CN108985144A - A kind of high efficiency, low cost image fire automatic identifying method and device - Google Patents
A kind of high efficiency, low cost image fire automatic identifying method and device Download PDFInfo
- Publication number
- CN108985144A CN108985144A CN201810526100.6A CN201810526100A CN108985144A CN 108985144 A CN108985144 A CN 108985144A CN 201810526100 A CN201810526100 A CN 201810526100A CN 108985144 A CN108985144 A CN 108985144A
- Authority
- CN
- China
- Prior art keywords
- image
- fire source
- fire
- value
- coordinate
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 26
- 238000004458 analytical method Methods 0.000 claims description 36
- 238000001514 detection method Methods 0.000 claims description 19
- 108700041286 delta Proteins 0.000 claims description 18
- 238000012216 screening Methods 0.000 claims description 12
- 238000010191 image analysis Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000005070 sampling Methods 0.000 claims description 3
- 230000001419 dependent effect Effects 0.000 abstract 2
- 230000001737 promoting effect Effects 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 3
- 238000012423 maintenance Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002354 daily effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 description 1
- 239000010931 gold Substances 0.000 description 1
- 229910052737 gold Inorganic materials 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000001769 paralizing effect Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 239000000779 smoke Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
-
- 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
- G06V10/56—Extraction of image or video features relating to colour
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Fire-Detection Mechanisms (AREA)
Abstract
The present invention provides a kind of high efficiency, low cost image fire automatic identifying method and devices, it is different from dependent on correlation technique before and after sequential frame image, the present invention can analyze fire source by single-frame images, and the quality of image is not required, it can be analyzed in the case where resolution ratio is very low, image lowest resolution is up to 320 × 240 pixels, the present invention is laid particular emphasis on to fire source color gamut, fire source shape, fire source feature is analyzed, and low-resolution image can be analyzed, it ensure that the generality of its use scope, also imply that domestic almost all of video image can be used to be analyzed at present；It requires characteristic, single frames can direct analytical characteristics the low of image resolution ratio, also determine that the present invention greatly reduces hardware-dependent degree, its cost is also greatly lower than domestic all image fire identification products at present, its product cost will reduce by 60% or so than domestic similar product at present, have very strong applicability and promotional value.
Description
Technical Field
The invention belongs to the field of computer image processing and identification, and particularly relates to a high-efficiency low-cost image fire automatic identification method and device.
Background
At present, the domestic related image fire automatic identification system is applied for years, the fire identification rate is about 60 percent in total, the false alarm rate is about 20 percent, and the system is characterized in that a high-definition camera with higher cost and special hardware are adopted, so that the overall cost is higher, and common users are difficult to apply. At present, video systems with different qualities are installed in most units, and the quantity of the cameras is large, so that great contribution is made to the safety and stability of the society. However, such a video camera is characterized by having only an image acquisition function and lacking an image analysis function, and in the field of fire fighting, videos also play an important role in the identification process of fire, and the discovery of fire fighting is the most effective and direct method by using video images except fire sensors such as smoke, temperature and infrared.
At present, the overall level of the false alarm rate of all domestic fire alarm sensors is high, many false alarms occur almost every day, people are easy to generate paralytic psychology when the false alarms occur frequently, and the best time for fighting fire is easy to miss once a fire disaster occurs in an atmosphere that the false alarms frequently occur and are real and false and difficult to distinguish; whether a large amount of existing video image data can be utilized to carry out automatic fire analysis is an original intention for researching the problem, if the automatic fire analysis can be carried out on the common video data, the prevention and control level of the existing fire management can be greatly improved, the working efficiency of the fire management is also improved, and reliable technical support is provided for timely finding, timely early warning and timely rescue.
Disclosure of Invention
In order to utilize the common video data installed by a large number of users to carry out image fire automatic analysis so as to improve the detection, discovery and extinguishing the fire in the bud state as far as possible when the fire of a fire-fighting user unit occurs, the invention provides a high-efficiency low-cost image fire automatic identification method and a device, a fire source is analyzed through a single-frame image, the quality of the image is not required, the analysis can be carried out under the condition of low resolution, the minimum resolution of the image can reach 320 multiplied by 240 pixels, the emphasis is placed on the analysis of the color range, the shape and the characteristics of the fire source, and the image fire can be automatically identified with high efficiency and low cost.
In order to solve the technical problem, the invention adopts the following technical scheme:
an efficient low-cost image fire automatic identification method comprises the following steps:
sampling from a common video stream (fire monitoring) to obtain a single-frame image analysis sample;
performing fire source color feature analysis on a single-frame image sample by using a 24-bit RGB bitmap, and determining a plurality of suspected fire source color feature image ranges;
performing fire source shape characteristic analysis in a plurality of suspected fire source color characteristic image ranges by adopting a boundary detection algorithm, comparing and analyzing a fire source boundary pixel point serving as a center with gray difference values of left and right pixels to obtain a fire source boundary, and screening out a plurality of suspected fire source ranges;
counting the range number of the suspected fire sources in the single-frame image sample, analyzing and screening the shape characteristics of the suspected fire sources, finding the suspected fire sources at the same position of more than three continuous frames of images, and determining the fire sources.
The technical scheme is further limited as follows: the color range of the suspected fire source color characteristic image is as follows: setting, a first component Δ 1= R-G, a second component Δ 2= R-B, a third component Δ 3= G-B; then there is a change in the number of,
(1) determination of the yellow component:
delta 1 is more than or equal to 10, delta 2 is more than 70, and delta 3 is more than 80; or,
-25 ≤⊿1≤25 ，⊿2＞120，⊿3＞120；
(2) determination of the red component:
r is more than 100, R is more than G, G is more than or equal to B, delta 2 is more than 80, and delta 1 is more than 30; or,
R＞100，⊿2＞100，⊿1＞100；
(3) determination of the gray component:
-15 ≤⊿1≤15，-15≤⊿2≤15，-15 ≤⊿3≤15；
(4) determination of white component:
R＞150，B＞150，G＞150；
and when the yellow component, the red component and the gray component in the color range are positioned in the range, the images are determined as the suspected fire source color characteristic images, wherein the yellow component, the red component and the gray component are analyzed by the daytime images, and the white component and the gray component are analyzed by the night images.
The technical scheme is further limited as follows: and (3) analyzing the shape of the fire source by adopting a boundary detection algorithm, wherein the method comprises the following steps:
the gray value of a single frame image sample is converted, the gray value of the image adopts an average value algorithm, namely, the gray value calculation method of image pixels comprises the following steps: an image pixel gradation VALUE GRAY _ VALUE = (R + G + B)/3;
establishing a rectangular boundary detection model in a plurality of suspected fire source color characteristic image ranges, analyzing fire source boundaries by using the rectangular boundary detection model, acquiring fire source boundaries by using the rectangular boundary detection model when the gray difference value between the interior of a suspected fire source color characteristic image range buffer area and boundary pixels is more than 30, and performing recursion by using the characteristic algorithm to determine the range as the suspected fire source range.
The technical scheme is further limited as follows: in the fire source shape analysis, after obtaining a fire source boundary through a pixel gray level difference value, further performing boundary smoothness and boundary trend analysis, wherein the main method is to use angle change between two points (spanning 5 pixel points and 10 pixel points) to count the boundary trend and the smoothness, and if the smoothness and the boundary trend proportion are less than 50% of the fire source model boundary, preliminarily determining the fire source; and continuously analyzing the aspect ratio of more than 50% of the suspected fire source area, determining the suspected fire source with the aspect ratio less than 1 as a non-fire source, and eliminating.
The technical scheme is further limited as follows: in the fire source shape analysis, the internal smoothness analysis is carried out on the object for analyzing the fire source, and the analysis method is as follows: continuously selecting 6 pixels in the horizontal or vertical coordinate, respectively taking the gray difference, and defining as follows,
a value1 coordinate (X1, Y1), a value2 coordinate (X2, Y2), a value3 coordinate (X3, Y3), a value4 coordinate (X4, Y4), a value5 coordinate (X5, Y5), a value6 coordinate (X6, Y6), wherein X1= X2= X3= X4= X5= X6, or Y1= Y2= Y3= Y4= Y5= Y6;
taking the Y coordinate equal, i.e. the X coordinate calculation as an example:
Δ 1= value 1-value 2= difference in X coordinate;
Δ 2= value 1-value 3= difference in X coordinate;
Δ 3= value 1-value 5= difference in X coordinate;
Δ 4= value 1-value 6= difference in X coordinate;
Δ 5= value 2-value 5= difference in X coordinate;
Δ 6= value 3-value 6= difference in X coordinate;
then, the conditions for meeting the fire source smoothness requirement are as follows:
Δ 1, Δ 2, Δ 3, and Δ 4 are less than or equal to 30, or,
less than or equal to 30 delta 1, less than or equal to delta 5 and less than or equal to 30 delta 6, or,
less than or equal to 30 delta 2, less than or equal to 30 delta 3 and less than or equal to delta 4, or,
and delta 3, delta 4 and delta 6 are less than or equal to 30.
The technical scheme is defined as follows: the image samples are 320 x 240 low resolution images.
The invention discloses a high-efficiency low-cost image fire automatic identification device, which comprises:
the image acquisition device samples and obtains a single-frame image analysis sample from a common video stream;
the flame color characteristic analysis device is used for carrying out fire source color characteristic analysis on a single-frame image sample by utilizing a 24-bit RGB bitmap and determining a plurality of suspected fire source color characteristic image ranges;
the flame shape characteristic analysis device is used for carrying out fire source shape characteristic analysis in a plurality of suspected fire source color characteristic image ranges by adopting a boundary detection algorithm, comparing and analyzing the gray difference value of a central pixel point and left and right pixels to obtain a fire source boundary and screening out a plurality of suspected fire source ranges;
the flame determining device is used for counting the range number of the suspected fire sources in the single-frame image samples, analyzing and screening the shape characteristics of the suspected fire sources, finding the suspected fire sources at the same position of more than three continuous frames of images and determining the fire sources.
The invention has the beneficial effects that: the fire source is analyzed through the single-frame image, the quality of the image is not required, the image can be analyzed under the condition of low resolution, and the image fire can be automatically identified with high efficiency and low cost through analyzing the color range, the shape and the characteristics of the fire source.
Drawings
FIG. 1 is a diagram of a boundary model of a pixel point in an image sample according to the present invention.
Fig. 2 is a structural configuration diagram of an image fire automatic identification device of the present invention with high efficiency and low cost.
Fig. 3 is a functional block diagram of an application of the present invention.
Detailed Description
In order to facilitate understanding of the present invention, the following description will be further made with reference to fig. 1 for describing an efficient and low-cost image fire automatic identification method of the present invention:
sampling from a common video stream to obtain a single-frame image analysis sample;
performing fire source color feature analysis on a single-frame image sample by using a 24-bit RGB bitmap, and determining a plurality of suspected fire source color feature image ranges; the color range of the suspected fire source color characteristic image is as follows: setting, a first component Δ 1= R-G, a second component Δ 2= R-B, a third component Δ 3= G-B; then there is a change in the number of,
(1) determination of the yellow component:
delta 1 is more than or equal to 10, delta 2 is more than 70, and delta 3 is more than 80; or,
-25 ≤⊿1≤25 ，⊿2＞120，⊿3＞120；
(2) determination of the red component:
r is more than 100, R is more than G, G is more than or equal to B, delta 2 is more than 80, and delta 1 is more than 30; or,
R＞100，⊿2＞100，⊿1＞100；
(3) determination of the gray component:
-15 ≤⊿1≤15，-15≤⊿2≤15，-15≤⊿3≤15；
(4) determination of white component:
R＞150，B＞150，G＞150；
when the yellow component, the red component and the gray component in the color range are positioned in the range, the color range is determined as a suspected fire source color characteristic image, wherein the yellow component, the red component and the gray component are analyzed by the image in the daytime, and the white component and the gray component are analyzed by the image in the evening;
performing fire source shape characteristic analysis in a plurality of suspected fire source color characteristic image ranges by adopting a boundary detection algorithm, comparing and analyzing a fire source boundary pixel point serving as a center with gray difference values of left and right pixels to obtain a fire source boundary, and screening out a plurality of suspected fire source ranges;
and (3) analyzing the shape of the fire source by adopting a boundary detection algorithm, wherein the method comprises the following steps:
the gray value of a single frame image sample is converted, the gray value of the image adopts an average value algorithm, namely, the gray value calculation method of image pixels comprises the following steps: an image pixel gradation VALUE GRAY _ VALUE = (R + G + B)/3;
establishing a rectangular boundary detection model in a plurality of suspected fire source color characteristic image ranges, analyzing fire source boundaries by using the rectangular boundary detection model, acquiring fire source boundaries by using the rectangular boundary detection model when the gray difference value between the interior of a suspected fire source color characteristic image range buffer area and boundary pixels is more than 30, and performing recursion by using a characteristic algorithm to determine the fire source boundaries as the suspected fire source ranges;
counting the range number of the suspected fire sources in the single-frame image sample, analyzing and screening the shape characteristics of the suspected fire sources, finding the suspected fire sources at the same position of more than three continuous frames of images, and determining the fire sources.
In the fire source shape analysis, after obtaining a fire source boundary through a pixel gray level difference value, further performing boundary smoothness and boundary trend analysis, wherein the main method is to use angle change between two points (spanning 5 pixels and 10 pixels) to count the boundary trend and the smoothness, and if the smoothness and the boundary trend proportion are less than 50% of the fire source model boundary, the fire source is preliminarily determined; continuously analyzing the length-width ratio of more than 50% of the suspected fire source area, determining the suspected fire source with the length-width ratio less than 1 as a non-fire source, and removing the non-fire source;
in the fire source shape analysis, the internal smoothness analysis is carried out on the object for analyzing the fire source, and the analysis method comprises the following steps: continuously selecting 6 pixels in the horizontal or vertical coordinate, respectively taking the gray difference, and defining as follows,
a value1 coordinate (X1, Y1), a value2 coordinate (X2, Y2), a value3 coordinate (X3, Y3), a value4 coordinate (X4, Y4), a value5 coordinate (X5, Y5), a value6 coordinate (X6, Y6), wherein X1= X2= X3= X4= X5= X6, or Y1= Y2= Y3= Y4= Y5= Y6;
taking the Y coordinate equal, i.e. the X coordinate calculation as an example:
Δ 1= value 1-value 2= difference in X coordinate;
Δ 2= value 1-value 3= difference in X coordinate;
Δ 3= value 1-value 5= difference in X coordinate;
Δ 4= value 1-value 6= difference in X coordinate;
Δ 5= value 2-value 5= difference in X coordinate;
Δ 6= value 3-value 6= difference in X coordinate;
then, the conditions for meeting the fire source smoothness requirement are as follows:
Δ 1, Δ 2, Δ 3, and Δ 4 are less than or equal to 30, or,
less than or equal to 30 delta 1, less than or equal to delta 5 and less than or equal to 30 delta 6, or,
less than or equal to 30 delta 2, less than or equal to 30 delta 3 and less than or equal to delta 4, or,
and delta 3, delta 4 and delta 6 are less than or equal to 30.
The technical scheme is defined as follows: the image samples are 320 x 240 low resolution images.
Referring to fig. 2, in the present invention, an image fire automatic identification apparatus with high efficiency and low cost comprises,
the image acquisition device samples and obtains a single-frame image analysis sample from a common video stream;
the flame color characteristic analysis device is used for carrying out fire source color characteristic analysis on a single-frame image sample by utilizing a 24-bit RGB bitmap and determining a plurality of suspected fire source color characteristic image ranges;
the flame shape characteristic analysis device is used for carrying out fire source shape characteristic analysis in a plurality of suspected fire source color characteristic image ranges by adopting a boundary detection algorithm, comparing and analyzing the gray difference value of a central pixel point and left and right pixels to obtain a fire source boundary and screening out a plurality of suspected fire source ranges;
the flame determining device is used for counting the range number of the suspected fire sources in the single-frame image samples, analyzing and screening the shape characteristics of the suspected fire sources, finding the suspected fire sources at the same position of more than three continuous frames of images and determining the fire sources.
According to the invention, after the fire source in the image is judged, if the fire is found, the image data is sent to relevant personnel for processing at a first time, so that the purpose of automatically analyzing and eliminating hidden dangers before the fire is achieved and ensuring that rescue facilities are reliably available in daily life is achieved; the functions of timely discovering and timely and quickly evacuating and rescuing after the disaster are realized, and the gold three-minute fire-fighting concept after the disaster is embodied.
The work flow of the present invention is described below with reference to fig. 3:
1. acquiring video image data (of various types existing in a user) by a secondary development kit provided by a national standard Internet video standard protocol or a video manufacturer;
2. data are transmitted into an image fire automatic identification server system which utilizes the image fire automatic identification system, and the server system extracts some image frames for automatic analysis after acquiring image resources;
3. the server system sends the image resource to a mobile terminal of an operator on duty or a manager of a fire-fighting user unit at the first time after a fire source (fire) is found;
4. after receiving the information, the fire-fighting user unit manager or the attendant mobile terminal dispatches the information to maintenance and repair personnel; and submitting the result to the server system after the maintenance personnel finish working.
Claims (7)
1. An efficient low-cost image fire automatic identification method comprises the following steps:
sampling from a common video stream to obtain a single-frame image analysis sample;
performing fire source color feature analysis on a single-frame image sample by using a 24-bit RGB bitmap, and determining a plurality of suspected fire source color feature image ranges;
performing fire source shape characteristic analysis in a plurality of suspected fire source color characteristic image ranges by adopting a boundary detection algorithm, comparing and analyzing a fire source boundary pixel point serving as a center with gray difference values of left and right pixels to obtain a fire source boundary, and screening out a plurality of suspected fire source ranges;
counting the range number of the suspected fire sources in the single-frame image sample, analyzing and screening the shape characteristics of the suspected fire sources, finding the suspected fire sources at the same position of more than three continuous frames of images, and determining the fire sources.
2. The method for automatically identifying an image fire according to claim 1, wherein the method comprises the following steps: the color range of the suspected fire source color characteristic image is as follows: setting, a first component Δ 1= R-G, a second component Δ 2= R-B, a third component Δ 3= G-B; then there is a change in the number of,
(1) determination of the yellow component:
delta 1 is more than or equal to 10, delta 2 is more than 70, and delta 3 is more than 80; or,
-25 ≤⊿1≤ 25 ，⊿2＞120，⊿3＞120；
(2) determination of the red component:
r is more than 100, R is more than G, G is more than or equal to B, delta 2 is more than 80, and delta 1 is more than 30; or,
R＞100，⊿2＞100，⊿1＞100；
(3) determination of the gray component:
-15 ≤⊿1≤15，-15≤⊿2≤15，-15 ≤⊿3≤15；
(4) determination of white component:
R＞150，B＞150，G＞150；
and when the yellow component, the red component and the gray component in the color range are positioned in the range, the images are determined as the suspected fire source color characteristic images, wherein the yellow component, the red component and the gray component are analyzed by the daytime images, and the white component and the gray component are analyzed by the night images.
3. The method for automatically identifying an image fire according to claim 1, wherein the method comprises the following steps: and (3) analyzing the shape of the fire source by adopting a boundary detection algorithm, wherein the method comprises the following steps:
the gray value of a single frame image sample is converted, the gray value of the image adopts an average value algorithm, namely, the gray value calculation method of image pixels comprises the following steps: an image pixel gradation VALUE GRAY _ VALUE = (R + G + B)/3;
establishing a rectangular boundary detection model in a plurality of suspected fire source color characteristic image ranges, analyzing fire source boundaries by using the rectangular boundary detection model, acquiring fire source boundaries by using the rectangular boundary detection model when the gray difference value between the interior of a suspected fire source color characteristic image range buffer area and boundary pixels is more than 30, and performing recursion by using the characteristic algorithm to determine the range as the suspected fire source range.
4. An efficient low-cost image fire automatic identification method as claimed in claim 3, characterized in that: after acquiring the fire source boundary through the pixel gray level difference, further performing boundary smoothness and boundary trend analysis, wherein the main method is to use the angle change between two points (spanning 5 pixels and 10 pixels) to count the boundary trend and smoothness, and if the smoothness and the boundary trend proportion are less than 50% of the fire source model boundary, preliminarily determining the fire source; and continuously analyzing the aspect ratio of more than 50% of the suspected fire source area, determining the suspected fire source with the aspect ratio less than 1 as a non-fire source, and eliminating.
5. The method for automatically identifying an image fire according to claim 4, wherein the method comprises the following steps: the internal smoothness analysis was performed on the object from which the fire source was analyzed as follows:
continuously selecting 6 pixels in the horizontal or vertical coordinate, respectively taking the gray difference, and defining as follows,
a value1 coordinate (X1, Y1), a value2 coordinate (X2, Y2), a value3 coordinate (X3, Y3), a value4 coordinate (X4, Y4), a value5 coordinate (X5, Y5), a value6 coordinate (X6, Y6), wherein X1= X2= X3= X4= X5= X6, or Y1= Y2= Y3= Y4= Y5= Y6;
taking the Y coordinate equal, i.e. the X coordinate calculation as an example:
Δ 1= value 1-value 2= difference in X coordinate;
Δ 2= value 1-value 3= difference in X coordinate;
Δ 3= value 1-value 5= difference in X coordinate;
Δ 4= value 1-value 6= difference in X coordinate;
Δ 5= value 2-value 5= difference in X coordinate;
Δ 6= value 3-value 6= difference in X coordinate;
then, the conditions for meeting the fire source smoothness requirement are as follows:
Δ 1, Δ 2, Δ 3, and Δ 4 are less than or equal to 30, or,
less than or equal to 30 delta 1, less than or equal to delta 5 and less than or equal to 30 delta 6, or,
less than or equal to 30 delta 2, less than or equal to 30 delta 3 and less than or equal to delta 4, or,
and delta 3, delta 4 and delta 6 are less than or equal to 30.
6. The method for automatically identifying an image fire according to claim 1, wherein the method comprises the following steps:
the single frame image sample is a 320 x 240 low resolution image.
7. An efficient low-cost image fire automatic identification device, comprising:
the image acquisition device samples and obtains a single-frame image analysis sample from a common video stream;
the flame color characteristic analysis device is used for carrying out fire source color characteristic analysis on a single-frame image sample by utilizing a 24-bit RGB bitmap and determining a plurality of suspected fire source color characteristic image ranges;
the flame shape characteristic analysis device is used for carrying out fire source shape characteristic analysis in a plurality of suspected fire source color characteristic image ranges by adopting a boundary detection algorithm, comparing and analyzing the gray difference value of a central pixel point and left and right pixels to obtain a fire source boundary and screening out a plurality of suspected fire source ranges;
the flame determining device is used for counting the range number of the suspected fire sources in the single-frame image samples, analyzing and screening the shape characteristics of the suspected fire sources, finding the suspected fire sources at the same position of more than three continuous frames of images and determining the fire sources.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810526100.6A CN108985144B (en) | 2018-05-29 | 2018-05-29 | Efficient low-cost image fire automatic identification method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810526100.6A CN108985144B (en) | 2018-05-29 | 2018-05-29 | Efficient low-cost image fire automatic identification method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108985144A true CN108985144A (en) | 2018-12-11 |
| CN108985144B CN108985144B (en) | 2022-04-12 |
Family
ID=64542210
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810526100.6A Active CN108985144B (en) | 2018-05-29 | 2018-05-29 | Efficient low-cost image fire automatic identification method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108985144B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112071006A (en) * | 2020-09-11 | 2020-12-11 | 湖北德强电子科技有限公司 | High-efficiency low-resolution image area intrusion recognition algorithm and device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102034110A (en) * | 2010-12-09 | 2011-04-27 | 湘潭乐星电气有限公司 | Detection method of flame |
| 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 |
| CN103761529A (en) * | 2013-12-31 | 2014-04-30 | 北京大学 | Open fire detection method and system based on multicolor models and rectangular features |
| US20150030203A1 (en) * | 2013-07-23 | 2015-01-29 | Samsung Techwin Co., Ltd. | Method and apparatus for detecting smoke from image |
| CN108765833A (en) * | 2018-04-27 | 2018-11-06 | 天津大学 | Based on the incipient fire detection algorithm for improving mixed Gaussian and machine learning |
-
2018
- 2018-05-29 CN CN201810526100.6A patent/CN108985144B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102034110A (en) * | 2010-12-09 | 2011-04-27 | 湘潭乐星电气有限公司 | Detection method of flame |
| US20150030203A1 (en) * | 2013-07-23 | 2015-01-29 | Samsung Techwin Co., Ltd. | Method and apparatus for detecting smoke from 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 |
| CN103761529A (en) * | 2013-12-31 | 2014-04-30 | 北京大学 | Open fire detection method and system based on multicolor models and rectangular features |
| CN108765833A (en) * | 2018-04-27 | 2018-11-06 | 天津大学 | Based on the incipient fire detection algorithm for improving mixed Gaussian and machine learning |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112071006A (en) * | 2020-09-11 | 2020-12-11 | 湖北德强电子科技有限公司 | High-efficiency low-resolution image area intrusion recognition algorithm and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108985144B (en) | 2022-04-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7859419B2 (en) | Smoke detecting method and device | |
| KR100948128B1 (en) | Smoke detecting method and device | |
| US7868772B2 (en) | Flame detecting method and device | |
| KR101168760B1 (en) | Flame detecting method and device | |
| CN101846576B (en) | Video-based liquid leakage analyzing and alarming system | |
| CN104581081B (en) | Passenger flow analysing method based on video information | |
| KR100922784B1 (en) | Image base fire sensing method and system of crime prevention and disaster prevention applying method thereof | |
| CN108389359B (en) | Deep learning-based urban fire alarm method | |
| CN108734913A (en) | Forest fire patrol alarm system based on unmanned plane image recognition and method | |
| EP2000952B1 (en) | Smoke detecting method and device | |
| KR101204259B1 (en) | A method for detecting fire or smoke | |
| CN111476964A (en) | Remote forest fire prevention monitoring system and method | |
| EP2000998B1 (en) | Flame detecting method and device | |
| CN101315326B (en) | Smog detecting method and apparatus | |
| CN101316371A (en) | Flame detecting method and device | |
| CN114917519B (en) | Building intelligent fire control system | |
| CN108985144B (en) | Efficient low-cost image fire automatic identification method and device | |
| CN110120142B (en) | Fire smoke video intelligent monitoring early warning system and early warning method | |
| CN102479416A (en) | Method, system and device for eliminating false alarm in monitoring system | |
| CN113420695A (en) | Rapid flame detection method based on MSRCR and YOLOv4-Tiny algorithm | |
| CN104217520A (en) | Oil field fire-prevention prewarning system and fire magnitude judging method thereof | |
| KR20200008229A (en) | Intelligence Fire Detecting System Using CCTV | |
| KR20120138459A (en) | Fire detection device based on image processing with motion detect function | |
| CN110796826A (en) | Alarm method and system for identifying smoke flame | |
| TWI807354B (en) | Fire detection system and fire detection method based on artificial intelligence and image recognition |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |