CN114612828A - Construction site fire monitoring and early warning method based on image analysis - Google Patents
Construction site fire monitoring and early warning method based on image analysis Download PDFInfo
- Publication number
- CN114612828A CN114612828A CN202210240280.8A CN202210240280A CN114612828A CN 114612828 A CN114612828 A CN 114612828A CN 202210240280 A CN202210240280 A CN 202210240280A CN 114612828 A CN114612828 A CN 114612828A
- Authority
- CN
- China
- Prior art keywords
- risk
- area
- early warning
- processing module
- fire
- 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
- 238000012544 monitoring process Methods 0.000 title claims abstract description 106
- 238000010276 construction Methods 0.000 title claims abstract description 33
- 238000010191 image analysis Methods 0.000 title claims abstract description 21
- 238000000034 method Methods 0.000 title claims abstract description 20
- 238000000605 extraction Methods 0.000 claims abstract description 6
- 238000012795 verification Methods 0.000 claims description 47
- 239000003086 colorant Substances 0.000 claims description 3
- 239000000779 smoke Substances 0.000 description 7
- 239000000463 material Substances 0.000 description 2
- 238000005192 partition Methods 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 239000004793 Polystyrene Substances 0.000 description 1
- 229910000831 Steel Inorganic materials 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 239000010426 asphalt Substances 0.000 description 1
- 239000004566 building material Substances 0.000 description 1
- 239000000470 constituent Substances 0.000 description 1
- 239000011162 core material Substances 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000003973 paint Substances 0.000 description 1
- 239000004033 plastic Substances 0.000 description 1
- 229920002223 polystyrene Polymers 0.000 description 1
- 239000010959 steel Substances 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 239000002023 wood Substances 0.000 description 1
Images
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
Abstract
The invention relates to a construction site fire monitoring and early warning method based on image analysis, in particular to the technical field of image processing, which comprises the following steps that S1, a fire monitoring video file of a construction site is obtained in real time through an acquisition module; step S2, extracting a video frame file from the video file through an extraction module, extracting video frames in the video file by taking seconds as a unit when extracting the video frames from the obtained video file, and taking the extracted video frames as fire point monitoring images; step S3, image processing is carried out on the fire point monitoring image through a processing module, and the processing module equally divides the fire point monitoring image into 3 multiplied by 3 grid areas; and step S4, performing fire point early warning according to the image processing result through an early warning module, performing corresponding early warning according to the regional risk coefficient L0 of the fire point monitoring image by the early warning module, and adjusting the regional risk coefficient L0 according to the risk target region ratio H. The invention effectively improves the monitoring efficiency of the fire point.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a construction site fire point monitoring and early warning method based on image analysis.
Background
When the construction site enters a high-temperature time section, the construction site is in a high-temperature construction stage in summer, and a fire disaster is easily caused by carelessness. At present, temporary buildings on construction sites of a plurality of construction projects are built by adopting color steel plates with polystyrene as core materials, and the building materials have low fire resistance grade but low price, so the temporary buildings are favored by a plurality of construction units. Scaffolds and safety shields for construction are also commonly made of combustible materials. A large amount of combustible and combustible objects such as asphalt felt, wood, paint, plastic products, decorative materials and the like are stored and used in a construction site, and the stacking phenomenon is serious.
At present, a camera is usually installed for manual monitoring on fire points in project construction sites, but the manual monitoring has hysteresis and is influenced by manual attention, and the optimal rescue time is easy to miss when a fire happens, so that the fire point monitoring efficiency is low.
Disclosure of Invention
Therefore, the invention provides a construction site fire monitoring and early warning method based on image analysis, which is used for solving the problem of low fire monitoring efficiency caused by the fact that risk early warning cannot be carried out through accurate analysis of monitored images in the prior art.
In order to achieve the above object, the present invention provides a method for monitoring and warning a fire point at a construction site based on image analysis, comprising,
step S1, acquiring a fire monitoring video file of the construction site in real time through an acquisition module;
step S2, extracting a video frame file from the video file through an extraction module, extracting video frames in the video file by taking seconds as a unit when extracting the video frames from the obtained video file, and taking the extracted video frames as fire point monitoring images;
step S3, image processing is carried out on the fire monitoring image through a processing module, when the image processing is carried out, the processing module equally divides the fire monitoring image into 3 x 3 grid areas, the central rectangular area of the grid areas is used as a primary area, other rectangular areas are used as secondary areas, the processing module divides each rectangular area according to different gray values, each area formed after division in the rectangular area is used as a target area, the processing module carries out risk area judgment on the rectangular area according to the number of the target areas in the rectangular area, and the processing module also carries out risk grade judgment on each target area so as to finish the risk area judgment on each rectangular area;
and step S4, performing fire point early warning according to the image processing result through an early warning module, performing corresponding early warning according to the regional risk coefficient L0 of the fire point monitoring image by the early warning module, and adjusting the regional risk coefficient L0 according to the risk target region ratio H.
Further, when the processing module determines the risk area of the rectangular area, the processing module obtains the number n of target areas in the rectangular area, and determines the risk area of the rectangular area according to the number n of the target areas, wherein,
when n is equal to 1, the processing module acquires the tone of the rectangular area, if the tone is white or red, the processing module judges that the rectangular area is a risk area, and if the tone is other colors, the processing module judges that the rectangular area is a safety area;
and when n is larger than 1, the processing module judges the risk level of the target area according to the area boundary of each target area in the rectangular area.
Further, the processing module takes the area boundary of each target area as a target shape curve and obtains the curvature A of the target shape curve, the processing module compares the curvature A of each target shape curve with each preset curvature and judges the risk grade of the target area according to the comparison result, wherein,
when A is less than A1, the processing module judges that the target area is a low risk area;
when A is not less than A1 and not more than A2, the processing module judges that the target area is a high risk area and carries out high risk area verification;
when A2 is less than A, the processing module carries out secondary risk grade judgment according to the gray value C of the target area;
wherein A1 is the first predetermined curvature, A2 is the second predetermined curvature, and A1 is less than A2.
Further, the processing module obtains the graph texture complexity P of the high risk region when performing the high risk region verification, compares the graph texture complexity P with a preset texture complexity P0, and performs the high risk region verification according to the comparison result, wherein,
when P is not more than P0, the processing module judges that the verification is successful and determines that the target area is a high risk area;
when P > P0, the processing module determines that the verification failed and makes a secondary risk level determination for the area.
Further, when the processing module determines the secondary risk level, the processing module compares the gray value C of the target area with each preset gray value, and determines the secondary risk level according to the comparison result, wherein,
when C is less than C1, the processing module judges that the target area is a low risk area;
when C1 is less than or equal to C2, the processing module judges the target area as a medium risk area;
when C2 < C, the processing module judges that the target area is a low risk area;
wherein, C1 is a first preset gray value, C2 is a second preset gray value, and C1 is less than C2.
Further, the processing module performs risk area determination on the rectangular area according to the risk level of each target area in the rectangular area, wherein,
when a high risk area exists in the rectangular area, the processing module judges that the rectangular area is a risk area;
when the number of the risk areas in the rectangular area is larger than m, the processing module judges that the rectangular area is a risk area, m is the number of preset risk areas, and m is larger than or equal to 1.
Further, the early warning module calculates a regional risk coefficient L0 of the fire point monitoring image according to the number of risk regions determined by the processing module, sets L0 to K1 × L + K2 × 0.1 × L, sets K1 to be the number of risk regions of the primary region, sets K2 to be the number of risk regions of the secondary region, and sets L to be a preset risk coefficient, where L is greater than 0, compares the calculated regional risk coefficient L0 with each preset regional risk coefficient, and performs early warning according to the comparison result, where,
when L0 is less than L1, the early warning module judges that the monitoring range is safe and does not perform early warning;
when the L1 is not less than L0 is not less than L2, the early warning module judges that a fire risk exists in the monitoring range and prompts a worker to check;
when L2 is less than L0, the early warning module judges that a fire disaster occurs in the monitoring range and prompts workers to carry out fire fighting;
wherein, L1 is the medium risk coefficient of the preset area, L2 is the high risk coefficient of the preset area, and L1 is less than L2.
Further, when the early warning module calculates the regional risk coefficient L0, the early warning module further adjusts the regional risk coefficient L0 according to the risk target region proportion H, sets H to R/T, where R is the total area of the medium-risk and high-risk target regions, and T is the area of the fire monitoring image, compares the risk target region proportion H with the preset region proportion H0, and adjusts the regional risk coefficient L0 according to the comparison result, where,
when H is less than or equal to H0, the early warning module judges that the calculation result of the regional risk coefficient is accurate and does not adjust;
when H is larger than H0, the early warning module adjusts the regional risk coefficient to be L0 ', sets L0 ═ L0+ L0 × (H-H0)/H, and carries out early warning according to the adjusted regional risk coefficient L0'.
Further, after the early warning module judges that a fire risk exists in the monitoring range, the early warning module acquires a fire monitoring image after t1 time for fire risk verification, t1 is set as first preset verification time, the early warning module calculates a regional risk coefficient L01 of the fire monitoring image after t1 time, if L01 is less than L0, verification fails without early warning, and if L01 is more than or equal to L0, verification succeeds and corresponding early warning is performed.
Further, after the early warning module judges that a fire occurs in the monitoring range, the early warning module acquires a fire monitoring image after t2 time for fire risk verification, t2 is set as second preset verification time, t1 is greater than t2, the early warning module calculates a regional risk coefficient L02 of the fire monitoring image after t2 time, if L02 is less than L0, verification fails and fire risk early warning is performed, a worker is prompted to check the fire risk, and if L01 is greater than or equal to L0, verification succeeds and corresponding early warning is performed.
Compared with the prior art, the invention has the advantages that when the processing module carries out image processing, firstly, the image is divided for partition processing, so that the image processing accuracy is improved, after the image is divided, the rectangular areas formed after the division are also graded, the central area and the edge area are graded, so that the fire monitoring accuracy is improved, the processing module also divides each rectangular area according to the gray value to obtain the target area, and carries out risk area judgment in different modes according to the number of the target areas in the rectangular area to judge whether the rectangular area has fire risks, so that the fire monitoring accuracy is improved, when the number of the target areas in the rectangular area is only one, if the color tone is white or red, the rectangular area is proved to have smoke or fire risks, if the number of the target areas is more than 1, further judgment is needed according to the area sidelines of the target areas so as to ensure the accuracy of the judgment result and improve the accuracy of fire point monitoring.
Particularly, when the risk level is determined, the processing module obtains the curvature of the area edge line of the target area, compares the curvature with a preset value, if the shape of the target area is proved to be not similar to the shape reflected by the preset curvature within the preset value, the target area is a low risk area, if the shape of the target area is proved to be similar to the shape reflected by the preset curvature within the preset range, the risk level is determined to be high risk, if the shape of the target area is higher than the preset value, smoke may exist, further accurate determination is needed, the fire point monitoring accuracy can be further improved by performing the risk level determination on the target area, the processing module also verifies the target area determined to be the high risk area to ensure the accuracy of the determination result, and when the verification is performed, the processing module compares the graphic texture complexity P of the high risk area with the preset value, if the target area is within the preset range, the verification is successful, if the target area is larger than the preset value, the target area is proved to be similar to the flame in shape, but the graph complexity is high, the target area needs to be further judged, so that the accuracy of the risk grade judgment result is ensured, and the fire point monitoring efficiency is further improved.
Particularly, when secondary risk level judgment is carried out, the processing module compares the gray value of the target area with a preset value, if the target area is in a preset value interval, the target area is proved to have smoke risk, the target area is judged to be a medium risk area, and secondary risk level judgment is carried out, so that the accuracy of the judgment result of the risk level of each target area can be effectively improved, and the fire point monitoring efficiency is further improved.
Particularly, after the risk level judgment of each target area is completed, the processing module judges the risk areas of the rectangular areas according to the target area level judgment conditions in the rectangular areas so as to improve the accuracy of the risk area judgment, because the high risk areas represent that the fire has been started and the middle risk areas represent that the smoke has been caused, the processing module judges the risk levels according to the number of different risk areas in the rectangular areas, the accuracy of the risk area judgment can be further ensured, and the accuracy and the efficiency of fire point monitoring are improved.
Particularly, after the processing module determines the risk areas of each rectangular area, the early warning module calculates an area risk coefficient L0 according to the distribution and number of the risk areas, compares the area risk coefficient L0 with a preset value, and performs early warning to different degrees according to the comparison result to improve the accuracy of the early warning, thereby further improving the fire monitoring efficiency, when calculating the area risk coefficient L0, different weights are set for different levels of areas to improve the accuracy of combustible monitoring of the first level area, thereby improving the fire monitoring efficiency, the early warning module compares the ratio H of a risk target area with the preset value, if the ratio is within the preset value, adjustment is not performed, if the ratio is greater than the preset value, more risk areas are proved, and an area risk coefficient L0 needs to be increased to make the early warning more accurate, the early warning module adjusts the area risk coefficient L0, the accuracy of early warning is further improved, and therefore the fire point monitoring efficiency is improved.
Especially, the early warning module still carries out fire risk verification through obtaining the image after the preset time for the degree of accuracy of guaranteeing the early warning, the early warning module is compared the regional danger coefficient of the fire monitoring image after the preset time with the regional danger coefficient L0 of the image that needs the early warning to verify whether carry out corresponding early warning, the early warning module is through carrying out fire risk verification, has further improved the degree of accuracy of early warning, thereby further improves the monitoring efficiency to the fire.
Drawings
FIG. 1 is a schematic structural diagram of a construction site fire monitoring and early warning system based on image analysis according to the present embodiment;
fig. 2 is a schematic flow chart of the construction site fire monitoring and early warning method based on image analysis according to the embodiment.
Detailed Description
In order that the objects and advantages of the invention will be more clearly understood, the invention is further described in conjunction with the following examples; it should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only for explaining the technical principle of the present invention, and do not limit the scope of the present invention.
Furthermore, it should be noted that, in the description of the present invention, unless otherwise explicitly specified or limited, the terms "mounted," "connected," and "connected" are to be construed broadly, and may be, for example, fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood by those skilled in the art according to specific situations.
Referring to fig. 1, it is shown that the construction site fire monitoring and early warning system based on image analysis of the present embodiment includes,
the acquisition module is used for acquiring a fire monitoring video file of the construction site by the camera device in real time;
the extraction module is used for extracting a video frame file from the video file and is connected with the acquisition module;
the processing module is used for processing the image of each frame of video frame and is connected with the extraction module;
and the early warning module is used for carrying out fire point early warning according to an image processing result and is connected with the processing module.
Referring to fig. 2, a schematic flow chart of the method for monitoring and warning of a fire point at a construction site based on image analysis according to the present embodiment is shown, the method includes,
step S1, acquiring a fire monitoring video file of a construction site in real time through an acquisition module;
step S2, extracting a video frame file from the video file through an extraction module; when extracting video frames from the acquired video file, extracting the video frames in the video file by taking seconds as a unit, and taking the extracted video frames as fire monitoring images;
step S3, image processing is carried out on the fire point monitoring image through a processing module;
and step S4, performing fire point early warning through an early warning module according to the image processing result.
Specifically, in step S3, when performing image processing on the fire monitoring image, the processing module equally divides the fire monitoring image into 3 × 3 grid regions, takes a central rectangular region of the grid regions as a primary region and takes other rectangular regions as secondary regions, the processing module divides each rectangular region according to different gray values, when dividing, the processing module obtains constituent pixels of the rectangular regions and obtains gray values of each pixel, takes a region in which a gray difference value of adjacent pixels is within a preset range as a same region, and takes each region formed after division in the rectangular region as a target region, the processing module obtains the number n of target regions in the rectangular region, and performs risk region determination on the rectangular region according to the number n of target regions, wherein,
when n is equal to 1, the processing module acquires the tone of the rectangular area, if the tone is white or red, the processing module judges that the rectangular area is a risk area, and if the tone is other colors, the processing module judges that the rectangular area is a safety area;
and when n is larger than 1, the processing module judges the risk level of the target area according to the area boundary of each target area in the rectangular area.
Specifically, when the processing module performs image processing, the image is divided first, and is divided into the grid areas of 3 × 3 for partition processing, and those skilled in the art can select other sizes according to the size of the object in the image to perform equal division, but it should be noted that the odd-by-odd form should be selected for equal division to ensure that the central area and the edge area exist in the image, and when the camera device is installed, the flammable substance should be ensured to be in the central area of the image to improve the accuracy of image analysis, so as to improve the fire monitoring efficiency, in this embodiment, after the image is divided, the rectangular area formed after division is further divided into levels, and by dividing the central area and the edge area into levels, the method comprises the steps that the accuracy of fire point monitoring is improved, the processing module divides each rectangular area according to gray values to obtain target areas, risk area judgment is carried out in different modes according to the number of the target areas in the rectangular areas, whether fire risks exist in the rectangular areas or not is judged, and therefore the accuracy of fire point monitoring is improved.
Specifically, the processing module takes the area boundary of each target area as a target shape curve and obtains the curvature a of the target shape curve, the processing module compares the curvature a of each target shape curve with each preset curvature and judges the risk level of the target area according to the comparison result, wherein,
when A is less than A1, the processing module judges that the target area is a low risk area;
when A is not less than A1 and not more than A2, the processing module judges that the target area is a high risk area and carries out high risk area verification;
when A2 is less than A, the processing module carries out secondary risk grade judgment according to the gray value C of the target area;
wherein A1 is the first predetermined curvature, A2 is the second predetermined curvature, and A1 is less than A2.
Specifically, the processing module obtains the graph texture complexity P of the high risk region when performing the high risk region verification, compares the graph texture complexity P with a preset texture complexity P0, and performs the high risk region verification according to the comparison result, wherein,
when P is not more than P0, the processing module judges that the verification is successful and determines that the target area is a high risk area;
when P > P0, the processing module determines that the verification failed and makes a secondary risk level determination for the area.
Specifically, in the embodiment, when the risk level determination is performed, the processing module obtains the curvature of the area boundary line of the target area, and compares the curvature with the preset value, in this embodiment, when the preset value is set, the setting is performed in consideration of the shape of the flame sufficiently to ensure that the curvature in the preset range includes curvatures of different flame shapes, so as to improve the accuracy of the determination of the risk area, if the shape of the target area is proved within the preset value not to be similar to the shape reflected by the preset curvature, the target area is a low risk area, if the shape of the target area is proved within the preset range to be similar to the shape reflected by the preset curvature, the risk level is determined to be a high risk, if the shape is above the preset value, smoke may exist, further accurate determination is required, and by performing the risk level determination on the target area, the accuracy of monitoring the fire point can be further improved, and the processing module is also used for verifying the target area judged as the high-risk area so as to ensure the accuracy of the judgment result, and when the verification is carried out, the processing module is used for comparing the graphic texture complexity P of the high-risk area with a preset value, if the graphic texture complexity P is within a preset range, the verification is successful, and if the graphic texture complexity P is greater than the preset value, the target area is proved to be similar to flame in shape but high in graphic complexity, and the target area needs to be further judged so as to ensure the accuracy of the risk grade judgment result, so that the fire point monitoring efficiency is further improved.
Specifically, when the processing module determines the secondary risk level, the processing module compares the gray value C of the target region with each preset gray value, and determines the secondary risk level according to the comparison result, wherein,
when C is less than C1, the processing module judges that the target area is a low risk area;
when C1 is less than or equal to C2, the processing module judges the target area as a medium risk area;
when C2 is less than C, the processing module judges that the target area is a low risk area;
wherein, C1 is a first preset gray value, C2 is a second preset gray value, and C1 is less than C2.
Specifically, when secondary risk level determination is performed, the processing module compares the gray value of the target area with a preset value, if the gray value is within a preset value interval, it is proved that the target area has a smoke risk, at this time, the target area is determined to be a medium risk area, and by performing the secondary risk level determination, the accuracy of the risk level determination result of each target area can be effectively improved, so that the fire point monitoring efficiency is further improved.
Specifically, the processing module performs risk area determination on the rectangular area according to the risk level of each target area in the rectangular area, wherein,
when a high risk area exists in the rectangular area, the processing module judges that the rectangular area is a risk area;
when the number of the risk areas in the rectangular area is larger than m, the processing module judges that the rectangular area is a risk area, m is the number of preset risk areas, and m is larger than or equal to 1.
Specifically, after the risk level judgment of each target area is completed, the processing module judges the risk areas of the rectangular areas according to the target area level judgment conditions in the rectangular areas so as to improve the accuracy of the risk area judgment, because the high risk areas represent that the fire has been started and the middle risk areas represent that the smoke has been caught, the processing module judges the risk levels according to the number of different risk areas in the rectangular areas, so that the accuracy of the risk area judgment can be further ensured, and the accuracy and the efficiency of fire point monitoring can be improved.
Specifically, the early warning module calculates a regional risk coefficient L0 of the fire point monitoring image according to the number of risk regions determined by the processing module, sets L0 to K1 × L + K2 × 0.1 × L, sets K1 to be the number of risk regions of the primary region, sets K2 to be the number of risk regions of the secondary region, and sets L to be a preset risk coefficient, where L is greater than 0, compares the calculated regional risk coefficient L0 with each preset regional risk coefficient, and performs early warning according to the comparison result, where,
when L0 is less than L1, the early warning module judges that the monitoring range is safe and does not perform early warning;
when L1 is not less than L0 is not less than L2, the early warning module judges that fire risks exist in the monitoring range and prompts workers to check;
when L2 is less than L0, the early warning module judges that a fire disaster occurs in the monitoring range and prompts workers to carry out fire fighting;
wherein, L1 is the preset region middle risk coefficient, L2 is the preset region high risk coefficient, and L1 is less than L2.
Specifically, when the early warning module calculates the regional risk coefficient L0, the early warning module further adjusts the regional risk coefficient L0 according to the risk target region proportion H, sets H to R/T, where R is the total area of the medium-risk and high-risk target regions, and T is the area of the fire monitoring image, compares the risk target region proportion H with the preset region proportion H0, and adjusts the regional risk coefficient L0 according to the comparison result, where,
when H is less than or equal to H0, the early warning module judges that the calculation result of the regional risk coefficient is accurate and does not adjust;
when H is larger than H0, the early warning module adjusts the regional risk coefficient to be L0 ', sets L0 ═ L0+ L0 × (H-H0)/H, and carries out early warning according to the adjusted regional risk coefficient L0'.
Specifically, after the processing module determines the risk areas of each rectangular area, the early warning module calculates an area risk coefficient L0 according to the distribution and number of the risk areas, compares the area risk coefficient L0 with a preset value, and performs early warning to different degrees according to the comparison result, so as to improve the accuracy of the early warning, thereby further improving the fire monitoring efficiency, when calculating the area risk coefficient L0, different weights are set for different levels of areas, so as to improve the accuracy of combustible monitoring for a level one area, thereby improving the fire monitoring efficiency, it can be understood that, in this embodiment, the preset values L1 and L2 are not specifically limited, but the accuracy of the early warning is not guaranteed, a person skilled in the art should set L1 to be between 0.1L and 0.3L, set L2 to be between 0.5L and 0.9L, and this range is an optimal implementation range, so as to guarantee the accuracy of the early warning, meanwhile, in the embodiment, after the regional risk coefficient L0 is calculated, the regional risk coefficient L0 is further adjusted according to the risk target region ratio H, so that the accuracy of early warning is further improved, and therefore the fire monitoring efficiency is improved, the early warning module compares the risk target region ratio H with a preset value, if the ratio is within the preset value, the adjustment is not performed, if the ratio is greater than the preset value, it is proved that the risk regions are more, the regional risk coefficient L0 needs to be increased so that the early warning is more accurate, and the early warning module adjusts the regional risk coefficient L0, so that the accuracy of early warning is further improved, and therefore the fire monitoring efficiency is improved.
Specifically, after the early warning module judges that a fire risk exists in a monitoring range, the early warning module acquires a fire monitoring image after t1 time for fire risk verification, t1 is set as first preset verification time, the early warning module calculates a regional risk coefficient L01 of the fire monitoring image after t1 time, if L01 is less than L0, verification fails without early warning, and if L01 is more than or equal to L0, verification succeeds and corresponding early warning is performed.
Specifically, after the early warning module judges that fire occurs in the monitoring range, the early warning module acquires a fire point monitoring image after t2 time for fire risk verification, t2 is set as second preset verification time, t1 is greater than t2, the early warning module calculates a regional risk coefficient L02 of the fire point monitoring image after t2 time, if L02 is less than L0, verification fails and fire risk early warning is performed, a worker is prompted to check the fire risk, and if L01 is greater than or equal to L0, verification succeeds and corresponding early warning is performed.
Specifically, in this embodiment, the early warning module is for guaranteeing the accuracy of early warning, still carries out fire risk verification through obtaining the image after the preset time, the early warning module compares the regional danger coefficient of the fire monitoring image after the preset time with the regional danger coefficient L0 of the image that needs the early warning to verify whether carry out corresponding early warning, when calculating regional danger coefficient L01 or L02 after the preset time, the same with the calculation mode of L0, in order to guarantee the validity of comparison result, the early warning module is through carrying out fire risk verification, has further improved the degree of accuracy of early warning, thereby further improves the monitoring efficiency to the fire.
So far, the technical solutions of the present invention have been described in connection with the preferred embodiments shown in the drawings, but it is easily understood by those skilled in the art that the scope of the present invention is obviously not limited to these specific embodiments. Equivalent changes or substitutions of related technical features can be made by those skilled in the art without departing from the principle of the invention, and the technical scheme after the changes or substitutions can be within the protection scope of the invention.
Claims (10)
1. A construction site fire point monitoring and early warning method based on image analysis is characterized by comprising the following steps,
step S1, acquiring a fire monitoring video file of the construction site in real time through an acquisition module;
step S2, extracting a video frame file from the video file through an extraction module, extracting video frames in the video file by taking seconds as a unit when extracting the video frames from the obtained video file, and taking the extracted video frames as fire point monitoring images;
step S3, image processing is carried out on the fire monitoring image through a processing module, when the image processing is carried out, the processing module equally divides the fire monitoring image into 3 x 3 grid areas, the central rectangular area of the grid areas is used as a primary area, other rectangular areas are used as secondary areas, the processing module divides each rectangular area according to different gray values, each area formed after division in the rectangular area is used as a target area, the processing module carries out risk area judgment on the rectangular area according to the number of the target areas in the rectangular area, and the processing module also carries out risk grade judgment on each target area so as to finish the risk area judgment on each rectangular area;
and step S4, performing fire point early warning according to the image processing result through an early warning module, performing corresponding early warning according to the regional risk coefficient L0 of the fire point monitoring image by the early warning module, and adjusting the regional risk coefficient L0 according to the risk target region ratio H.
2. The image analysis-based construction site fire monitoring and early warning method according to claim 1, wherein when the processing module performs risk region determination on a rectangular region, the processing module obtains the number n of target regions in the rectangular region and performs risk region determination on the rectangular region according to the number n of target regions, wherein,
when n is equal to 1, the processing module acquires the tone of the rectangular area, if the tone is white or red, the processing module judges that the rectangular area is a risk area, and if the tone is other colors, the processing module judges that the rectangular area is a safety area;
and when n is larger than 1, the processing module judges the risk level of the target area according to the area boundary of each target area in the rectangular area.
3. The construction site fire monitoring and early warning method based on image analysis as claimed in claim 2, wherein the processing module takes the area boundary line of each target area as a target shape curve and obtains the curvature A of the target shape curve, the processing module compares the curvature A of each target shape curve with each preset curvature and judges the risk grade of the target area according to the comparison result, wherein,
when A is less than A1, the processing module judges that the target area is a low risk area;
when A is not less than A1 and not more than A2, the processing module judges that the target area is a high risk area and carries out high risk area verification;
when A2 is less than A, the processing module carries out secondary risk grade judgment according to the gray value C of the target area;
wherein A1 is the first predetermined curvature, A2 is the second predetermined curvature, and A1 is less than A2.
4. The image analysis-based construction site fire monitoring and early warning method according to claim 3, wherein the processing module obtains the graphic texture complexity P of the high risk area during the high risk area verification, compares the graphic texture complexity P with a preset texture complexity P0, and performs the high risk area verification according to the comparison result, wherein,
when P is not more than P0, the processing module judges that the verification is successful and determines that the target area is a high risk area;
when P > P0, the processing module determines that the verification failed and makes a secondary risk level determination for the area.
5. The image analysis-based construction site fire monitoring and early warning method according to claim 4, wherein the processing module compares the gray value C of the target area with each preset gray value when performing the secondary risk level determination, and performs the secondary risk level determination according to the comparison result, wherein,
when C is less than C1, the processing module judges that the target area is a low risk area;
when C1 is less than or equal to C2, the processing module judges the target area as a medium risk area;
when C2 < C, the processing module judges that the target area is a low risk area;
wherein, C1 is a first preset gray value, C2 is a second preset gray value, and C1 is less than C2.
6. The image analysis-based worksite fire monitoring and early warning method according to claim 5, wherein the processing module performs risk region judgment on the rectangular region according to the risk level of each target region in the rectangular region, wherein,
when a high risk area exists in the rectangular area, the processing module judges that the rectangular area is a risk area;
when the number of the risk areas in the rectangular area is larger than m, the processing module judges that the rectangular area is a risk area, m is the number of preset risk areas, and m is larger than or equal to 1.
7. The image analysis-based construction site fire monitoring and early warning method according to claim 6, wherein the early warning module calculates the regional risk coefficient L0 of the fire monitoring image according to the number of risk regions determined by the processing module, sets the L0-K1 xL + K2 x0.1 xL, sets K1 as the number of risk regions of the primary region, sets K2 as the number of risk regions of the secondary region, sets L as a preset risk coefficient, and sets L > 0, compares the calculated regional risk coefficient L0 with each preset regional risk coefficient, and makes an early warning according to the comparison result, wherein,
when L0 is less than L1, the early warning module judges that the monitoring range is safe and does not perform early warning;
when L1 is not less than L0 is not less than L2, the early warning module judges that fire risks exist in the monitoring range and prompts workers to check;
when L2 is less than L0, the early warning module judges that a fire disaster occurs in the monitoring range and prompts workers to carry out fire fighting;
wherein, L1 is the medium risk coefficient of the preset area, L2 is the high risk coefficient of the preset area, and L1 is less than L2.
8. The image analysis-based construction site fire monitoring and early warning method according to claim 7, wherein the early warning module, when calculating the region risk coefficient L0, further adjusts the region risk coefficient L0 according to the risk target region proportion H, sets H to R/T, R is the total area of the regions of the medium-risk and high-risk target regions, and T is the area of the fire monitoring image, compares the risk target region proportion H with the preset region proportion H0, and adjusts the region risk coefficient L0 according to the comparison result, wherein,
when H is less than or equal to H0, the early warning module judges that the calculation result of the regional risk coefficient is accurate and does not adjust;
when H is larger than H0, the early warning module adjusts the regional risk coefficient to be L0 ', sets L0 ═ L0+ L0 × (H-H0)/H, and carries out early warning according to the adjusted regional risk coefficient L0'.
9. The construction site fire monitoring and early warning method based on image analysis as claimed in claim 8, wherein after the early warning module determines that a fire risk exists in the monitoring range, the early warning module obtains a fire monitoring image after t1 time for fire risk verification, sets t1 as a first preset verification time, calculates a regional risk coefficient L01 of the fire monitoring image after t1 time, if L01 < L0, verification fails without early warning, and if L01 is greater than or equal to L0, verification succeeds and corresponding early warning is performed.
10. The construction site fire monitoring and early warning method based on image analysis as claimed in claim 9, wherein after the early warning module determines that a fire occurs within a monitoring range, the early warning module obtains a fire monitoring image after t2 time for fire risk verification, sets t2 as second preset verification time, t1 > t2, calculates a regional risk coefficient L02 of the fire monitoring image after t2 time, if L02 < L0, the verification fails and fire risk early warning is performed, prompts a worker to check, and if L01 is greater than or equal to L0, the verification succeeds and corresponding early warning is performed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210240280.8A CN114612828B (en) | 2022-03-10 | 2022-03-10 | Construction site fire monitoring and early warning method based on image analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210240280.8A CN114612828B (en) | 2022-03-10 | 2022-03-10 | Construction site fire monitoring and early warning method based on image analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114612828A true CN114612828A (en) | 2022-06-10 |
| CN114612828B CN114612828B (en) | 2022-12-02 |
Family
ID=81862157
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210240280.8A Active CN114612828B (en) | 2022-03-10 | 2022-03-10 | Construction site fire monitoring and early warning method based on image analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114612828B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114119491A (en) * | 2021-10-29 | 2022-03-01 | 吉林医药学院 | Data processing system based on medical image analysis |
| CN115376268A (en) * | 2022-10-21 | 2022-11-22 | 山东太平天下智慧科技有限公司 | Monitoring alarm fire-fighting linkage system based on image recognition |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH09139870A (en) * | 1995-11-13 | 1997-05-27 | Nippon Dry Chem Co Ltd | Image distortion corrector and fire monitoring and firefighting system |
| CN108470424A (en) * | 2018-03-06 | 2018-08-31 | 深圳森阳环保材料科技有限公司 | A kind of forest safety monitoring system based on characteristics of image |
| CN109767591A (en) * | 2019-03-08 | 2019-05-17 | 郭弋硙 | A kind of forest fireproofing early warning system and method |
| US20200348446A1 (en) * | 2019-05-05 | 2020-11-05 | Vasily Antonovich Tremsin | Early-Warning Fire Detection System Based on a Multivariable Approach |
| CN113408495A (en) * | 2021-07-30 | 2021-09-17 | 广州汇图计算机信息技术有限公司 | Safety guard system for security |
| CN113505758A (en) * | 2021-09-07 | 2021-10-15 | 广州汇图计算机信息技术有限公司 | Smog early warning system based on visual detection |
| CN114022944A (en) * | 2022-01-05 | 2022-02-08 | 北京国信网联科技有限公司 | Intelligent monitoring system |
| CN114118682A (en) * | 2021-10-15 | 2022-03-01 | 山东联安信息科技有限责任公司 | Hydrogen-oil-ammonia-region intelligent monitoring and early warning and risk hidden danger identification method based on 5G technology |
-
2022
- 2022-03-10 CN CN202210240280.8A patent/CN114612828B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH09139870A (en) * | 1995-11-13 | 1997-05-27 | Nippon Dry Chem Co Ltd | Image distortion corrector and fire monitoring and firefighting system |
| CN108470424A (en) * | 2018-03-06 | 2018-08-31 | 深圳森阳环保材料科技有限公司 | A kind of forest safety monitoring system based on characteristics of image |
| CN109767591A (en) * | 2019-03-08 | 2019-05-17 | 郭弋硙 | A kind of forest fireproofing early warning system and method |
| US20200348446A1 (en) * | 2019-05-05 | 2020-11-05 | Vasily Antonovich Tremsin | Early-Warning Fire Detection System Based on a Multivariable Approach |
| CN113408495A (en) * | 2021-07-30 | 2021-09-17 | 广州汇图计算机信息技术有限公司 | Safety guard system for security |
| CN113505758A (en) * | 2021-09-07 | 2021-10-15 | 广州汇图计算机信息技术有限公司 | Smog early warning system based on visual detection |
| CN114118682A (en) * | 2021-10-15 | 2022-03-01 | 山东联安信息科技有限责任公司 | Hydrogen-oil-ammonia-region intelligent monitoring and early warning and risk hidden danger identification method based on 5G technology |
| CN114022944A (en) * | 2022-01-05 | 2022-02-08 | 北京国信网联科技有限公司 | Intelligent monitoring system |
Non-Patent Citations (2)
| Title |
|---|
| HA DAI DUONG ET AL: "A Novel Computational Approach for Fire Detection", 《2010 SECOND INTERNATIONAL CONFERENCE ON KNOWLEDGE AND SYSTEMS ENGINEERING》 * |
| 李玲: "基于视频图像的火灾识别研究", 《中国优秀博硕士学位论文全文数据库(硕士) 信息科技辑》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114119491A (en) * | 2021-10-29 | 2022-03-01 | 吉林医药学院 | Data processing system based on medical image analysis |
| CN114119491B (en) * | 2021-10-29 | 2022-09-13 | 吉林医药学院 | Data processing system based on medical image analysis |
| CN115376268A (en) * | 2022-10-21 | 2022-11-22 | 山东太平天下智慧科技有限公司 | Monitoring alarm fire-fighting linkage system based on image recognition |
| CN115376268B (en) * | 2022-10-21 | 2023-02-28 | 山东太平天下智慧科技有限公司 | Monitoring alarm fire-fighting linkage system based on image recognition |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114612828B (en) | 2022-12-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114612828B (en) | Construction site fire monitoring and early warning method based on image analysis | |
| US8442263B2 (en) | Quantification of energy loss from buildings | |
| CN104850693A (en) | Monitoring equipment layout method and device | |
| CN113505758B (en) | Smog early warning system based on visual detection | |
| CN115565330A (en) | Building construction site spot-distribution type fire monitoring system and method | |
| CN112927461B (en) | Early warning decision method and device for charging pile of new energy automobile | |
| CN103425959B (en) | Flame video detection method for identifying fire hazard | |
| CN104780361A (en) | Quality evaluation method for urban video monitoring system | |
| CN115077489B (en) | Unmanned aerial vehicle oblique photography earthwork volume calculation method | |
| CN109389333A (en) | The methods of risk assessment and device of power equipment | |
| US20180197504A1 (en) | Gamma Adjustment Method | |
| CN110610309A (en) | Subway station real-time monitoring method and emergency treatment rehearsal method based on three-dimensional structure | |
| CN112288692A (en) | Crack monitoring and management system and monitoring and management method | |
| KR20200026613A (en) | Indoor air quality evaluation system | |
| CN106412575A (en) | Method and device for detecting display | |
| CN116989679A (en) | Highway high slope displacement integrated monitoring method based on image processing | |
| CN110443316A (en) | Board turning recognition methods, device and the distribution network failure of fault detector monitor system | |
| KR102039048B1 (en) | Apparatus and method for generating of simulated satellite images | |
| US9997038B2 (en) | Smoke detection apparatus, method for detecting smoke and computer program | |
| CN116090838B (en) | Automatic building site system of patrolling and examining based on camera | |
| CN108681702A (en) | A kind of container handling stowage shellfish position information determines method and system | |
| KR20180114409A (en) | Energy preformance simulating system for existing building | |
| CN105141904B (en) | Security monitoring and personal identification method and device | |
| CN110131855A (en) | The Intelligentized control method and device of indoor air conditioner | |
| CN110543824A (en) | construction safety judgment method based on binocular vision |
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 |