CN104794738A - Method and device for detecting smoke and fire - Google Patents
Method and device for detecting smoke and fire Download PDFInfo
- Publication number
- CN104794738A CN104794738A CN201510141324.1A CN201510141324A CN104794738A CN 104794738 A CN104794738 A CN 104794738A CN 201510141324 A CN201510141324 A CN 201510141324A CN 104794738 A CN104794738 A CN 104794738A
- Authority
- CN
- China
- Prior art keywords
- image
- interest
- area
- frame
- foreground image
- 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
Abstract
An embodiment of the invention provides a method and a device for detecting smoke and fire. The smoke and fire detection accuracy can be improved in severe environments by the aid of the method and the device. The method includes carrying out preprocessing operation on each image in video; extracting and processing foreground of each preprocessed image to obtain a foreground image; marking suspected smoke regions in each foreground image as regions of interest; judging whether frequencies of occurrence of various characteristic values in all the regions of interest are within first frequency ranges or not by the aid of preset characteristic statistic decisions, and giving an alarm if the frequencies of the occurrence of the various characteristic values in all the regions of interest are within the first frequency ranges. The first frequency ranges correspond to the characteristic values. The method and the device have the advantage that the smoke and fire detection accuracy can be improved in the severe environments by the aid of the method and the device.
Description
Technical field
The present invention relates to communication electronics field, particularly relate to a kind of firework detecting method and device.
Background technology
Fire is one of modal major disaster, is a kind of disaster that out of control burning causes on space-time, and the moment all can form serious threat to the safety of social economy, natural, ecological and personal property.In order to prevention fire failure as early as possible burst and spread, people have utilized multiple sensors to develop numerous pyrotechnics detection system.Conventional detecting device is mainly for smog, heat and radiation 3 class object, and what majority adopted is all the point sensor of sampling for physics such as particle, temperature, air humidity, transparency, smog, also has traditional ultraviolet or infrared spectrometry instrument.But all there is limitation in these Proximity Sensors and optoelectronic device in operation easier, reliability, detection efficiency, place practicality, cost and versatility etc.Therefore the formation one that grows up gradually of video fire prevention inspection cigarette technology compares the technique direction in forward position, to solve the shortcoming of traditional sensing solutions.
And the present inventor finds, in current video pyrotechnics detection technique, there is following defect:
The first, be utilize existing motion principal direction to accumulate scheme in prior art, the method only uses the change information of two two field pictures to add up, and when the slow cigarette of distant view, adjacent two frame figure change often less, and movable information is not obvious, and therefore this kind of scheme has certain limitation in this respect.
The second, existing motion accumulation scheme is utilized in prior art, because only consider the frequent degree of the motion in certain space of picture, so easily cause the wrong report problem of moving and causing of hovering because of object.
Three, motion accumulation scheme of the prior art utilizes the binary map accumulation strategy of time sliding window, easily produces larger memory consumption.
Summary of the invention
The embodiment of the present invention provides a kind of firework detecting method and device, can dislike the accuracy improving pyrotechnics detection under environment slightly.
A kind of firework detecting method, described method comprises:
Pretreatment operation is carried out to each two field picture in video;
Foreground extraction process is carried out to each two field picture pretreated and obtains foreground image;
Doubtful smoke region in every frame foreground image is labeled as area-of-interest;
Utilize the characteristic statistics decision-making that pre-sets, judge number of times that the various eigenwerts in all area-of-interests occur whether in the first numbers range corresponding to this eigenwert, if, then report to the police.
A kind of fire detection device, described device comprises:
Pretreatment module, for carrying out pretreatment operation to each two field picture in video;
Perspective process module, obtains foreground image for carrying out foreground extraction process to each two field picture pretreated;
Mark module, for being labeled as area-of-interest by the doubtful smoke region in every frame foreground image;
Judge module, for utilizing the characteristic statistics decision-making pre-set, judges number of times that the various eigenwerts in all area-of-interests occur whether in the first numbers range corresponding to this eigenwert, if, then report to the police.
The characteristic statistics decision-making that the technical scheme that the embodiment of the present invention provides proposes, the program is different from prior art completely, can under environment compares rugged environment, and the correct smog that detects is reported to the police.And unlike the scheme utilizing motion tracking technology to coordinate resemblance, easily in the situation lose objects of slow cigarette, or to the irregular situation of motion profile, the non-rigid object activity such as Ru Shu, cloud, people, produces irritated wrong report.The technical scheme that the embodiment of the present invention provides is not only add up the change information of two two field pictures, therefore also can carry out pyrotechnics when the slow cigarette of distant view very accurately and detect judgement, avoids not clear dominant the brought drawback of frame difference; The interim prospect accumulation step that the embodiment of the present invention provides, does not need the foreground image stored in the time window of continuous k frame, therefore avoids the problem that memory consumption is large; Interference cancellation operation that what the embodiment of the present invention provided hover, can eliminate the false alarm problem that some artificial or other objects produce because of motion of hovering effectively.
Accompanying drawing explanation
In order to be illustrated more clearly in the present invention or technical scheme of the prior art, be briefly described to the accompanying drawing used required in embodiment or description of the prior art below.
The firework detecting method process flow diagram that Fig. 1 provides for the embodiment of the present invention;
The first firework detecting method process flow diagram that Fig. 2 provides for the embodiment of the present invention;
The fire detection device structural drawing that Fig. 3 provides for the embodiment of the present invention.
Embodiment
In order to make those skilled in the art person understand technical scheme in the embodiment of the present invention better, and enable above-mentioned purpose of the present invention, feature and advantage become apparent more, below in conjunction with accompanying drawing, technical scheme in the present invention is described in further detail.
The characteristic statistics decision-making that the technical scheme that the embodiment of the present invention provides proposes, the program is different from prior art completely, can under environment compares rugged environment, and the correct smog that detects is reported to the police.And unlike the scheme utilizing motion tracking technology to coordinate resemblance, easily in the situation lose objects of slow cigarette, or to the irregular situation of motion profile, the non-rigid object activity such as Ru Shu, cloud, people, produces irritated wrong report.The technical scheme that the embodiment of the present invention provides is not only add up the change information of two two field pictures, therefore also can carry out pyrotechnics when the slow cigarette of distant view very accurately and detect judgement, avoids not clear dominant the brought drawback of frame difference; The interim prospect accumulation step that the embodiment of the present invention provides, does not need the foreground image stored in the time window of continuous k frame, therefore avoids the problem that memory consumption is large; Interference cancellation operation that what the embodiment of the present invention provided hover, can eliminate the false alarm problem that some artificial or other objects produce because of motion of hovering effectively.As shown in Figure 1, concrete steps comprise:
Step 11, carries out pretreatment operation to each two field picture in video;
Step 12, carries out foreground extraction process to each two field picture pretreated and obtains foreground image;
Step 13, is labeled as area-of-interest by the doubtful smoke region in every frame foreground image;
Step 14, utilizes the characteristic statistics decision-making that pre-sets, judges number of times that the various eigenwerts in all area-of-interests occur whether in the first numbers range corresponding to this eigenwert, if, then report to the police.
Concrete, described pretreatment operation comprises the context update modeling process operation according to pre-setting, and determines background image; Describedly foreground extraction process is carried out to each two field picture pretreated obtain foreground image and comprise:
Current frame image and background image are done pixel value difference computing, and binary conversion treatment is carried out to operation result obtains foreground image.
Concrete, described obtain foreground image after and before the doubtful smoke region in every frame foreground image is labeled as area-of-interest, comprise further:
A frame foreground image is often obtained within the acquisition cycle, just binary conversion treatment is carried out to it, then the foreground image after the foreground image after this binary conversion treatment and former frame binary conversion treatment is accumulated, when the cycle of acquisition arrives, interim prospect accumulation image can be got; Describedly doubtful smoke region in every frame foreground image be labeled as area-of-interest comprise:
Area-of-interest is labeled as to the doubtful smoke region in described interim prospect accumulation image.
Concrete, the doubtful smoke region in every frame foreground image is labeled as area-of-interest comprises described after interim prospect accumulation image result:
Interim prospect accumulation image is hovered interference cancellation operation; Describedly doubtful smoke region in every frame foreground image be labeled as area-of-interest be:
Doubtful smoke region in interim prospect accumulation image after completing interference cancellation operation of hovering is labeled as area-of-interest.
Concrete, described pretreatment operation also comprises carries out pixel value difference calculating by current frame image and previous frame image, then contrasts with the empirical value pre-set and obtain the poor binary map of adjacent frame; Described interference cancellation operation of hovering comprises:
Using the pixel flipchart of adjacent frame difference binary map as mask, the interference of hovering in elimination phase prospect accumulation image result.
Concrete, if the number of times that the various eigenwerts in a certain area-of-interest occur is in the second numbers range, then remove the mark of area-of-interest.
Concrete, described area-of-interest at the appointed time interior not confirmed is smog, then remove the mark of area-of-interest.
Be introduced with specific embodiment below:
Embodiment one:
The embodiment of the present invention provides the first firework detecting method, after the method can read the video of camera Real-time Collection or the video file preserved, analyze whether there is fire, judge whether to report to the police according to analysis result, as shown in Figure 2, detailed process is as follows:
Step 21, intercepts each two field picture, and carries out pretreatment operation to every two field picture from the video flowing obtained;
The application that the data obtained after pretreatment operation in this step will be used in follow-up flow process;
In this step pretreatment operation mainly comprise to every two field picture carry out dark primary model treatment, the process of Sobel gradient image, adjacent frame difference process, context update modeling process;
Dark primary model treatment refers to that the minimum value in each pixel using image in RGB tri-passages takes out the gray-scale value as this pixel, form dark primary image, carry out a mini-value filtering process again, after then binary conversion treatment being done to this width dark primary image OTSU algorithm, just obtain dark primary model;
The process of Sobel gradient image refers to carries out Sobel filtering process to former figure, obtains the Soebl gradient image that namely gradient magnitude corresponding to each pixel obtain this image;
The process of adjacent frame difference refers to carries out pixel value difference calculating by current frame image and previous frame image, then with the experience gray-scale value pre-set [20,30], contrast obtains the poor binary map of adjacent frame of this current frame image;
Context update modeling process refers to and utilizes runtime context update algorithm, goes out non-athletic scenery in present image, determine background image with this according to historical frames image statistics.
Step 22, carries out foreground extraction process to each two field picture pretreated and obtains foreground image, and using foreground image as image to be operated;
In this step, current frame image and background image are done pixel value difference computing, and binary conversion treatment is carried out to operation result obtain foreground image;
Step 23, is labeled as area-of-interest by the doubtful smoke region in every frame foreground image;
Step 24, utilizes the characteristic statistics decision-making that pre-sets, judges number of times that the various eigenwerts in all area-of-interests occur whether in the first numbers range corresponding to this eigenwert, if, then report to the police.
Concrete characteristic statistics decision-making is:
By in foreground image positive result be labeled as area-of-interest, be labeled as ROI region by doubtful smoke region;
ROI region will carry the eigenwert that some are marked as smog, and certain hour of surviving;
Eigenwert comprises: adjacent frame similar features value, spatial similarity eigenwert, Graded Density eigenwert and color model eigenwert;
Adjacent frame similar features value, refers to that the pixel value sum of each Color Channel in the identical ROI region of a few frame in front and back is substantially constant;
Spatial similarity eigenwert, refer to that in the ROI region containing smog, pixel value size is more stable, the variance of regional area is less;
Graded Density eigenwert, refers to the Sobel gradient magnitude sum in identical ROI region and ROI area ratio, and the grain that this eigenwert expresses smog is poor, therefore requires that Graded Density is little, and keeps stable within a period of time;
Color model eigenwert, refers to the area meeting smoke model in dark primary model.
In this step, the first numbers range is [15,50] frame, when the value of all eigenwerts is all within the scope of this, then ROI region is defined as smoke region, reports to the police.
Embodiment two:
The embodiment of the present invention two provides the second firework detecting method, the difference of embodiment two and embodiment one is that embodiment two with the addition of interim prospect and accumulates step between step 22 and step 23, the scheme that the present embodiment two proposes, can solve the problem that memory consumption that in prior art, prospect of the application accumulation algorithm brings is large;
Also prospect accumulation step can be comprised in current existing pyrotechnics detection scheme, its process be by detected by motion the foreground image empirical value two-value of continuous k frame obtained turn to non-zero then 1 image, then that this k is secondary non-zero, the imaging importing of 1 forms the prospect accumulation image that a pixel maximal value is k.Whether prospect accumulation image can a certain region of Description Image have object moving always.Because prior art is that the continuous foreground image accumulation of k frame in a time slide window obtains, therefore always need the foreground image stored in the time window of continuous k frame when algorithm realization, therefore memory consumption is large.But fire hazard aerosol fog is continuously process, until burning terminates just stopping, so the storage of time window is extremely wasted.
The interim foreground image accumulation algorithm that the present invention proposes is then the storage eliminating window during this period of time, adopt the strategy of periodic accumulation, namely several foreground images of k frame time window are kept in the temporary result that result becomes the interim prospect accumulation image of 1 frame, its process is: be in the time of k frame an acquisition cycle, often obtain a frame foreground image, just binary conversion treatment is carried out to it, then the foreground image after the foreground image after this binary conversion treatment and former frame binary conversion treatment is accumulated, it is final when k+1 frame (both the having obtained end cycle frame) moment arrives, temporary interim prospect accumulation image can be read out, and reset temporary for this result, prepare the result storing the next acquisition cycle.
Then in step 23, be then that the doubtful smoke region in interim prospect accumulation image is labeled as area-of-interest.
Embodiment three:
The embodiment of the present invention three provides the third firework detecting method, and embodiment three with the addition of interference cancellation operation of hovering on the basis of embodiment two, and achieve the operation of wrong report of moving and causing of hovering eliminated due to object, detailed process is:
Get interim prospect accumulation image in embodiment two after, hover to disturb to interim prospect accumulation image and eliminate, both the adjacent frame difference binary map obtained in step 21 was reversed, and using the adjacent frame difference binary map after reversion as mask, to complete the interference cancellation operation of hovering to foreground image;
Then in step 23, be then that the doubtful smoke region in the every frame foreground image after completing interference cancellation operation of hovering is labeled as area-of-interest.
Concrete, embodiment two and the interim prospect provided in embodiment three accumulate step with hover interference cancellation operation can different time appear in a workflow, such as embodiment one combines with the technical scheme in embodiment two, or embodiment one combines with the technical scheme of embodiment three, or embodiment one combines with the technical scheme in embodiment two and embodiment three; When embodiment one combines with the technical scheme in embodiment two and embodiment three, interim prospect accumulation step can sequentially be carried out in no particular order with interference cancellation operation of hovering, but preferably scheme is that advanced row interim prospect accumulation step carries out hovering interference cancellation operation again.
In above-described embodiment, if the number of times that the various eigenwerts in a certain area-of-interest occur is in the second numbers range [20,80] frame, not confirmed is smog, is not also identified into interference, then remove the mark of area-of-interest.
As shown in Figure 3, the embodiment of the present invention provides a kind of fire detection device, and this device comprises:
Pretreatment module, for carrying out pretreatment operation to each two field picture in video;
Perspective process module, obtains foreground image for carrying out foreground extraction process to each two field picture pretreated;
Mark module, for being labeled as area-of-interest by the doubtful smoke region in every frame foreground image;
Judge module, for utilizing the characteristic statistics decision-making pre-set, judges number of times that the various eigenwerts in all area-of-interests occur whether in the first numbers range corresponding to this eigenwert, if, then report to the police.
This device also comprises:
Accumulation module, for often obtaining a frame foreground image within the acquisition cycle, just binary conversion treatment is carried out to it, then the foreground image after the foreground image after this binary conversion treatment and former frame binary conversion treatment is accumulated, when the cycle of acquisition arrives, interim prospect accumulation image can be got;
Described mark module, specifically for being labeled as area-of-interest to the doubtful smoke region in described interim prospect accumulation image.
This device also comprises:
To hover interference cancellation module, for interference cancellation operation of hovering to interim prospect accumulation image;
Described mark module, specifically for being labeled as area-of-interest by the doubtful smoke region in the interim prospect accumulation image after completing interference cancellation operation of hovering.
Described interference cancellation module of hovering, specifically for using the pixel flipchart of adjacent frame difference binary map as mask, the interference of hovering in elimination phase prospect accumulation image result.
Described device also comprises:
Remove module, if the number of times occurred for the various eigenwerts in a certain area-of-interest is in the second numbers range, then remove the mark of area-of-interest.
In sum, beneficial effect:
The characteristic statistics decision-making that the technical scheme that the embodiment of the present invention provides proposes, the program is different from prior art completely, can under environment compares rugged environment, and the correct smog that detects is reported to the police.And unlike the scheme utilizing motion tracking technology to coordinate resemblance, easily in the situation lose objects of slow cigarette, or to the irregular situation of motion profile, the non-rigid object activity such as Ru Shu, cloud, people, produces irritated wrong report.
The technical scheme that the embodiment of the present invention provides is not only add up the change information of two two field pictures, therefore also can carry out pyrotechnics when the slow cigarette of distant view very accurately and detect judgement, avoids not clear dominant the brought drawback of frame difference;
The interim prospect accumulation step that the embodiment of the present invention provides, does not need the foreground image stored in the time window of continuous k frame, therefore avoids the problem that memory consumption is large;
Interference cancellation operation that what the embodiment of the present invention provided hover, can eliminate the false alarm problem that some artificial or other objects produce because of motion of hovering effectively.
The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, all any amendments done within the spirit and principles in the present invention, equivalent replacement and improvement etc., all should be included within protection scope of the present invention.
Claims (10)
1. a firework detecting method, is characterized in that, described method comprises:
Pretreatment operation is carried out to each two field picture in video;
Foreground extraction process is carried out to each two field picture pretreated and obtains foreground image;
Doubtful smoke region in every frame foreground image is labeled as area-of-interest;
Utilize the characteristic statistics decision-making that pre-sets, judge number of times that the various eigenwerts in all area-of-interests occur whether in the first numbers range corresponding to this eigenwert, if, then report to the police.
2. the method for claim 1, is characterized in that, described pretreatment operation comprises the context update modeling process operation according to pre-setting, and determines background image;
Describedly foreground extraction process is carried out to each two field picture pretreated obtain foreground image and comprise:
Current frame image and background image are done pixel value difference computing, and binary conversion treatment is carried out to operation result obtains foreground image.
3. the method for claim 1, is characterized in that, described obtain foreground image after and before the doubtful smoke region in every frame foreground image is labeled as area-of-interest, comprise further:
A frame foreground image is often obtained within the acquisition cycle, just binary conversion treatment is carried out to it, then the foreground image after the foreground image after this binary conversion treatment and former frame binary conversion treatment is accumulated, when the cycle of acquisition arrives, interim prospect accumulation image can be got;
Describedly doubtful smoke region in every frame foreground image be labeled as area-of-interest comprise:
Area-of-interest is labeled as to the doubtful smoke region in described interim prospect accumulation image.
4. method as claimed in claim 3, is characterized in that, the doubtful smoke region in every frame foreground image is labeled as area-of-interest comprises after interim prospect accumulation image result described:
Interim prospect accumulation image is hovered interference cancellation operation;
Describedly doubtful smoke region in every frame foreground image be labeled as area-of-interest be:
Doubtful smoke region in interim prospect accumulation image after completing interference cancellation operation of hovering is labeled as area-of-interest.
5. method as claimed in claim 4, it is characterized in that, described pretreatment operation also comprises carries out pixel value difference calculating by current frame image and previous frame image, then contrasts with the empirical value pre-set and obtain the poor binary map of adjacent frame;
Described interference cancellation operation of hovering comprises:
Using the pixel flipchart of adjacent frame difference binary map as mask, the interference of hovering in elimination phase prospect accumulation image result.
6. the method for claim 1, is characterized in that, if the number of times that the various eigenwerts in a certain area-of-interest occur is in the second numbers range, then removes the mark of area-of-interest.
7. the method for claim 1, is characterized in that, described area-of-interest at the appointed time interior not confirmed is smog, then remove the mark of area-of-interest.
8. a fire detection device, is characterized in that, described device comprises:
Pretreatment module, for carrying out pretreatment operation to each two field picture in video;
Perspective process module, obtains foreground image for carrying out foreground extraction process to each two field picture pretreated;
Mark module, for being labeled as area-of-interest by the doubtful smoke region in every frame foreground image;
Judge module, for utilizing the characteristic statistics decision-making pre-set, judges number of times that the various eigenwerts in all area-of-interests occur whether in the first numbers range corresponding to this eigenwert, if, then report to the police.
9. device as claimed in claim 8, it is characterized in that, this device also comprises:
Accumulation module, for often obtaining a frame foreground image within the acquisition cycle, just binary conversion treatment is carried out to it, then the foreground image after the foreground image after this binary conversion treatment and former frame binary conversion treatment is accumulated, when the cycle of acquisition arrives, interim prospect accumulation image can be got;
Described mark module, specifically for being labeled as area-of-interest to the doubtful smoke region in described interim prospect accumulation image.
10. method as claimed in claim 9, it is characterized in that, described device also comprises:
To hover interference cancellation module, for interference cancellation operation of hovering to interim prospect accumulation image;
Described mark module, specifically for being labeled as area-of-interest by the doubtful smoke region in the interim prospect accumulation image after completing interference cancellation operation of hovering.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510141324.1A CN104794738A (en) | 2015-03-27 | 2015-03-27 | Method and device for detecting smoke and fire |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510141324.1A CN104794738A (en) | 2015-03-27 | 2015-03-27 | Method and device for detecting smoke and fire |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104794738A true CN104794738A (en) | 2015-07-22 |
Family
ID=53559517
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510141324.1A Pending CN104794738A (en) | 2015-03-27 | 2015-03-27 | Method and device for detecting smoke and fire |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104794738A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112598712A (en) * | 2020-12-28 | 2021-04-02 | 浙江天衡五维电子科技有限公司 | Solar blind ultraviolet plume detection method and warning system |
| CN112861676A (en) * | 2021-01-28 | 2021-05-28 | 济南和普威视光电技术有限公司 | Smoke and fire identification marking method, system, terminal and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102163361A (en) * | 2011-05-16 | 2011-08-24 | 公安部沈阳消防研究所 | Image-type fire detection method based on cumulative prospect image |
| CN102163358A (en) * | 2011-04-11 | 2011-08-24 | 杭州电子科技大学 | Smoke/flame detection method based on video image analysis |
| CN102819735A (en) * | 2012-08-17 | 2012-12-12 | 深圳辉锐天眼科技有限公司 | Flame detection method based on video frame image |
| CN103440484A (en) * | 2013-09-12 | 2013-12-11 | 沈阳聚德视频技术有限公司 | Flame detection method adaptive to large outdoor space |
| CN103473788A (en) * | 2013-07-31 | 2013-12-25 | 中国电子科技集团公司第三十八研究所 | Indoor fire and flame detection method based on high-definition video images |
| CN103983574A (en) * | 2014-06-03 | 2014-08-13 | 上海安维尔信息科技有限公司 | Smoke detection method |
-
2015
- 2015-03-27 CN CN201510141324.1A patent/CN104794738A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102163358A (en) * | 2011-04-11 | 2011-08-24 | 杭州电子科技大学 | Smoke/flame detection method based on video image analysis |
| CN102163361A (en) * | 2011-05-16 | 2011-08-24 | 公安部沈阳消防研究所 | Image-type fire detection method based on cumulative prospect image |
| CN102819735A (en) * | 2012-08-17 | 2012-12-12 | 深圳辉锐天眼科技有限公司 | Flame detection method based on video frame image |
| CN103473788A (en) * | 2013-07-31 | 2013-12-25 | 中国电子科技集团公司第三十八研究所 | Indoor fire and flame detection method based on high-definition video images |
| CN103440484A (en) * | 2013-09-12 | 2013-12-11 | 沈阳聚德视频技术有限公司 | Flame detection method adaptive to large outdoor space |
| CN103983574A (en) * | 2014-06-03 | 2014-08-13 | 上海安维尔信息科技有限公司 | Smoke detection method |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112598712A (en) * | 2020-12-28 | 2021-04-02 | 浙江天衡五维电子科技有限公司 | Solar blind ultraviolet plume detection method and warning system |
| CN112861676A (en) * | 2021-01-28 | 2021-05-28 | 济南和普威视光电技术有限公司 | Smoke and fire identification marking method, system, terminal and storage medium |
| CN112861676B (en) * | 2021-01-28 | 2022-06-21 | 济南和普威视光电技术有限公司 | Smoke and fire identification marking method, system, terminal and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10366509B2 (en) | Setting different background model sensitivities by user defined regions and background filters | |
| EP2118864B1 (en) | Behavioral recognition system | |
| CN105940430B (en) | Personnel's method of counting and its device | |
| CN103150549B (en) | A kind of road tunnel fire detection method based on the early stage motion feature of smog | |
| CN109409256A (en) | A kind of forest rocket detection method based on 3D convolutional neural networks | |
| CN103729858B (en) | A kind of video monitoring system is left over the detection method of article | |
| CN111881853B (en) | Method and device for identifying abnormal behaviors in oversized bridge and tunnel | |
| CN104463869B (en) | A kind of video flame image composite identification method | |
| CN110264495A (en) | A kind of method for tracking target and device | |
| CN110874592A (en) | Forest fire smoke image detection method based on total bounded variation | |
| Lian et al. | A novel method on moving-objects detection based on background subtraction and three frames differencing | |
| CN102663778B (en) | A kind of method for tracking target based on multi-view point video and system | |
| CN106778488B (en) | Low-light (level) smog video detecting method based on image correlation | |
| CN106296708B (en) | Car tracing method and apparatus | |
| Malhi et al. | Vision based intelligent traffic management system | |
| CA2825506A1 (en) | Spectral scene simplification through background subtraction | |
| CN104794738A (en) | Method and device for detecting smoke and fire | |
| CN110852241A (en) | Small target detection method applied to nursing robot | |
| CN107729811B (en) | Night flame detection method based on scene modeling | |
| CN108961287A (en) | Intelligent commodity shelf triggering method, intelligent commodity shelf system, storage medium and electronic equipment | |
| CN104751591A (en) | Early fire smoke detection method based on semi transparent feature | |
| Muniruzzaman et al. | Deterministic algorithm for traffic detection in free-flow and congestion using video sensor | |
| CN105740819A (en) | Integer programming based crowd density estimation method | |
| CN110858392A (en) | Monitoring target positioning method based on fusion background model | |
| CN105718881B (en) | The zero illumination environment monitoring smoke dust method based on infrared video gray level image |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| EXSB | Decision made by sipo to initiate substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20150722 |