unit, for by depth convolutional neural networks algorithm, being identified to the target image；
Determining unit, for will identify in the target image that carrying the fiery image for suffering from feature is defined as first figure
Picture.
So, image recognition is carried out by depth convolutional neural networks, the image for so identifying is more accurate, further
Improve the accuracy rate that fire suffers from monitoring.
In one embodiment, the analysis submodule may include：
Input block, for the target image is input in object module in chronological order, the object module includes
Following any one：Recurrent neural networks model, conditional random field models；
Analytic unit, for obtaining the related analysis knot of described first image according to the output result of the object module
Really.
It can be seen that, image Time-Series analyses module can be calculated according to input target image, it is determined that suffering from feature with fire is carried
The related analysis result of image, so that it is determined that target area has the danger classes that fire is suffered from, so, image Time-Series analyses can be adopted
Various models, so, the mode that monitoring fire is suffered from is more flexible, also relatively more accurate.
In one embodiment, the warning module may include：
Submodule is searched, for obtaining target database, the target database indicates that danger classes operates it with early warning
Between corresponding relation；
Early warning submodule, for obtaining early warning operation corresponding with the danger classes according to the target database.
It can be seen that, various danger classes can be preset, the default different early warning operation of correspondence, so as to by current target
After image determines that target area has the danger classes that fire is suffered from, different danger classes can carry out different early warning operations so that
It is more careful that fire suffers from monitoring process, more with level, more with practicality.
In one embodiment, the warning module is carried out early warning operation include it is following at least one：
Behavior early warning, fire reply operation；
Wherein, the behavior early warning indicate it is following at least one：Behavioral guidance, Behavior adjustment, key monitoring region,
The key monitoring time；
It is described reply operation include it is following at least one：Report to the police, start extinguishing device.
It can be seen that, early warning operation has various, different danger classes, can carry out different early warning operations so that fire suffers from monitoring
Process more has level, more with practicality.
According to the third aspect of the embodiment of the present disclosure, there is provided a kind of fire suffers from the device of monitoring, and for terminal, its feature exists
In, including：
Processor；
For storing the memorizer of processor executable；
Wherein, the processor is configured to：
Obtain the current target image in target area；
The target area is obtained according to the target image and there is the danger classes that fire is suffered from；
Corresponding early warning operation is performed according to the danger classes.
The technical scheme that embodiment of the disclosure is provided can include following beneficial effect：
In above-mentioned technical proposal, according to the target image of target area, it may be determined that target area is suffered from the presence or absence of fire, this
Sample, can be carried out fire and suffers from early warning, improve the intelligent of fire alarm system by the target image in monitoring objective region automatically,
Personnel and economic loss can be reduced.
It should be appreciated that the general description of the above and detailed description hereinafter are only exemplary and explanatory, not
The disclosure can be limited.
Description of the drawings
During accompanying drawing herein is merged in description and the part of this specification is constituted, show the enforcement for meeting the disclosure
Example, and be used for explaining the principle of the disclosure together with description.
Fig. 1 is the flow chart that the fire according to an exemplary embodiment suffers from monitoring method.
Fig. 2 is the flow chart that the fire according to an exemplary embodiment one suffers from monitoring method.
Fig. 3 is the flow chart that the fire according to an exemplary embodiment two suffers from monitoring method.
Fig. 4 is the block diagram that the fire according to an exemplary embodiment suffers from monitoring device.
Fig. 5 is the block diagram of the analysis module 420 according to an exemplary embodiment.
Fig. 6 is the block diagram of the identification submodule 421 according to an exemplary embodiment.
Fig. 7 is the block diagram of the analysis submodule 422 according to an exemplary embodiment.
Fig. 8 is the block diagram of the warning module 430 according to an exemplary embodiment.
Fig. 9 is the block diagram that the fire according to an exemplary embodiment three suffers from monitoring device.
Figure 10 is a kind of block diagram of the device 1200 for suffering from monitoring for fire according to an exemplary embodiment.
Specific embodiment
Here in detail exemplary embodiment will be illustrated, its example is illustrated in the accompanying drawings.Explained below is related to
During accompanying drawing, unless otherwise indicated, the same numbers in different accompanying drawings represent same or analogous key element.Following exemplary embodiment
Described in embodiment do not represent all embodiments consistent with the disclosure.Conversely, they be only with as appended by
The example of consistent apparatus and method in terms of some described in detail in claims, the disclosure.
At present, fire alarm system can carry out automatic alarm for there is spark, flare or fire, but for one
A bit can fire hazard hidden danger, for example：Lighter, unburnt cigarette end etc. these can fire hazard hidden danger, can't
Carry out automatic alarm.And the technical scheme that the embodiment of the present disclosure is provided, according to the target image of target area, it may be determined that target area
Domain so, can be carried out fire and suffers from early warning, improve fire by the target image in monitoring objective region automatically with the presence or absence of fire trouble
Warning system it is intelligent, personnel and economic loss can be reduced.
Fig. 1 is the flow chart that the fire according to an exemplary embodiment suffers from monitoring method, as shown in figure 1, including following
Step S101-S103：
In step S101, the current target image of target area is obtained.
Consideration for public security, fire-fighting etc. in terms of some, at present many regions be fitted with monitoring camera, for example：Building
Space, railway station, gas station etc., therefore, monitor video can be obtained by existing monitoring camera, so as to obtain target area
Current target image.
In step s 102, target area is obtained according to target image and there is the danger classes that fire is suffered from.
In the embodiment of the present disclosure, image Time-Series analyses can be carried out to the current target image for obtaining, determine target area
There is the danger classes that fire is suffered from.
The relation between the currency and past value of a variable is mainly found in Time-Series analyses, is a kind of longitudinal relation.
The variable of image Time-Series analyses is image, according to known image, calculates unknown image.
Figure Time-Series analyses can be carried out to target image, i.e., according to known image, calculate unknown image, if unknown
Image is it is possible that fire, then can determine that target area has fire and suffers from, also, the probability that fire occurs in unknown image is got over
Greatly, then the danger classes that presence fire in target area is suffered from is higher.
In the embodiment of the present disclosure, first every image in current target image can be identified, it is determined that carry fire suffering from
First image of feature；Then, using image Time-Series analyses model, Time-Series analyses are carried out to current target image, it is determined that with
The related analysis result of first image；Finally, according to analysis result, determine that target area has the danger classes that fire is suffered from.That is root
Target area is obtained according to target image and there is the danger classes that fire is suffered from, it may include：By being identified to target image, from target
The first image for carrying that fire suffers from feature is obtained in image；By carrying out image Time-Series analyses to target image, obtain and the first figure
As related analysis result；Danger classes is obtained according to analysis result.
It can be seen that, first recognize whether the scene of every image is suffered from fire and close, if the image for identifying carries fire and suffers from feature,
Can determine that the image for identifying is the first image.Wherein, image recognition particular technique or model can be it is varied, so,
Depth convolutional neural networks can be adopted, every image in current target image is identified；If the present image of identification
Include that the fire for setting suffers from feature, determine that present image is to carry the first image that fire suffers from feature.Pass through depth convolutional Neural
Network algorithm, is identified to target image；To identify in target image that carrying the fiery image for suffering from feature is defined as the first figure
Picture.Wherein, fire is suffered from feature and can be preset, for example：Medicated cigarette image, lighter image, flare image etc., so, work as knowledge
Medicated cigarette image is included in the present image not gone out, then can determine that present image is to carry the first image that fire suffers from feature.So,
After carrying out image recognition, in current target image, be likely to there is no the first image, it is also possible to have one, two or multiple first
Image.
Suffer from feature due to carrying fire in the first image, therefore, by carrying out image sequential point to current target image
Analysis, after analyzing the analysis result related to the first image, you can according to analysis result, determines that target area has the danger that fire is suffered from
Dangerous grade.And various image Time-Series analyses models can be adopted, Time-Series analyses are carried out to current target image, for example, are adopted
Recurrent neural networks model, or conditional random field models are adopted, by image Time-Series analyses model to current target image
Carry out Time-Series analyses, you can obtain the analysis result related to the first image.I.e. by carrying out image sequential point to target image
Analysis, obtains the analysis result related to the first image.
Here, every image in current target image can be input into the recurrent neural network of setting in chronological order
In model or conditional random field models；Recurrent neural networks model or conditional random field models are calculated, it is determined that with
The related analysis result of one image.I.e. by carrying out image Time-Series analyses to target image, related to the first image dividing is obtained
Analysis result, it may include：Target image is input in object module in chronological order, object module includes following any one：Recurrence
Neural network model, conditional random field models；The related analysis result of first image is obtained according to the output result of object module.
When wherein, using recurrent neural networks model, can be temporally suitable by every image in current target image
Sequence input setting recurrent neural networks model after, by recurrent neural networks model in back- end to end
Propagation training methodes are calculated, it is determined that the analysis result related to the first image.
For example：After being identified to every image in current target image, it is determined that carry the first image of medicated cigarette,
And carry the first image of lighter.Then, when carrying out image using recurrent neural networks model to current target image
Sequence is analyzed, and as two the first image sequential are adjacent, then by the reckoning of recurrent neural networks model, can be obtained and the first image
Related analysis result is the scene that lights a cigarette.
After obtaining the analysis result related to the first image, can determine that target area has what fire was suffered from according to analysis result
Danger classes.Corresponding relation between the danger classes that predeterminable analysis result and fire are suffered from.So, obtain related to the first image
Analysis result after, can according to default corresponding relation, it is determined that fire suffer from danger classes.
For example：Corresponding relation between the danger classes that default analysis result and fire are suffered from is as shown in table 1：So, according to
Table 1, you can it is determined that target area corresponding with analysis result has the danger classes that fire is suffered from.For example：Analysis result is to light a cigarette
Scene, according to as shown in table 1, then can determine that the danger classes that target area presence fire is suffered from is 2 grades.
| Analysis result | The danger classes that fire is suffered from |
| Without igniting scene | 0 |
| May be lit a cigarette scene | 1 |
| Lighted a fire scene, lit a cigarette scene | 2 |
| Throw away pyrotechnics scene | 3 |
| …… | …… |
Table 1
In step s 103, corresponding early warning operation is performed according to danger classes.
In the embodiment of the present disclosure, the corresponding relation between danger classes and early warning operation can be preset in target database,
The corresponding relation of danger classes and early warning operation i.e. in target database, is indicated, so, is grasped with early warning in the danger classes for indicating
In the corresponding relation of work, current early warning operation corresponding with the danger classes for obtaining is searched；Then, perform current early warning operation.
Therefore, in the embodiment of the present disclosure, corresponding early warning operation is performed according to danger classes, including：Obtain target database, number of targets
The corresponding relation between danger classes and early warning operation is indicated according to storehouse；Obtain corresponding with danger classes pre- according to target database
Alert operation.
Early warning operation can have various, different danger classes, can carry out different early warning operations so that fire trouble monitored
Cheng Gengjia has level, more with practicality.Wherein, early warning operation include it is following at least one：Behavior early warning, fire
Calamity reply operation；And wherein, behavior early warning indicate it is following at least one：Behavioral guidance, Behavior adjustment, key monitoring area
Domain, key monitoring time；Fire reply operation include it is following at least one：Report to the police, start extinguishing device.
And behavior early warning can be pointed out by the captions on voice or display device.And can in fire reply operation
Fire disasters protection is carried out including the extinguishing device for starting intelligence, the loss of personnel and property is further reduced.
For example：The corresponding relation that the danger classes of preservation is operated with early warning is as shown in table 2.
| The danger classes that fire is suffered from | Early warning is operated |
| 0 | Early warning operation 1, for example：The early warning in key monitoring region |
| 1 | Early warning operation 2, for example：The early warning of Behavior adjustment |
| 2 | Early warning operation 3, for example：Behavior adjustment and the early warning in terminal monitoring region |
| 3 | Early warning operation 4, for example：Send alarm of fire |
| …… | …… |
Table 2
As shown in table 2, when it is determined that the danger classes suffered from of fire be 1 when, you can determine that corresponding early warning operation is early warning behaviour
Make 2, so as to carry out the early warning of Behavior adjustment.
It can be seen that, according to the target image of target area, it may be determined that target area so, can pass through prison with the presence or absence of fire trouble
Control target area target image, carry out automatically fire and suffer from early warning, improve the intelligent of fire alarm system, can reduce personnel and
Economic loss.Also, different danger classes, can carry out different early warning operations so that it is more careful that fire suffers from monitoring process, more
With level, more with practicality.
Below by operating process set in specific embodiment, the method that the embodiment of the present disclosure is provided is illustrated.
Embodiment one, in the present embodiment, target area can be gas station, and due to gas station, dust operation room etc., these are special
Determining region can not have any Mars to occur, therefore, once occur causing the condition of Mars all be reminded or be corrected,
Therefore, here, the image of Mars can be caused all predeterminable for carrying the image that fire suffers from feature, for example：Medicated cigarette is taken out, movement is taken out
Terminal, takes out lighter, answers mobile terminal etc..
Fig. 2 is the flow chart that the fire according to an exemplary embodiment one suffers from monitoring method, as shown in Fig. 2 including with
Lower step S201-S205：
In step s 201, obtain the current target image of gas station.
In step S202, by being identified to target image, carry that fire suffers from feature the is obtained from target image
One image.
Here it is possible to pass through depth convolutional neural networks, every image in current target image is identified；If
The present image of identification includes that the fire for setting suffers from feature, determines that present image is to carry the first image that fire suffers from feature.Will
Identify in target image that carrying the fiery image for suffering from feature is defined as the first image.
For example：The present image of identification includes taking out medicated cigarette evidence, then can determine that present image is to carry fire to suffer from feature
First image.Or, the present image of identification includes taking out mobile terminal, then can determine that present image for carrying fire and suffer from feature
The first image.Or, the present image of identification includes smoking scene, then can determine that present image is to carry fire to suffer from feature
First image.It is of course also possible to not include in any image of identification that the fire of setting suffers from feature, i.e. current monitor video data
In there is no the first image.
In step S203, by recurrent neural networks model, image Time-Series analyses are carried out to current target image, is obtained
Take the analysis result related to the first image.
Here, use recurrent neural networks model.By the data of every image in current target image, on time
Between sequentially input in the recurrent neural networks model of setting, by recurrent neural networks model in back- end to end
Propagation training methodes are calculated, obtain the analysis result related to the first image.
For example：There is no the first image in current target image, then by the reckoning of recurrent neural networks model, it is determined that
Analysis result is without igniting scene.If there is the first image for taking out cigarette scene in current target image, sequentially in time,
Do not have the first image in successive image, then analysis result is can determine that for the scene that may light a cigarette.If having in current target image
The first image for taking out cigarette scene and the first image for taking out lighter scene, then according to Time-Series analyses, it is determined that analysis knot
Fruit is the scene that lights a cigarette.If there is the first image of scene of lighting up in current target image, and in successive image without with
The first related scene of medicated cigarette, then can determine that analysis result for throwing away pyrotechnics scene.
In step S204, according to analysis result, obtain gas station and there is the danger classes that fire is suffered from.
Here, the corresponding relation between the danger classes that predeterminable analysis result and fire are suffered from, can be as shown in table 1, so as to root
According to table 1, obtain target area corresponding with analysis result and there is the danger classes that fire is suffered from.For example：Analysis result is possible to light a cigarette
Scene, then can determine that the danger classes of fire trouble is 1 grade；Analysis result then can determine that danger of fire trouble etc. for throwing away pyrotechnics scene
Level is 3 grades.
In step S205, corresponding early warning operation is performed according to danger classes.
Here, target database can be obtained, and in target database, indicates danger classes and the early warning operation of fire trouble
Corresponding relation, can be as shown in table 2, so as to according to table 2, it may be determined that early warning operation corresponding with danger classes.For example：Current
There is the first image for taking out cigarette scene in target image, then sequentially in time, in successive image, there is no the first image, then can be true
Setting analysis result for lighting a cigarette scene, so that it is determined that fire suffer from danger classes for 1 grade, according to table 2, then can determine that corresponding pre-
Alert operation is early warning operation 2, so as to perform early warning operation 2, here it is possible to be the early warning for carrying out Behavior adjustment.For example, send out
Go out the voice message of Behavior adjustment.And if have the first image of scene of lighting up in current target image, and successive image
In without first scene related to medicated cigarette, then can determine that analysis result for throwing away pyrotechnics scene, so that it is determined that fire suffer from danger
Grade is 3 grades, according to table 2, then can determine that corresponding early warning operation for early warning operation 4, so as to perform early warning operation 4, here, can
Being to send alarm of fire.
It can be seen that, in the present embodiment, identify in the target image in monitoring after carrying the image that fire suffers from feature, can lead to
Image Time-Series analyses model is crossed, correspondence analysis result is obtained, so that it is determined that the danger classes that corresponding fire is suffered from, so, Ren Heke
The feature of fire hazard can all be identified, and be analyzed and obtain corresponding analysis result, further increase fire and suffer from monitoring
Accuracy, reduce and bring the probability of fire because of fire trouble, reduce personnel and economic loss.
In embodiment two, the present embodiment, target area can be the larger region of artificial abortion, for example, airport, railway station etc.,
The larger region of artificial abortion causes the requirement of fire not as gas station etc. to Mars, but, because artificial abortion is larger, also need emphasis to prevent
Model fire, will otherwise cause larger casualties.
Fig. 3 is the flow chart that the fire according to an exemplary embodiment two suffers from monitoring method, as shown in figure 3, including with
Lower step S301-S305：
In step S301, the current target image in the larger region of artificial abortion is obtained.
In step s 302, by being identified to target image, carry that fire suffers from feature the is obtained from target image
One image.
In the present embodiment, corresponding fire is set for the larger region of artificial abortion and suffer from characteristic image, for example：Smoking image, throws away
Cigarette end image is abandoned, images of items etc. is burnt.Here, by depth convolutional neural networks, current target image is identified；
To identify in target image that carrying the fiery image for suffering from feature is defined as the first image.
In step S303, using conditional random field models, image Time-Series analyses are carried out to current target image, it is determined that
The analysis result related to the first image.
Condition random field (conditional random fields, CRF) is a kind of discriminant probabilistic model, is random
One kind of field, can be used in Time-Series analyses.Typically, condition random field is using a kind of probability graph model, with expression over long distances
The ability of dependency and overlapping property feature, can preferably solve the advantage of the problems such as marking (classification) biasing, and all spies
Levying to carry out global normalization, can try to achieve the optimal solution of the overall situation.For example：There is scene of lighting up in current target image
First image, and conditional random field models can be then adopted without the first scene related to medicated cigarette in successive image, it is determined that analysis
As a result it is to throw away pyrotechnics scene.And in current target image burn article scene the first image, then can adopt condition random field
Model, determines that analysis result is fire scene.
In step s 304, according to the analysis result for determining, determine that the larger region of artificial abortion has the danger classes that fire is suffered from.
Here, the corresponding relation between the danger classes that also predeterminable analysis result and fire are suffered from, can be as shown in table 3：
| Analysis result | The danger classes that fire is suffered from |
| Without igniting scene | 0 |
| Light a cigarette or put usual articles scene | 1 |
| Point inflammable articles scene | 2 |
| Point is more than set point or setting number of articles scene | 3 |
| …… | …… |
Table 3
Here, it is for the feature in the larger region of artificial abortion, corresponding between presupposition analysis result and the danger classes that fire is suffered from
Relation.If analysis result is to light a cigarette or put a piece of paper scene, can determine that the danger classes of fire trouble is 1 grade according to table 3, if analysis
As a result it is to burn the scenes such as substantial amounts of paper, cloth, chest, then can determine that the danger classes of fire trouble is 3 grades according to table 3.
In step S305, according to danger classes, corresponding early warning operation is performed.
Here, target database can be obtained, and the danger classes for indicating fire trouble in target data is right with what early warning was operated
Should be related to, also can be as shown in table 2, then, early warning operation corresponding with danger classes is obtained according to target database.For example：Danger
Dangerous grade is 0, can carry out the captions prompting in key monitoring region on a display screen.Or, danger classes is 5, due to artificial abortion compared with
Big region such as railway station or airport, may all be equipped with automatic sprinkler, and the device can be intelligent extinguishing device, because
This, corresponding early warning operation is fire reply operation, including：Alarm of fire is sent, and starts automatic sprinkler.Perform
Alarm of fire, and start the fire reply operation of automatic sprinkler.
So as to, different danger classes, different early warning operations can be carried out so that it is more careful that fire suffers from monitoring process, more
With level, more with practicality.
It can be seen that, in the present embodiment, by image Time-Series analyses being carried out to current target image, determine artificial abortion's large area
With the presence or absence of fire trouble, so, fire can be carried out automatically and suffers from early warning, improve the intelligence of fire alarm system by monitoring image video
Energy property, and according to image Time-Series analyses, determine that the larger region of artificial abortion has the danger classes that fire is suffered from, improve fire trouble and eliminate
Accuracy so that fire suffers from monitoring more accurately, can more reduce personnel and economic loss.
It is following for disclosure device embodiment, can be used for performing method of disclosure embodiment.
Fig. 4 be according to an exemplary embodiment fire suffer from monitoring block diagram, the device can by software, hardware or
Being implemented in combination with of both persons becomes some or all of of electronic equipment.As shown in figure 4, the fire suffers from monitoring device including：Obtain
Module 410, analysis module 420 and warning module 430.Wherein,
Acquisition module 410, is configured to obtain the current target image of target area.
Analysis module 420, is connected with acquisition module 410, is configured to obtain target area presence fire according to target image
The danger classes of trouble.
Warning module 430, is connected with analysis module 420, is configured to perform corresponding early warning operation according to danger classes.
It can be seen that, fire suffers from the device of monitoring can be according to the target image of target area, it may be determined that target area is with the presence or absence of fire
Suffer from, so, fire can be carried out automatically and suffer from early warning, improve the intelligence of fire alarm system by the target image in monitoring objective region
Energy property, can reduce personnel and economic loss.
Fig. 5 is the block diagram of the analysis module 420 according to an exemplary embodiment.As shown in figure 5, analysis module 420
May include：Identification submodule 421, analysis submodule 422 and determination sub-module 423, wherein,
Identification submodule 421, is configured to be identified target image, obtains and carry fire trouble from target image
First image of feature.
Analysis submodule 422, is configured to carry out target image image Time-Series analyses, obtains and the first image phase
The analysis result of pass.
Determination sub-module 423, is configured to obtain danger classes according to analysis result.
It can be seen that, analysis module Yi Dan identify in the target image carry fire suffer from feature image after, can pass through image when
Sequence analysis model, obtains correspondence analysis result, so that it is determined that the danger classes that corresponding fire is suffered from, so, any to cause fire
The feature of calamity can all be identified, and be analyzed and obtain corresponding analysis result, further increase the accuracy that fire suffers from monitoring,
Reduce and the probability of fire is brought because of fire trouble, reduce personnel and economic loss.
Fig. 6 is the block diagram of the identification submodule 421 according to an exemplary embodiment.As shown in fig. 6, identification submodule
421 may also include：Recognition unit 4211 and determining unit 4212, wherein,
Recognition unit 4211, is configured to depth convolutional neural networks algorithm, target image is identified.
Determining unit 4212, is configured to identify in target image that carrying the fiery image for suffering from feature is defined as the first figure
Picture.
So, image recognition is carried out by depth convolutional neural networks, the image for so identifying is more accurate, further
Improve the accuracy rate that fire suffers from monitoring.
Fig. 7 is the block diagram of the analysis submodule 422 according to an exemplary embodiment.As shown in fig. 7, analysis submodule
422 may include：Input block 4221 and analytic unit 4222, wherein,
Input block 4221, is configured to target image is input in object module in chronological order, and object module includes
Following any one：Recurrent neural networks model, conditional random field models.
Analytic unit 4222, is configured to obtain the related analysis knot of the first image according to the output result of object module
Really.
It can be seen that, image Time-Series analyses module can be calculated according to input target image, it is determined that suffering from feature with fire is carried
The related analysis result of image, so that it is determined that target area has the danger classes that fire is suffered from, so, image Time-Series analyses can be adopted
Various models, so, the mode that monitoring fire is suffered from is more flexible, also relatively more accurate.
Certainly, analytic unit, the back- end to end in being also configured to by recurrent neural networks model
Propagation training methodes are calculated, it is determined that the analysis result related to the first image.
It can be seen that, recurrent neural networks model is relatively common more easy model so that it is fairly simple that fire suffers from monitoring process, easily
Realize, save flow process.
Fig. 8 is the block diagram of the warning module 430 according to an exemplary embodiment.As shown in figure 8, warning module 430
May include：Submodule 431 and early warning submodule 432 is searched, wherein,
Submodule 431 is searched, is configured to obtain target database, target database indicates that danger classes is operated with early warning
Between corresponding relation.
Early warning submodule 432, is configured to obtain early warning operation corresponding with danger classes according to target database.
It can be seen that, various danger classes can be preset, the default different early warning operation of correspondence, so as to by current target
After image determines that target area has the danger classes that fire is suffered from, different danger classes can carry out different early warning operations so that
It is more careful that fire suffers from monitoring process, more with level, more with practicality.
In one embodiment of the disclosure, early warning that warning module 430 is carried out operation include it is following at least one：
Behavior early warning, fire reply operation；Wherein, behavior early warning indicate it is following at least one：Behavioral guidance, behavior
Correction, key monitoring region, key monitoring time；Reply operation include it is following at least one：Report to the police, start extinguishing device.This
Sample, early warning operation have various, different danger classes, can carry out different early warning operations so that fire is suffered from monitoring process and more had
Having levels property, more with practicality.
Below by operating process set in specific embodiment, the device that the embodiment of the present disclosure is provided is illustrated.
Embodiment three, Fig. 9 are the block diagrams that the fire according to an exemplary embodiment three suffers from monitoring, as shown in figure 9, the dress
Put including：Acquisition module 410, analysis module 420 and warning module 430, wherein, analysis module 420 may include to recognize submodule
Block 421, analysis submodule 422 and determination sub-module 423, and recognize that submodule 421 includes：Recognition unit 4211 and determining unit
4212.Analysis submodule 422 includes：Input block 4221 and analytic unit 4222.
In the present embodiment, target area can be gas station.
So, acquisition module 410 can gas station current target image.And the identification submodule in analysis module 420
Every image in 421 pairs of current target images is identified, it is determined that carry the first image that fire suffers from feature, i.e., by mesh
Logo image is identified, and obtains the first image for carrying that fire suffers from feature from target image.Wherein, recognition unit 4211 can lead to
Depth convolutional neural networks algorithm is crossed, target image is identified, and if wrapping in the present image of the identification of determining unit 4212
The fire for including setting suffers from feature, determines that present image is to carry the first image that fire suffers from feature.
So, the analysis submodule 422 in analysis module 420 adopts recurrent neural networks model, to current target figure
As carrying out image Time-Series analyses, it is determined that the analysis result related to the first image.Wherein, target image is pressed by input block 4221
In the recurrent neural networks model of time sequencing input setting, and analytic unit 4222 is by the end in recurrent neural networks model
Back-propagation training methodes to end are calculated, it is determined that the analysis result related to the first image.
Then, according to analysis result, the determination sub-module 423 in analysis module 420 determines that gas station has the danger that fire is suffered from
Dangerous grade.Here, the corresponding relation between the danger classes that predeterminable analysis result and fire are suffered from, can be as shown in table 1, so as to root
According to table 1, it is determined that there is the danger classes that fire is suffered from target area corresponding with analysis result.
So as to warning module 430 can perform corresponding early warning operation according to danger classes.Here, target can be obtained in advance
Data base, and target database indicates the danger classes of fire trouble and the corresponding relation of early warning operation, can be as shown in table 2, so as to,
According to table 2, it may be determined that early warning operation corresponding with danger classes.For example：Have the of scene of lighting up in current target image
One image, and in successive image, there is no first scene related to medicated cigarette, then analysis module 420 can determine that analysis result for throwing away
Pyrotechnics scene, so as to confirmable fire suffers from danger classes for 3 grades, then, according to table 2, then warning module 430 can determine that corresponding
Early warning operation is early warning operation 4, and so as to perform early warning operation 4, here, warning module 430 can send alarm of fire.
It can be seen that, in the present embodiment, after analysis module is identified in the target image and carries the image that fire suffers from feature, can
By image Time-Series analyses model, correspondence analysis result is obtained, so that it is determined that the danger classes that corresponding fire is suffered from, so, any
Can the feature of fire hazard can all be identified, and be analyzed and obtain corresponding analysis result, further increase fire and suffer from prison
The accuracy of survey, reduces and brings the probability of fire because of fire trouble, reduce personnel and economic loss.Also, warning module can be according to not
With danger classes, carry out different early warning operations so that it is more careful that fire suffers from monitoring process, more with level, more has
There is practicality.
The embodiment of the present disclosure provides the device that a kind of fire suffers from monitoring, is configured to terminal, including：
Processor；
It is configured to store the memorizer of processor executable；
Wherein, processor is configured to：
Obtain the current target image in target area；
The target area is obtained according to the target image and there is the danger classes that fire is suffered from；
Corresponding early warning operation is performed according to the danger classes.
The technical scheme that embodiment of the disclosure is provided can include following beneficial effect：
The above-mentioned technical proposal that embodiment of the disclosure is provided, according to the target image of target area, it may be determined that target area
Domain so, can be carried out fire and suffers from early warning, improve fire by the target image in monitoring objective region automatically with the presence or absence of fire trouble
Warning system it is intelligent, personnel and economic loss can be reduced.
With regard to the device in above-described embodiment, wherein modules perform the concrete mode of operation in relevant the method
Embodiment in be described in detail, explanation will be not set forth in detail herein.
Figure 10 is a kind of for the fiery block diagram for suffering from the device 1200 monitored, the device according to an exemplary embodiment
Suitable for terminal unit.For example, device 1200 can be mobile phone, computer, digital broadcast terminal, messaging devices,
Game console, tablet device, armarium, body-building equipment, personal digital assistant etc..
With reference to Figure 10, device 1200 can include following one or more assemblies：Process assembly 1202, memorizer 1204,
Power supply module 1206, multimedia groupware 1208, audio-frequency assembly 1210, the interface 1212 of input/output (I/O), sensor cluster
1214, and communication component 1216.
The integrated operation of 1202 usual control device 1200 of process assembly, such as with display, call, data communication,
The associated operation of camera operation and record operation.Process assembly 1202 can include one or more processors 1220 to perform
Instruction, to complete all or part of step of above-mentioned method.Additionally, process assembly 1202 can include one or more moulds
Block, the interaction being easy between process assembly 1202 and other assemblies.For example, process assembly 1202 can include multi-media module,
To facilitate the interaction between multimedia groupware 1208 and process assembly 1202.
Memorizer 1204 is configured to store various types of data to support the operation in equipment 1200.These data
Example includes the instruction of any application program for operating on device 1200 or method, contact data, telephone book data,
Message, picture, video etc..Memorizer 1204 can by any kind of volatibility or non-volatile memory device or they
Combination realizes, such as static RAM (SRAM), Electrically Erasable Read Only Memory (EEPROM), it is erasable can
Program read-only memory (EPROM), programmable read only memory (PROM), read only memory (ROM), magnetic memory, flash memory
Reservoir, disk or CD.
Power supply module 1206 provides electric power for the various assemblies of device 1200.Power supply module 1206 can include power management
System, one or more power supplys, and other generate, manage and distribute the component that electric power is associated with for device 1200.
Multimedia groupware 1208 is included in the screen of one output interface of offer between device 1200 and user.At some
In embodiment, screen can include liquid crystal display (LCD) and touch panel (TP).If screen includes touch panel, screen
Touch screen is may be implemented as, to receive the input signal from user.Touch panel includes one or more touch sensors
With the gesture on sensing touch, slip and touch panel.Touch sensor can not only sensing touch or sliding action border,
But also detect the persistent period point related to touch or slide and pressure.In certain embodiments, multimedia groupware
1208 include a front-facing camera and/or post-positioned pick-up head.When equipment 1200 is in operator scheme, such as screening-mode or video
During pattern, front-facing camera and/or post-positioned pick-up head can receive outside multi-medium data.Each front-facing camera and rearmounted
Photographic head can be a fixed optical lens system or have focusing and optical zoom capabilities.
Audio-frequency assembly 1210 is configured to output and/or input audio signal.For example, audio-frequency assembly 1210 includes a wheat
Gram wind (MIC), when device 1200 is in operator scheme, such as call model, logging mode and speech recognition mode, mike quilt
It is configured to receive external audio signal.The audio signal for being received can be further stored in memorizer 1204 or via communication
Component 1216 sends.In certain embodiments, audio-frequency assembly 1210 also includes a speaker, for exports audio signal.
I/O interfaces 1212 are to provide interface, above-mentioned peripheral interface module between process assembly 1202 and peripheral interface module
Can be keyboard, click wheel, button etc..These buttons may include but be not limited to：Home button, volume button, start button and
Locking press button.
Sensor cluster 1214 includes one or more sensors, and the state for providing various aspects for device 1200 is commented
Estimate.For example, sensor cluster 1214 can detect the opening/closed mode of equipment 1200, such as relative localization of component, group
Display and keypad of the part for device 1200, sensor cluster 1214 can be with 1,200 1 groups of detection means 1200 or device
The position change of part, user are presence or absence of with what device 1200 was contacted, 1200 orientation of device or acceleration/deceleration and device
1200 temperature change.Sensor cluster 1214 can include proximity transducer, be configured to do not having any physics to connect
The presence of object nearby is detected when tactile.Sensor cluster 1214 can also include optical sensor, such as CMOS or ccd image sensing
Device, for used in imaging applications.In certain embodiments, the sensor cluster 1214 can also include acceleration sensing
Device, gyro sensor, Magnetic Sensor, pressure transducer or temperature sensor.
Communication component 1216 is configured to facilitate the communication of wired or wireless way between device 1200 and other-end.Dress
Put 1200 and can access wireless network based on communication standard, such as WiFi, 2G or 3G, or combinations thereof.It is exemplary at one
In embodiment, communication component 1216 receives broadcast singal or the broadcast correlation from external broadcasting management system via broadcast channel
Information.In one exemplary embodiment, communication component 1216 also includes near-field communication (NFC) module, to promote junction service.
For example, RF identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra broadband (UWB) skill can be based in NFC module
Art, bluetooth (BT) technology and other technologies are realizing.
In the exemplary embodiment, device 1200 can be by one or more application specific integrated circuits (ASIC), numeral
Signal processor (DSP), digital signal processing appts (DSPD), PLD (PLD), field programmable gate array
(FPGA), controller, microcontroller, microprocessor or other electronic components are realized, for performing said method.
In the exemplary embodiment, a kind of non-transitorycomputer readable storage medium including instruction, example are additionally provided
Such as include the memorizer 1204 of instruction, above-mentioned instruction can be performed to complete said method by the processor 820 of device 1200.Example
Such as, non-transitorycomputer readable storage medium can be ROM, random access memory (RAM), CD-ROM, tape, floppy disk and
Optical data storage devices etc..
A kind of non-transitorycomputer readable storage medium, when the instruction in storage medium is held by the processor of device 1200
During row so that device 1200 is able to carry out the method shown in Fig. 1, method includes：
Obtain the current target image in target area；
The target area is obtained according to the target image and there is the danger classes that fire is suffered from；
Corresponding early warning operation is performed according to the danger classes.
It is described to obtain the danger classes that the target area has fire trouble according to the target image, it may include：
By being identified to the target image, the first figure for carrying that fire suffers from feature is obtained from the target image
Picture；
By carrying out image Time-Series analyses to the target image, the analysis result related to described first image is obtained；
The danger classes is obtained according to the analysis result.
It is described by being identified to the target image, from the target image obtain carry fire suffers from the first of feature
Image, it may include：
By depth convolutional neural networks algorithm, the target image is identified；
To identify in the target image that carrying the fiery image for suffering from feature is defined as described first image.
It is described by carrying out image Time-Series analyses to the target image, obtain the analysis related to described first image and tie
Really, it may include：
The target image is input in object module in chronological order, the object module includes following any one：Pass
Return neural network model, conditional random field models；
The related analysis result of described first image is obtained according to the output result of the object module.
It is described that corresponding early warning operation is performed according to the danger classes, including：
Target database is obtained, the target database indicates the corresponding relation between danger classes and early warning operation；
Early warning operation corresponding with the danger classes is obtained according to the target database.
Early warning operation include it is following at least one：
Behavior early warning, fire reply operation；
Wherein, the behavior early warning indicate it is following at least one：Behavioral guidance, Behavior adjustment, key monitoring region,
The key monitoring time；
It is described reply operation include it is following at least one：Report to the police, start extinguishing device.
Those skilled in the art will readily occur to its of the disclosure after considering description and putting into practice disclosure disclosed herein
Its embodiment.The application is intended to any modification, purposes or the adaptations of the disclosure, these modifications, purposes or
Person's adaptations follow the general principle of the disclosure and including the undocumented common knowledge in the art of the disclosure
Or conventional techniques.Description and embodiments are considered only as exemplary, and the true scope of the disclosure and spirit are by following
Claim is pointed out.
It should be appreciated that the disclosure is not limited to the precise results for being described above and being shown in the drawings, and
And various modifications and changes can be being carried out without departing from the scope.The scope of the present disclosure is limited only by appended claim.
Claims (13)
1. a kind of method that fire suffers from monitoring, it is characterised in that include：
Obtain the current target image in target area；
The target area is obtained according to the target image and there is the danger classes that fire is suffered from；
Corresponding early warning operation is performed according to the danger classes.
2. the method for claim 1, it is characterised in that described the target area is obtained according to the target image to deposit
Fire suffer from danger classes, including：
By being identified to the target image, the first image for carrying that fire suffers from feature is obtained from the target image；
By carrying out image Time-Series analyses to the target image, the analysis result related to described first image is obtained；
The danger classes is obtained according to the analysis result.
3. method as claimed in claim 2, it is characterised in that described by being identified to the target image, from described
The first image for carrying that fire suffers from feature is obtained in target image, including：
By depth convolutional neural networks algorithm, the target image is identified；
To identify in the target image that carrying the fiery image for suffering from feature is defined as described first image.
4. method as claimed in claim 2, it is characterised in that described by carrying out image sequential point to the target image
Analysis, obtains the analysis result related to described first image, including：
The target image is input in object module in chronological order, the object module includes following any one：Recurrence god
Jing network modeies, conditional random field models；
The analysis result is obtained according to the output result of the object module.
5. the method for claim 1, it is characterised in that described that corresponding early warning behaviour is performed according to the danger classes
Make, including：
Target database is obtained, the target database indicates the corresponding relation between danger classes and early warning operation；
Early warning operation corresponding with the danger classes is obtained according to the target database.
6. the method for claim 1, it is characterised in that the early warning operation include it is following at least one：
Behavior early warning, fire reply operation；
Wherein, the behavior early warning indicate it is following at least one：Behavioral guidance, Behavior adjustment, key monitoring region, emphasis
Monitoring period；
It is described reply operation include it is following at least one：Report to the police, start extinguishing device.
7. a kind of fire suffers from the device of monitoring, it is characterised in that include：
Acquisition module, for obtaining the current target image of target area；
Analysis module, is connected with the acquisition module, is existed for obtaining the target area according to the target image
The danger classes that fire is suffered from；
Warning module, is connected with the analysis module, for performing corresponding early warning operation according to the danger classes.
8. device as claimed in claim 7, it is characterised in that the analysis module includes：
Identification submodule, it is special for by being identified to the target image, obtaining carrying fire trouble from the target image
The first image levied；
Analysis submodule, for by carrying out image Time-Series analyses to the target image, obtaining related to described first image
Analysis result；
Determination sub-module, for obtaining the danger classes according to the analysis result.
9. device as claimed in claim 8, it is characterised in that the identification submodule includes：
Recognition unit, for by depth convolutional neural networks algorithm, being identified to the target image；
Determining unit, for will identify in the target image that carrying the fiery image for suffering from feature is defined as described first image.
10. device as claimed in claim 8, it is characterised in that the analysis submodule includes：
Input block, for the target image is input in object module in chronological order, the object module includes following
Any one：Recurrent neural networks model, conditional random field models；
Analytic unit, for obtaining the analysis result according to the output result of the object module.
11. devices as claimed in claim 7, it is characterised in that the warning module includes：
Submodule is searched, for obtaining target database, the target database is indicated between danger classes and early warning operation
Corresponding relation；
Early warning submodule, for obtaining early warning operation corresponding with the danger classes according to the target database.
12. devices as claimed in claim 7, it is characterised in that early warning that the warning module is carried out operation include with down to
One item missing：
Behavior early warning, fire reply operation；
Wherein, the behavior early warning indicate it is following at least one：Behavioral guidance, Behavior adjustment, key monitoring region, emphasis
Monitoring period；
It is described reply operation include it is following at least one：Report to the police, start extinguishing device.
A kind of 13. fire suffer from the device of monitoring, for terminal, it is characterised in that include：
Processor；
For storing the memorizer of processor executable；
Wherein, the processor is configured to：
Obtain the current target image in target area；
The target area is obtained according to the target image and there is the danger classes that fire is suffered from；
Corresponding early warning operation is performed according to the danger classes.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610902204.3A CN106530569A (en) | 2016-10-17 | 2016-10-17 | Method and device for fire monitoring |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610902204.3A CN106530569A (en) | 2016-10-17 | 2016-10-17 | Method and device for fire monitoring |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106530569A true CN106530569A (en) | 2017-03-22 |
Family
ID=58332597
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610902204.3A Pending CN106530569A (en) | 2016-10-17 | 2016-10-17 | Method and device for fire monitoring |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106530569A (en) |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107770244A (en) * | 2017-09-07 | 2018-03-06 | 深圳市盛路物联通讯技术有限公司 | Data transfer control method and Related product |
| CN108447212A (en) * | 2018-03-22 | 2018-08-24 | 河南科技大学 | A kind of intelligent repository fire alarm system and its alarm method |
| CN108830305A (en) * | 2018-05-30 | 2018-11-16 | 西南交通大学 | A kind of real-time fire monitoring method of combination DCLRN network and optical flow method |
| CN109561281A (en) * | 2018-11-15 | 2019-04-02 | 陕西瑞海工程智慧数据科技有限公司 | Industrial equipment method for safety monitoring, device, control device and readable storage medium storing program for executing |
| CN109920214A (en) * | 2017-12-12 | 2019-06-21 | 杭州海康威视系统技术有限公司 | A kind of picture pick-up device and information cuing method |
| CN110051954A (en) * | 2019-04-19 | 2019-07-26 | 辽宁科技大学 | A kind of split type Super High full-automatic fire-extinguishing machine people and its control method |
| CN110102002A (en) * | 2019-05-15 | 2019-08-09 | 上海荷福人工智能科技（集团）有限公司 | The early warning of fire-fighting robot based on artificial intelligence and fire extinguishing system and method |
| CN110115819A (en) * | 2019-05-14 | 2019-08-13 | 上海荷福人工智能科技（集团）有限公司 | Fire water monitor orientation early warning and extinguishing method, storage medium and fire water monitor based on artificial intelligence |
| CN110379121A (en) * | 2019-05-07 | 2019-10-25 | 深圳市泰和安科技有限公司 | A kind of hazard alarm method and server |
| CN110775930A (en) * | 2019-10-18 | 2020-02-11 | 北京地平线机器人技术研发有限公司 | Security protection device, method, computer-readable storage medium, and electronic device |
| CN111105582A (en) * | 2019-11-12 | 2020-05-05 | 西安天和防务技术股份有限公司 | Forest fire prevention monitoring method and system, computer equipment and readable storage medium |
| EP3667632A1 (en) * | 2018-12-13 | 2020-06-17 | Robert Bosch GmbH | Method for identifying a person smoking |
| CN111583567A (en) * | 2020-06-01 | 2020-08-25 | 江西憶源多媒体科技有限公司 | Forest fire prevention early warning method and device |
| CN111753594A (en) * | 2019-03-29 | 2020-10-09 | 杭州海康威视数字技术股份有限公司 | Danger identification method, device and system |
| CN111767771A (en) * | 2019-11-06 | 2020-10-13 | 母宗军 | Target site identification system and method |
| CN112418004A (en) * | 2020-11-05 | 2021-02-26 | 烟台艾睿光电科技有限公司 | Fire safety detection method and related components |
| CN113392710A (en) * | 2021-05-19 | 2021-09-14 | 上海可深信息科技有限公司 | Intelligent recognition method and system for smoking behavior |
| CN113469156A (en) * | 2021-09-03 | 2021-10-01 | 南通海腾铜业有限公司 | Artificial intelligence-based safety risk early warning assessment method, device and equipment |
| CN113807216A (en) * | 2021-09-01 | 2021-12-17 | 重庆中科云从科技有限公司 | Monitoring early warning method, equipment and computer storage medium |
| CN115548900A (en) * | 2022-10-21 | 2022-12-30 | 上海蓝箭电控设备南通有限公司 | Outdoor power distribution cabinet with high-temperature fire extinguishing function |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH11134579A (en) * | 1997-10-30 | 1999-05-21 | Nohmi Bosai Ltd | Flame detector |
| CN201117049Y (en) * | 2007-08-10 | 2008-09-17 | 山东省科学院自动化研究所 | Double-waveband image identification fire hazard detecting alarming system |
| CN101527073A (en) * | 2009-02-17 | 2009-09-09 | 丁国锋 | Fire detection system and method thereof |
| CN101632855A (en) * | 2009-05-27 | 2010-01-27 | 北京奥腾讯达科技有限公司 | Fire fighting system |
| CN101667322A (en) * | 2009-09-24 | 2010-03-10 | 丁国锋 | Disaster relieving system and method |
| CN101867790A (en) * | 2010-04-23 | 2010-10-20 | 刘文萍 | Millimeter-wave image analysis method, fire monitoring method and system |
| CN201876973U (en) * | 2010-11-10 | 2011-06-22 | 东华大学 | Image smoke detection system for rail transit station fire alarm |
| CN102236947A (en) * | 2010-04-29 | 2011-11-09 | 中国建筑科学研究院 | Flame monitoring method and system based on video camera |
| CN103400111A (en) * | 2013-07-10 | 2013-11-20 | 重庆大学 | Method for detecting fire accident on expressway or in tunnel based on video detection technology |
| CN105427507A (en) * | 2015-12-04 | 2016-03-23 | 广东欧珀移动通信有限公司 | Fire monitoring method and device |
| CN105575034A (en) * | 2014-10-14 | 2016-05-11 | 哈尔滨新光光电科技有限公司 | Image processing and analysis method of double-waveband forest fireproof intelligent monitoring software |
| CN105654127A (en) * | 2015-12-30 | 2016-06-08 | 成都数联铭品科技有限公司 | End-to-end-based picture character sequence continuous recognition method |
| CN205541250U (en) * | 2016-04-15 | 2016-08-31 | 林勇河 | Intelligence fire control control system |
-
2016
- 2016-10-17 CN CN201610902204.3A patent/CN106530569A/en active Pending
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH11134579A (en) * | 1997-10-30 | 1999-05-21 | Nohmi Bosai Ltd | Flame detector |
| CN201117049Y (en) * | 2007-08-10 | 2008-09-17 | 山东省科学院自动化研究所 | Double-waveband image identification fire hazard detecting alarming system |
| CN101527073A (en) * | 2009-02-17 | 2009-09-09 | 丁国锋 | Fire detection system and method thereof |
| CN101632855A (en) * | 2009-05-27 | 2010-01-27 | 北京奥腾讯达科技有限公司 | Fire fighting system |
| CN101667322A (en) * | 2009-09-24 | 2010-03-10 | 丁国锋 | Disaster relieving system and method |
| CN101867790A (en) * | 2010-04-23 | 2010-10-20 | 刘文萍 | Millimeter-wave image analysis method, fire monitoring method and system |
| CN102236947A (en) * | 2010-04-29 | 2011-11-09 | 中国建筑科学研究院 | Flame monitoring method and system based on video camera |
| CN201876973U (en) * | 2010-11-10 | 2011-06-22 | 东华大学 | Image smoke detection system for rail transit station fire alarm |
| CN103400111A (en) * | 2013-07-10 | 2013-11-20 | 重庆大学 | Method for detecting fire accident on expressway or in tunnel based on video detection technology |
| CN105575034A (en) * | 2014-10-14 | 2016-05-11 | 哈尔滨新光光电科技有限公司 | Image processing and analysis method of double-waveband forest fireproof intelligent monitoring software |
| CN105427507A (en) * | 2015-12-04 | 2016-03-23 | 广东欧珀移动通信有限公司 | Fire monitoring method and device |
| CN105654127A (en) * | 2015-12-30 | 2016-06-08 | 成都数联铭品科技有限公司 | End-to-end-based picture character sequence continuous recognition method |
| CN205541250U (en) * | 2016-04-15 | 2016-08-31 | 林勇河 | Intelligence fire control control system |
Cited By (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107770244A (en) * | 2017-09-07 | 2018-03-06 | 深圳市盛路物联通讯技术有限公司 | Data transfer control method and Related product |
| CN109920214A (en) * | 2017-12-12 | 2019-06-21 | 杭州海康威视系统技术有限公司 | A kind of picture pick-up device and information cuing method |
| CN108447212A (en) * | 2018-03-22 | 2018-08-24 | 河南科技大学 | A kind of intelligent repository fire alarm system and its alarm method |
| CN108830305A (en) * | 2018-05-30 | 2018-11-16 | 西南交通大学 | A kind of real-time fire monitoring method of combination DCLRN network and optical flow method |
| CN109561281A (en) * | 2018-11-15 | 2019-04-02 | 陕西瑞海工程智慧数据科技有限公司 | Industrial equipment method for safety monitoring, device, control device and readable storage medium storing program for executing |
| EP3667632A1 (en) * | 2018-12-13 | 2020-06-17 | Robert Bosch GmbH | Method for identifying a person smoking |
| CN111753594B (en) * | 2019-03-29 | 2023-09-29 | 杭州海康威视数字技术股份有限公司 | Dangerous identification method, device and system |
| CN111753594A (en) * | 2019-03-29 | 2020-10-09 | 杭州海康威视数字技术股份有限公司 | Danger identification method, device and system |
| CN110051954A (en) * | 2019-04-19 | 2019-07-26 | 辽宁科技大学 | A kind of split type Super High full-automatic fire-extinguishing machine people and its control method |
| CN110379121A (en) * | 2019-05-07 | 2019-10-25 | 深圳市泰和安科技有限公司 | A kind of hazard alarm method and server |
| CN110115819A (en) * | 2019-05-14 | 2019-08-13 | 上海荷福人工智能科技（集团）有限公司 | Fire water monitor orientation early warning and extinguishing method, storage medium and fire water monitor based on artificial intelligence |
| CN110102002A (en) * | 2019-05-15 | 2019-08-09 | 上海荷福人工智能科技（集团）有限公司 | The early warning of fire-fighting robot based on artificial intelligence and fire extinguishing system and method |
| CN110775930A (en) * | 2019-10-18 | 2020-02-11 | 北京地平线机器人技术研发有限公司 | Security protection device, method, computer-readable storage medium, and electronic device |
| CN110775930B (en) * | 2019-10-18 | 2021-07-06 | 北京地平线机器人技术研发有限公司 | Security protection device, method, computer-readable storage medium, and electronic device |
| CN111767771A (en) * | 2019-11-06 | 2020-10-13 | 母宗军 | Target site identification system and method |
| CN111105582A (en) * | 2019-11-12 | 2020-05-05 | 西安天和防务技术股份有限公司 | Forest fire prevention monitoring method and system, computer equipment and readable storage medium |
| CN111105582B (en) * | 2019-11-12 | 2021-08-06 | 西安天和防务技术股份有限公司 | Forest fire prevention monitoring method and system, computer equipment and readable storage medium |
| CN111583567A (en) * | 2020-06-01 | 2020-08-25 | 江西憶源多媒体科技有限公司 | Forest fire prevention early warning method and device |
| CN112418004A (en) * | 2020-11-05 | 2021-02-26 | 烟台艾睿光电科技有限公司 | Fire safety detection method and related components |
| CN113392710A (en) * | 2021-05-19 | 2021-09-14 | 上海可深信息科技有限公司 | Intelligent recognition method and system for smoking behavior |
| CN113807216A (en) * | 2021-09-01 | 2021-12-17 | 重庆中科云从科技有限公司 | Monitoring early warning method, equipment and computer storage medium |
| CN113469156A (en) * | 2021-09-03 | 2021-10-01 | 南通海腾铜业有限公司 | Artificial intelligence-based safety risk early warning assessment method, device and equipment |
| CN113469156B (en) * | 2021-09-03 | 2021-11-26 | 南通海腾铜业有限公司 | Artificial intelligence-based safety risk early warning assessment method, device and equipment |
| CN115548900A (en) * | 2022-10-21 | 2022-12-30 | 上海蓝箭电控设备南通有限公司 | Outdoor power distribution cabinet with high-temperature fire extinguishing function |
| CN115548900B (en) * | 2022-10-21 | 2023-12-22 | 上海蓝箭电控设备南通有限公司 | Outdoor switch board with high temperature function of putting out a fire |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106530569A (en) | Method and device for fire monitoring | |
| US20210056981A1 (en) | Systems and methods for managing an emergency situation | |
| US9704377B2 (en) | Systems and methods for managing an emergency situation | |
| US8606316B2 (en) | Portable blind aid device | |
| CN104822042B (en) | A kind of pedestrains safety detection method and device based on camera | |
| CN106127141A (en) | Warning message generates method and device | |
| US10616396B2 (en) | Danger detection system | |
| TW202133034A (en) | Violation detection method, electronic device and computer readable storage medium | |
| US11133021B2 (en) | Minimizing gunshot detection false positives | |
| US20150347079A1 (en) | Electronic device for determining when an officer is in a foot pursuit, a fight, has been incapacitated, or shots have been fired | |
| CN105701952B (en) | The method and device of air abnormality alarming | |
| US20200327347A1 (en) | Person monitoring system and person monitoring method | |
| CN104349133A (en) | Intelligent monitoring device and method | |
| US8955421B1 (en) | Methods and systems for detecting a gun and/or bullet within one's vicinity via an electronic device | |
| KR101602753B1 (en) | emergency call system using voice | |
| CN104954543A (en) | Automatic alarm method and device and mobile terminal | |
| KR102152123B1 (en) | Video Processing System of Intelligent Mobile Device with Monitoring Application to Detect and Control Commercial Fire | |
| US11450098B2 (en) | Firearm detection system and method | |
| CN110493474A (en) | A kind of data processing method, device and electronic equipment | |
| CN106375728B (en) | Monitoring method and device | |
| CN110636444A (en) | Earthquake early warning method, earthquake early warning device and storage medium | |
| US11093757B2 (en) | Firearm detection system and method | |
| CN106412227A (en) | Anti-lost alarm method and device and device for anti-lost alarm | |
| CN202221609U (en) | Networking alarm device based on audio identification | |
| CN111353454A (en) | Data processing method and device and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20181107 Address after: 100085 Huarun Qingcai Street 68, Haidian District, Beijing, two stage, 9 floor, 01 rooms. Applicant after: BEIJING XIAOMI MOBILE SOFTWARE Co.,Ltd. Applicant after: Anhui Huami Information Technology Co.,Ltd. Address before: 100085 Huarun Qingcai Street 68, Haidian District, Beijing, two stage, 9 floor, 01 rooms. Applicant before: BEIJING XIAOMI MOBILE SOFTWARE Co.,Ltd. Applicant before: XI'AN HAIDAO INFORMATION TECHNOLOGY CO., LTD. |
|
| TA01 | Transfer of patent application right | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170322 |
|
| RJ01 | Rejection of invention patent application after publication |