CN114627610B - Disaster condition processing method and device based on image recognition - Google Patents
Disaster condition processing method and device based on image recognition Download PDFInfo
- Publication number
- CN114627610B CN114627610B CN202210248429.7A CN202210248429A CN114627610B CN 114627610 B CN114627610 B CN 114627610B CN 202210248429 A CN202210248429 A CN 202210248429A CN 114627610 B CN114627610 B CN 114627610B
- Authority
- CN
- China
- Prior art keywords
- image
- flame
- environment image
- fire
- environment
- 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.)
- Active
Links
- 238000003672 processing method Methods 0.000 title claims abstract description 22
- 238000012545 processing Methods 0.000 claims description 47
- 230000007613 environmental effect Effects 0.000 claims description 27
- 238000000034 method Methods 0.000 claims description 16
- 238000009826 distribution Methods 0.000 claims description 10
- 230000001960 triggered effect Effects 0.000 claims description 9
- 230000008859 change Effects 0.000 claims description 8
- 230000003247 decreasing effect Effects 0.000 claims description 5
- 230000000750 progressive effect Effects 0.000 claims description 3
- 238000001514 detection method Methods 0.000 abstract description 16
- 230000007480 spreading Effects 0.000 abstract description 4
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 6
- 239000007788 liquid Substances 0.000 description 5
- 239000000463 material Substances 0.000 description 5
- 230000007547 defect Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000002485 combustion reaction Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000010355 oscillation Effects 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000005375 photometry Methods 0.000 description 1
- 239000000779 smoke Substances 0.000 description 1
- 230000001629 suppression Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B17/00—Fire alarms; Alarms responsive to explosion
- G08B17/12—Actuation by presence of radiation or particles, e.g. of infrared radiation or of ions
- G08B17/125—Actuation by presence of radiation or particles, e.g. of infrared radiation or of ions by using a video camera to detect fire or smoke
-
- A—HUMAN NECESSITIES
- A62—LIFE-SAVING; FIRE-FIGHTING
- A62C—FIRE-FIGHTING
- A62C37/00—Control of fire-fighting equipment
Landscapes
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Public Health (AREA)
- Fire-Detection Mechanisms (AREA)
Abstract
The invention relates to a disaster situation processing method based on image recognition, which comprises the following steps: collecting an environment image in real time, and judging whether the collected environment image is a suspected flame image or not; if the acquired environment image contains a suspected flame image, carrying out secondary judgment on the suspected flame image in the environment image, and determining whether flame exists in the environment image; if the flame exists in the environment image, carrying out image information identification on the environment image to judge the fire trend in the environment image; and corresponding operation is performed according to the identified fire trend in the environment image so as to prevent the expansion of the fire. The invention can judge, verify and identify the flame points in the surrounding environment in real time, greatly reduces the possibility of false detection and missing detection of the flame points, improves the detection rate and the identification precision, can identify the fire trend, can remedy the fire in time, and greatly reduces the possibility of fire spreading.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a disaster situation processing method and device based on image recognition.
Background
Fire is one of the most common disasters, and is a disaster caused by burning which is out of control in space-time; in a complex environment with large space, such as a residential area, a forest, a warehouse and the like, once a fire disaster occurs, huge economic losses are caused, and the normal life of surrounding residents is seriously influenced.
Fire detectors are currently in widespread use for detecting building smoke and fire generation. Because they need particle arrival by ionization or photometry, they cannot be used in open spaces and large coverage areas, and because of the large monitoring range, the cost of applying the original monitoring is very high, and it is difficult to popularize and apply.
With the development of computer technology, the fire detection method through digital image processing overcomes the defects of small coverage range, short effective detection distance and the like of the traditional detection technology.
However, the fire detection method by digital image processing has the defects of easy false detection, omission detection and the like, has low detection and identification precision, and cannot predict the fire spreading trend, so improvement is urgently needed.
Disclosure of Invention
In order to overcome the technical defects in the prior art, the invention provides a disaster processing method and device based on image recognition, which can effectively solve the problems in the background art.
In order to solve the technical problems, the technical scheme provided by the invention is as follows:
the embodiment of the invention discloses a disaster condition processing method based on image recognition, which comprises the following steps:
Collecting an environment image in real time, and judging whether the collected environment image is a suspected flame image or not;
If the acquired environment image contains a suspected flame image, carrying out secondary judgment on the suspected flame image in the environment image, and determining whether flame exists in the environment image;
if the flame exists in the environment image, carrying out image information identification on the environment image to judge the fire trend in the environment image;
and corresponding operation is performed according to the identified fire trend in the environment image so as to prevent the expansion of the fire.
In any of the above schemes, preferably, the method includes detecting flame pixels in the collected environmental image by establishing an RGB probability model based on a single peak gaussian to determine whether the collected environmental image is a suspected flame image.
In any of the above schemes, it is preferable that whether the acquired environmental image is a suspected flame image is judged by:
setting the RGB channel distribution of each pixel in the acquired environment image to be independent, and establishing Gaussian probability distribution: Wherein, I i (x, y) is the color value of the (x, y) coordinate in the ith channel, mu i is the average value of I i (x, y), and sigma i is the standard value of I i (x, y);
By the formula: Mu i and sigma i are estimated, where N is the number of channels, t is the time point,/> The color value of the coordinates (x, y) in the ith channel at the time point t;
And setting a color threshold tau, if P i is larger than tau, judging as suspected flames, and if P i is smaller than tau, judging as non-suspected flames, wherein P i is a probability density function of Gaussian distribution of color values with coordinates (x, y) in an ith channel.
In any of the above aspects, it is preferable that the secondary judgment is performed on the environmental image determined to be the suspected flame, and whether the environmental image is the environmental image for determining the flame is determined, wherein the secondary judgment method includes the steps of:
The image storage module judges that the environment image of the suspected flame takes the upper left corner of the image as a searching starting point, searches in a progressive scanning mode, and starts edge scanning after the first suspected flame pixel point of the target area is searched;
encoding the direction code when performing edge scanning until all edge points of the target area are scanned out, and obtaining an edge chain code;
The relative length of the points with even number of edge chain codes is marked as 1, and the relative length of the points with odd number of edge chain codes is marked as Obtaining the perimeter L Ω of the target area and the area S Ω of the target area;
Setting a circularity judging threshold value xi, and passing through a formula Calculating the circularity C Ω of the target area, judging the suspected flame as the determined flame if C Ω is larger than the circularity judgment threshold value xi, and judging the suspected flame as the non-determined flame if C Ω is smaller than the circularity judgment threshold value xi.
In any of the above aspects, it is preferable that the image information recognition is performed on the environmental image in which the determined flame exists to judge the tendency of the fire, wherein the image information recognition method is as follows:
Assuming that a target area in the acquired environment image is unique, and acquiring the environment image of k frames as a statistical sample;
setting the area average value of the target area as The change trend is R (M), and the formula is adoptedCalculate the area average as/>And the change trend is R (M), wherein S n is the area of the target area in the nth frame of image;
When R (M) > 1, it is determined that the area of the target area is in an increasing trend, the image information of the environment image which determines flame is in an increasing trend, and when R (M) <1, it is determined that the area of the target area is in a decreasing trend, the image information of the environment image which determines flame is in a decreasing trend.
In any of the above schemes, preferably, when the suspected flame in the collected environmental image is determined to be the determined flame, the disaster situation processing device is driven to move to the determined flame point to perform the fire extinguishing operation.
In any of the above aspects, it is preferable that the first fire processing time threshold T 1 and the second fire processing time threshold T 2 are set, and after the fire extinguishing operation T 1 is performed, an alarm is triggered if it is determined that the determined flame trend in the environmental image is an increase in fire.
In any of the above embodiments, preferably, after the fire extinguishing operation T 1 is performed, if it is determined that the determined flame trend in the environmental image is a decrease in fire, the alarm system is not triggered, and the fire extinguishing operation is continued.
In any of the above schemes, it is preferable that if the fire extinguishing operation is continued until the time period reaches T 2, a certain flame point still exists in the environment image, and an alarm is triggered.
In a second aspect, an image recognition-based disaster situation processing apparatus includes:
The acquisition module is used for acquiring the environment image in real time and judging whether the acquired environment image is a suspected flame image or not;
The secondary judging module is used for carrying out secondary judgment on the suspected flame image in the environment image if the collected environment image contains the suspected flame image, and determining whether flame exists in the environment image;
The identification module is used for carrying out image information identification on the environment image if the flame exists in the environment image, so as to judge the fire trend in the environment image;
And the control module is used for performing corresponding operation according to the identified fire trend in the environment image so as to prevent the expansion of the fire.
Compared with the prior art, the invention has the beneficial effects that:
The method comprises the steps of collecting an environment image in real time and judging whether the collected environment image is a suspected flame image or not; if the acquired environment image contains a suspected flame image, carrying out secondary judgment on the suspected flame image in the environment image, and determining whether flame exists in the environment image; if the flame exists in the environment image, carrying out image information identification on the environment image to judge the fire trend in the environment image; corresponding operation is carried out according to the identified fire trend in the environment image so as to prevent the expansion of the fire; the flame point detecting device can judge, verify and identify flame points in a surrounding environment in real time, greatly reduces the possibility of false detection and missing detection of the flame points, improves the detection rate and the identification precision, can identify the fire trend, can timely remedy the fire, and greatly reduces the possibility of fire spreading.
Drawings
The accompanying drawings are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification.
FIG. 1 is a schematic flow chart of a disaster situation processing method based on image recognition;
FIG. 2 is a schematic block diagram of a disaster situation processing device based on image recognition according to the present invention;
FIG. 3 is a schematic view showing an internal structure of a disaster processing device based on an image recognition method of the present invention;
FIG. 4 is a schematic view of an external structure of a disaster processing device based on an image recognition method of the present invention;
Fig. 5 is a schematic view of a slide block of a disaster processing device based on an image recognition method of the present invention.
The reference numerals in the figures illustrate:
1. A fire suppression assembly; 11. a case; 12. an adjustment assembly; 13. a fire extinguishing pipe; 14. a material storage assembly; 15. a limit spring; 121. adjusting a motor; 122. a threaded rod; 123. a sliding block; 124. an electric telescopic rod; 1231. a groove; 141. a storage tank; 142. a feed pipe; 143. a water pump; 144. and (3) a hose.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
It will be understood that when an element is referred to as being "mounted" or "disposed" on another element, it can be directly on the other element or be indirectly on the other element. When an element is referred to as being "connected to" another element, it can be directly connected to the other element or be indirectly connected to the other element.
In the description of the present invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," and the like indicate orientations or positional relationships based on the orientation or positional relationships shown in the drawings, merely to facilitate describing the present invention and simplify the description, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be configured and operated in a specific orientation, and therefore should not be construed as limiting the present invention.
Furthermore, the terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include one or more such feature. In the description of the present invention, the meaning of "a plurality" is two or more, unless explicitly defined otherwise.
In order to better understand the above technical scheme, the following detailed description of the technical scheme of the present invention will be given with reference to the accompanying drawings of the specification and the specific embodiments.
As shown in fig. 1, the present invention provides a disaster situation processing method based on image recognition, which includes the following steps:
Step one, acquiring an environment image in real time, and judging whether the acquired environment image is a suspected flame image or not.
Specifically, in the invention, environmental images are acquired in real time through a disaster processing device, wherein the disaster processing device is provided with an image storage module, an image processing module and a central processing module, the image storage module is used for acquiring environmental image information and judging suspected flames of the environmental images, if the environmental images are judged to be suspected flames, the environmental images are stored, and the acquired environmental image information is sent to the image processing module; the image processing module is used for performing image processing on the environment image stored by the image storage module so as to improve the processing operation on the environment image; the central processing module is used for receiving the environment image processed by the image processing module and identifying the environment image so as to obtain fire information in the environment image.
Furthermore, a camera can be arranged on the disaster processing device, and can collect images under various conditions of daytime, night, backlight and forward light and transmit data to a central processing module; in addition, a plurality of high-definition cameras can be arranged on the disaster processing device, so that the driving route of the disaster processing device can be conveniently identified, meanwhile, the driving route of the disaster processing device can be further accurately determined by means of an infrared sensor, or the driving track of the disaster processing device can be determined by a GPS or Beidou positioning system in a wireless communication mode, and then data information is transmitted to a central processing module.
Preferably, the traveling assembly of the disaster situation processing device may be a traveling tool driven by a power system, for example, a two-wheel, three-wheel or four-wheel passenger car driven by a power source or other energy sources as an engine power source, or a robot driven by a power source or other energy sources as an engine power source.
Further, since the video image collected by the camera is based on the RGB color model, the flame pixels in the collected environmental image can be detected by establishing the RGB probability model based on the single peak gaussian so as to determine whether the collected environmental image is a suspected flame image.
Further, the image storage module judges whether the collected environmental image is a suspected flame image or not through the following steps:
setting the RGB channel distribution of each pixel in the acquired environment image to be independent, and establishing Gaussian probability distribution: Wherein, I i (x, y) is the color value of the (x, y) coordinate in the ith channel, mu i is the average value of I i (x, y), and sigma i is the standard value of I i (x, y);
By the formula: Mu i and sigma i are estimated, where N is the number of channels, t is the time point,/> The color value of the coordinates (x, y) in the ith channel at the time point t;
And setting a color threshold tau, if P i is larger than tau, judging as suspected flames, and if P i is smaller than tau, judging as non-suspected flames, wherein P i is a probability density function of Gaussian distribution of color values with coordinates (x, y) in an ith channel.
For example, when the product of 3 probability distributions of a pixel with coordinates (x, y) is greater than the color threshold τ, the pixel is regarded as a flame pixel, and the suspected flame region can be detected through the steps, so as to improve the early warning capability of fire occurrence.
If the acquired environment image contains a suspected flame image, performing secondary judgment on the suspected flame image in the environment image to determine whether flame exists in the environment image.
Specifically, in the present invention, the central processing module performs a secondary judgment on the environmental image determined to be a suspected flame by the image storage module, and determines whether the environmental image is an environmental image for determining a flame, where the secondary judgment method of the central processing module is as follows:
The image storage module judges that the environment image of the suspected flame takes the upper left corner of the image as a searching starting point, searches in a progressive scanning mode, and starts edge scanning after the first suspected flame pixel point of the target area is searched;
encoding the direction code when performing edge scanning until all edge points of the target area are scanned out, and obtaining an edge chain code;
The relative length of the points with even number of edge chain codes is marked as 1, and the relative length of the points with odd number of edge chain codes is marked as Obtaining the perimeter L Ω of the target area and the area S Ω of the target area;
Setting a circularity judging threshold value xi, and passing through a formula Calculating the circularity C Ω of the target area, judging the suspected flame as the determined flame if C Ω is larger than the circularity judgment threshold value xi, and judging the suspected flame as the non-determined flame if C Ω is smaller than the circularity judgment threshold value xi.
In the disaster condition processing method based on image recognition, as the circularity of the interference light source such as a candle, a flashlight, sunlight and the like is very close to 1, preferably, the circularity judgment threshold value xi is more than 2, so that fire flames and partial interference officers can be accurately distinguished, and the recognition accuracy is improved.
And thirdly, if the flame exists in the environment image, carrying out image information identification on the environment image to judge the fire trend in the environment image.
In the disaster processing method based on image recognition, because flames are greatly influenced by environmental factors, a target area is not necessarily unique, omega 1,Ω2,…,ΩN is marked in sequence, and the area of the corresponding target area is marked as S 1,S2,…,Sn; furthermore, since combustion is a dynamic unstable process, the flame area is subject to continuous oscillation change, and image recognition is required for determining the environment image of the flame, so as to determine the trend of the fire, and facilitate control of the fire.
Further, in the present invention, after the secondary judgment is performed on the environmental image, the central processing module performs image information identification on the environmental image for determining flame so as to judge the trend of fire, wherein the image information identification method of the central processing module is as follows:
Assuming that a target area in the acquired environment image is unique, and acquiring the environment image of k frames as a statistical sample;
setting the area average value of the target area as The change trend is R (M), and the formula is adoptedCalculate the area average as/>And the change trend is R (M), wherein S n is the area of the target area in the nth frame of image;
When R (M) > 1, the area of the target area is judged to be an increasing trend, the image information of the environment image of the flame is determined to be a fire increase, and when R (M) < 1, the area of the target area is judged to be a decreasing trend, the image information of the environment image of the flame is determined to be a fire decrease.
In the disaster processing method based on image recognition, the range of n is continuous k frames of images from a certain frame, after each calculation is completed, the images removed again are not k+1 to 2k frames, but the image of the (k+1) th frame is used for replacing the image of the first frame, and the like is performed, so that the influence caused by the vibration change of the flame area is reduced.
In the disaster processing method based on image recognition, as the rapid requirement on flame recognition is provided, the number of frames is not easy to be excessive, preferably k is 10, so that the rapid determination of the flame environment image is realized to judge the fire, and the flame can be accurately distinguished from a lighting tool such as a candle.
And step four, corresponding operation is carried out according to the identified fire trend in the environment image so as to prevent the expansion of the fire.
Specifically, when the central processing module judges that the suspected flame in the collected environment image is the determined flame, the central processing module drives the disaster processing device to move to the determined flame point to perform fire extinguishing operation.
Further, a first fire processing time threshold T 1 and a second fire processing time threshold T 2 are set in the central processing module, after the fire extinguishing operation is performed for a period of time T 1, if the central processing module determines that the determined flame trend in the environment image is a decrease in fire, the alarm system is not triggered, the fire extinguishing operation is continued until the time T 2 is reached, and if the determined flame point still exists in the environment image at this time, an alarm is triggered.
Further, after the fire extinguishing operation T 1 is performed, if the central processing module determines that the determined flame trend in the environmental image is an increase in fire, an alarm is triggered.
Further, as shown in fig. 3 to 5, the disaster treatment device comprises a fire extinguishing assembly 1, the fire extinguishing assembly 1 comprises a box body 11, an adjusting assembly 12, a fire extinguishing pipe 13, a material storage assembly 14 and a limiting spring 15, the adjusting assembly 12 is arranged inside the box body 11, the material storage assembly 14 is arranged inside the box body 11, the output end of the material storage assembly 14 is connected with the fire extinguishing pipe 13, one end of the fire extinguishing pipe 13 is positioned outside the box body 11, the other end of the fire extinguishing pipe 13 extends to the inside of the box body 11 and is connected with the output end of the adjusting assembly 12, one end of the limiting spring 15 is fixedly connected with the outer wall of the fire extinguishing pipe 13, and the other end of the limiting spring 15 is fixedly connected with the box body 11.
Further, the adjusting component 12 includes accommodate motor 121, threaded rod 122, sliding block 123, electric telescopic rod 124, accommodate motor 121 is fixed in the box 11 outer wall, threaded rod 122 one end with accommodate motor 121 output fixed connection, the threaded rod 122 other end extends to the box 11 is inside and with the inside lateral wall rotation of box 11 is connected, the sliding block 123 bottom is provided with recess 1231, the sliding block 123 passes through recess 1231 with the inside bottom surface sliding connection of box 1, just the sliding block 123 cover is located threaded rod 122 is last and with threaded rod 122 threaded connection, electric telescopic rod 124 one end with sliding block 123 top fixed connection, the electric telescopic rod 124 other end with fire-extinguishing tube 13 extends to the inside one end bottom fixed connection of box 11.
In the disaster situation processing method based on image recognition according to the embodiment of the present invention, the adjusting motor 121 works to rotate the threaded rod 122, so as to drive the sliding block 123 to move in the case 11, so as to adjust the horizontal angle of the fire extinguishing pipe 13, and further, the electric telescopic rod 124 works to drive the fire extinguishing pipe 13 to move, so as to adjust the vertical angle of the fire extinguishing pipe 13, so that one end of the fire extinguishing pipe 13 outside the case 11 faces exactly towards the fire spot.
Further, the material storage assembly 14 includes a storage tank 141, a feed pipe 142, a water pump 143, and a hose 144, wherein the bottom of the storage tank 141 is fixedly connected with the bottom surface inside the tank 11, one end of the feed pipe 142 is communicated with the storage tank 141, the other end of the feed pipe 142 extends to the outside of the tank 11, the water pump 143 is fixed at the top of the storage tank 141, the input end of the water pump 143 is communicated with the storage tank 141, the output end of the water pump 143 is communicated with one end of the hose 144, and the other end of the hose 144 is communicated with the fire extinguishing pipe 13.
In the disaster situation processing method based on image recognition according to the embodiment of the present invention, the fire extinguishing liquid is stored in the storage tank 141, and the fire extinguishing liquid in the storage tank 141 can be replenished when the fire extinguishing liquid in the storage tank 141 is insufficient through the feeding pipe 142, and the fire extinguishing liquid in the storage tank 141 can be transferred into the fire extinguishing pipe 13 through the operation of the water pump 143, and then the fire extinguishing liquid is sprayed to a fire spot through the fire extinguishing pipe 13, so as to process the fire.
Further, the disaster situation processing device further includes a traveling assembly, where the traveling assembly is disposed at the bottom of the box 11, and the traveling assembly may be a traveling tool driven by a power system, for example, a two-wheel, three-wheel or four-wheel passenger car driven by a power source or other energy sources as an engine power source, or a robot driven by a power source or other energy sources as an engine power source.
In a second aspect, as shown in fig. 2, the present invention provides a disaster situation processing apparatus based on image recognition, comprising:
The acquisition module is used for acquiring the environment image in real time and judging whether the acquired environment image is a suspected flame image or not;
The secondary judging module is used for carrying out secondary judgment on the suspected flame image in the environment image if the collected environment image contains the suspected flame image, and determining whether flame exists in the environment image;
The identification module is used for carrying out image information identification on the environment image if the flame exists in the environment image, so as to judge the fire trend in the environment image;
And the control module is used for performing corresponding operation according to the identified fire trend in the environment image so as to prevent the expansion of the fire.
Compared with the prior art, the invention has the beneficial effects that:
The method comprises the steps of collecting an environment image in real time and judging whether the collected environment image is a suspected flame image or not; if the acquired environment image contains a suspected flame image, carrying out secondary judgment on the suspected flame image in the environment image, and determining whether flame exists in the environment image; if the flame exists in the environment image, carrying out image information identification on the environment image to judge the fire trend in the environment image; corresponding operation is carried out according to the identified fire trend in the environment image so as to prevent the expansion of the fire; the flame point detecting device can judge, verify and identify flame points in a surrounding environment in real time, greatly reduces the possibility of false detection and missing detection of the flame points, improves the detection rate and the identification precision, can identify the fire trend, can timely remedy the fire, and greatly reduces the possibility of fire spreading.
The above is only a preferred embodiment of the present invention, and the present invention is not limited thereto, but it is to be understood that the present invention is described in detail with reference to the foregoing embodiments, and modifications and equivalents of some of the technical features described in the foregoing embodiments may be made by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (9)
1. A disaster condition processing method based on image recognition is characterized in that: the method comprises the following steps:
Collecting an environment image in real time, and judging whether the collected environment image is a suspected flame image or not;
If the acquired environment image contains a suspected flame image, carrying out secondary judgment on the suspected flame image in the environment image to determine whether flame exists in the environment image, wherein the secondary judgment method comprises the following steps of:
Searching the environment image which is judged to be suspected flame in a progressive scanning mode by taking the upper left corner of the image as a searching starting point, and starting edge scanning after searching the first suspected flame pixel point of the target area;
encoding the direction code when performing edge scanning until all edge points of the target area are scanned out, and obtaining an edge chain code;
The relative length of the points with even number of edge chain codes is marked as 1, and the relative length of the points with odd number of edge chain codes is marked as Obtaining the perimeter L Ω of the target area and the area S Ω of the target area;
Setting a circularity judging threshold value xi, and passing through a formula Calculating the circularity C Ω of the target area, judging the suspected flame as a determined flame if C Ω is larger than a circularity judgment threshold value xi, and judging the suspected flame as a non-determined flame if C Ω Small size is smaller than the circularity judgment threshold value xi;
if the flame exists in the environment image, carrying out image information identification on the environment image to judge the fire trend in the environment image;
and corresponding operation is performed according to the identified fire trend in the environment image so as to prevent the expansion of the fire.
2. The disaster situation processing method based on image recognition according to claim 1, wherein: and detecting flame pixels in the acquired environment image by establishing an RGB probability model based on single peak Gaussian so as to judge whether the acquired environment image is a suspected flame image.
3. The disaster situation processing method based on image recognition according to claim 2, wherein: judging whether the acquired environment image is a suspected flame image or not through the following steps:
setting the RGB channel distribution of each pixel in the acquired environment image to be independent, and establishing Gaussian probability distribution: Wherein, I i (x, y) is the color value of the (x, y) coordinate in the ith channel, mu i is the average value of I i (x, y), and sigma i is the standard value of I i (x, y);
By the formula: Mu i and sigma i are estimated, where N is the number of channels, t is the time point,/> The color value of the coordinates (x, y) in the ith channel at the time point t;
And setting a color threshold tau, if P i is larger than tau, judging as suspected flames, and if P i is smaller than tau, judging as non-suspected flames, wherein P i is a probability density function of Gaussian distribution of color values with coordinates (x, y) in an ith channel.
4. The disaster situation processing method based on image recognition according to claim 3, wherein: and carrying out image information identification on the environment image with the determined flame so as to judge the trend of the fire, wherein the image information identification method comprises the following steps:
Assuming that a target area in the acquired environment image is unique, and acquiring the environment image of k frames as a statistical sample;
setting the area average value of the target area as The change trend is R (M), and the formula/> Calculate the area average as/>And the change trend is R (M), wherein S n is the area of the target area in the nth frame of image;
When R (M) > 1, it is determined that the area of the target area is in an increasing trend, the image information of the environment image which determines flame is in an increasing trend, and when R (M) <1, it is determined that the area of the target area is in a decreasing trend, the image information of the environment image which determines flame is in a decreasing trend.
5. The disaster situation processing method based on image recognition according to claim 4, wherein: when the suspected flames in the collected environment images are judged to be the determined flames, driving the disaster processing device to move to the determined flame points to perform fire extinguishing operation.
6. The disaster situation processing method based on image recognition according to claim 5, wherein: after the first fire processing time threshold T 1 and the second fire processing time threshold T 2 are set and the fire extinguishing operation T 1 is performed, if it is determined that the determined flame trend in the environmental image is an increase in fire, an alarm is triggered.
7. The disaster situation processing method based on image recognition according to claim 6, wherein: after the fire extinguishing operation T 1 is performed, if the determined flame trend in the environment image is determined to be the decrease of the fire, the alarm system is not triggered, and the fire extinguishing operation is continued.
8. The disaster situation processing method based on image recognition according to claim 7, wherein: if the duration of the fire extinguishing operation is continued to reach T 2, a determined flame point still exists in the environment image, and an alarm is triggered.
9. The utility model provides a disaster situation processing apparatus based on image recognition which characterized in that: the method applied in claim 1, comprising:
The acquisition module is used for acquiring the environment image in real time and judging whether the acquired environment image is a suspected flame image or not;
The secondary judging module is used for carrying out secondary judgment on the suspected flame image in the environment image if the collected environment image contains the suspected flame image, and determining whether flame exists in the environment image;
The identification module is used for carrying out image information identification on the environment image if the flame exists in the environment image, so as to judge the fire trend in the environment image;
And the control module is used for performing corresponding operation according to the identified fire trend in the environment image so as to prevent the expansion of the fire.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210248429.7A CN114627610B (en) | 2022-03-14 | 2022-03-14 | Disaster condition processing method and device based on image recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210248429.7A CN114627610B (en) | 2022-03-14 | 2022-03-14 | Disaster condition processing method and device based on image recognition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114627610A CN114627610A (en) | 2022-06-14 |
| CN114627610B true CN114627610B (en) | 2024-06-14 |
Family
ID=81901454
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210248429.7A Active CN114627610B (en) | 2022-03-14 | 2022-03-14 | Disaster condition processing method and device based on image recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114627610B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115359617A (en) * | 2022-08-26 | 2022-11-18 | 新创碳谷控股有限公司 | Oxidation furnace flame detection method, computer equipment and storage medium |
| CN115713833A (en) * | 2022-08-30 | 2023-02-24 | 新创碳谷集团有限公司 | Flame detection method and device based on area characteristics and storage medium |
| CN116071883A (en) * | 2022-12-13 | 2023-05-05 | 华能山西综合能源有限责任公司 | Fire alarm system and method for photovoltaic power station |
| CN116308975A (en) * | 2023-05-17 | 2023-06-23 | 山东金桥保安器材有限公司 | Security data processing method and system based on image recognition |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN110516609A (en) * | 2019-08-28 | 2019-11-29 | 南京邮电大学 | A kind of fire video detection and method for early warning based on image multiple features fusion |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101315667B (en) * | 2008-07-04 | 2011-01-19 | 南京航空航天大学 | Multi-characteristic synthetic recognition method for outdoor early fire disaster |
| CN105788142B (en) * | 2016-05-11 | 2018-08-31 | 中国计量大学 | A kind of fire detection system and detection method based on Computer Vision |
| JP6722041B2 (en) * | 2016-05-18 | 2020-07-15 | 株式会社日立国際電気 | Monitoring system |
| CN111340746A (en) * | 2020-05-19 | 2020-06-26 | 深圳应急者安全技术有限公司 | Fire fighting method and fire fighting system based on Internet of things |
-
2022
- 2022-03-14 CN CN202210248429.7A patent/CN114627610B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN110516609A (en) * | 2019-08-28 | 2019-11-29 | 南京邮电大学 | A kind of fire video detection and method for early warning based on image multiple features fusion |
Non-Patent Citations (1)
| Title |
|---|
| 基于小波变换和支持向量机的火灾识别算法;邹婷;王慧琴;胡燕;梁俊山;殷颖;;计算机工程与应用(14);第250-253页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114627610A (en) | 2022-06-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114627610B (en) | Disaster condition processing method and device based on image recognition | |
| CN210233046U (en) | Rail mounted utility tunnel patrols and examines robot and system | |
| CN101498889B (en) | Multi-eye stereo camera shooting method and device | |
| CN110081982B (en) | Unmanned aerial vehicle target positioning method based on double-spectrum photoelectric search | |
| CN108897342B (en) | Positioning and tracking method and system for fast-moving civil multi-rotor unmanned aerial vehicle | |
| CN115019512A (en) | Road event detection system based on radar video fusion | |
| CN210899299U (en) | Tunnel monitoring system | |
| CN111985365A (en) | Straw burning monitoring method and system based on target detection technology | |
| CN114265418A (en) | Unmanned aerial vehicle inspection and defect positioning system and method for photovoltaic power station | |
| CN113819881B (en) | Method for detecting distance and map azimuth of fire source for reconnaissance inspection robot | |
| CN111323767B (en) | System and method for detecting obstacle of unmanned vehicle at night | |
| CN113203409A (en) | Method for constructing navigation map of mobile robot in complex indoor environment | |
| CN115512307A (en) | Wide-area space infrared multi-point real-time fire detection method and system and positioning method | |
| CN112257554A (en) | Forest fire recognition method, system, program and storage medium | |
| CN115035470A (en) | Low, small and slow target identification and positioning method and system based on mixed vision | |
| CN112145976A (en) | Detection system and method based on infrared gas cloud imaging and robot applying system | |
| CN112489125A (en) | Automatic detection method and device for storage yard pedestrians | |
| CN114051093B (en) | Portable navigation mark lamp field detection system based on image processing technology | |
| CN114913323A (en) | Method for detecting open fire at night in charging pile area | |
| CN111667656A (en) | System and method for discriminating forest fire points of power transmission line | |
| CN110648419A (en) | Inspection system and method for pipe gallery inspection robot | |
| CN117372549A (en) | Multi-view license plate color recognition method and system | |
| CN115014669B (en) | Automatic response type tunnel water leakage detection method, system and device | |
| Yao et al. | Automatic concrete tunnel inspection robot system | |
| CN214277928U (en) | Detection robot based on infrared gas cloud imaging |
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 | ||
| CB03 | Change of inventor or designer information |
Inventor after: Zhou Ying Inventor after: Li Xiao Inventor before: Zhou Ying |
|
| CB03 | Change of inventor or designer information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |