CN113298048B - Forest fire detection and early warning system based on computer vision - Google Patents
Forest fire detection and early warning system based on computer vision Download PDFInfo
- Publication number
- CN113298048B CN113298048B CN202110770087.0A CN202110770087A CN113298048B CN 113298048 B CN113298048 B CN 113298048B CN 202110770087 A CN202110770087 A CN 202110770087A CN 113298048 B CN113298048 B CN 113298048B
- Authority
- CN
- China
- Prior art keywords
- module
- picture
- forest
- flame
- early warning
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 20
- 238000004458 analytical method Methods 0.000 claims abstract description 33
- 238000004891 communication Methods 0.000 claims abstract description 11
- 238000012544 monitoring process Methods 0.000 claims description 11
- 239000006185 dispersion Substances 0.000 claims description 8
- 238000007906 compression Methods 0.000 claims description 5
- 238000012800 visualization Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000011161 development Methods 0.000 claims description 3
- 230000008635 plant growth Effects 0.000 claims description 3
- 230000001932 seasonal effect Effects 0.000 claims description 3
- 238000000034 method Methods 0.000 description 14
- 238000005070 sampling Methods 0.000 description 10
- 239000011159 matrix material Substances 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 238000012545 processing Methods 0.000 description 4
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 230000006378 damage Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000012502 risk assessment Methods 0.000 description 2
- 241001270131 Agaricus moelleri Species 0.000 description 1
- 241001465754 Metazoa Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 230000001066 destructive effect Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 239000012535 impurity Substances 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 239000000779 smoke Substances 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/188—Vegetation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/22—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B17/00—Fire alarms; Alarms responsive to explosion
- G08B17/005—Fire alarms; Alarms responsive to explosion for forest fires, e.g. detecting fires spread over a large or outdoors area
-
- 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 provides a forest fire detection and early warning system based on computer vision, which comprises a central service station, a video acquisition module, a video storage module, a picture comparison module, an acquisition control module, a depth analysis module and a communication module, wherein the video acquisition module is used for acquiring image data around the forest early warning station, the video storage module is used for storing a comparison picture group, the picture comparison module is used for comparing the comparison picture group with real-time image data to obtain a suspicious region, the acquisition control module controls the video acquisition module to acquire a detail picture of the suspicious region, the depth analysis module is used for carrying out flame identification on the detail picture, and the communication module is used for sending the image data to the central service station after the flame is identified. The system divides the flame detection into two levels of image contrast and depth analysis, can accurately detect the initial fire under limited computing resources, and reduces the loss caused by fire.
Description
Technical Field
The invention relates to the technical field of forest fire detection, in particular to a forest fire detection and early warning system based on computer vision.
Background
Forest fires are the most dangerous enemies of forests and the most terrible disasters of forests, can bring the most harm to the forests and have destructive consequences, the forest fires not only burn out a large number of forests and damage animals in the forests, but also reduce the updating capacity of the forests, cause the soil to be barren and destroy the forest water conservation source, even cause the ecological environment to be unbalanced, and how to effectively identify the fire behavior at the initial stage of the fire, so that the deployment is carried out as soon as possible to inactivate, and the problems faced by the current fire early warning systems when the loss caused by the fire is reduced.
A plurality of fire early warning systems are developed at present, and through a large number of searches and references, the existing early warning systems are found to be disclosed as KR101146474B1, KR101085975B1, CN101719298B and KR101532055B1, the information expressive force of satellite remote sensing data on a large-scale space is utilized, the remote sensing data is used as the basis, fire risk assessment on a forest grassland ecological system in a large space range and detection on abnormal high-temperature points of the forest grassland are included, and the forest grassland fire remote sensing monitoring early warning information in point-to-point combination is generated through assimilation processing of the detected data of the abnormal high-temperature points and the fire risk assessment data, so that effective decision indication information is provided for fire prevention and fire emergency reflection of the forest grassland. However, the system adopts a remote sensing technology, although the detection range is wide, the system cannot detect small-scale fire, and the early warning efficiency is not high enough, so that fire fighters cannot extinguish the fire at an early stage.
Disclosure of Invention
The invention aims to provide a forest fire detection and early warning system based on computer vision aiming at the existing defects,
in order to overcome the defects of the prior art, the invention adopts the following technical scheme:
a forest fire detection and early warning system based on computer vision is characterized by comprising a plurality of forest early warning stations and a central service station, wherein the forest early warning stations are used for collecting image data of a front line and carrying out fire analysis, and the central service station is used for overall planning the data of all the forest early warning stations and visualization of a fire map;
the forest early warning station comprises a video acquisition module, a video storage module, a picture comparison module, an acquisition control module, a depth analysis module and a communication module, wherein the video acquisition module is used for acquiring image data around the forest early warning station, the video storage module is used for storing a comparison picture group, the picture comparison module is used for comparing the comparison picture group with real-time image data to obtain a suspicious region, the acquisition control module controls the video acquisition module to acquire a detail picture of the suspicious region, the depth analysis module performs flame identification on the detail picture, and the communication module transmits the image data to the central service station after the flame is identified;
the picture comparison module selects a comparison picture closest to the real-time image from the comparison picture group according to the pattern, and the calculation mode of the pattern is as follows:
obtaining gray level features g according to pixel points of the picture, and obtaining a gray level feature sequence { g ] from the whole picture in a fixed sequenceiAnd i is an element serial number, and the gray characteristic sequence is compressed according to the following modes:
wherein j is the element serial number of the compressed gray level feature sequence;
continuously repeating the compression process to enable the number of elements of the final gray characteristic sequence to be within a preset range, wherein the gray characteristic sequence is the pattern of the picture;
the depth analysis module retrieves suspected flame pixel points from two continuous detail graphs, the overlapped suspected flame pixel points are set { H1}, the non-overlapped suspected flame pixel points are set { H2}, and the dispersion Q is calculated for the sets { H1} and { H2} respectively:
wherein N is the total number of the pixel points in the set, and (x, y) is the coordinate of the pixel points in the set,
the coordinate average value of all pixel points in the set is obtained;
the depth analysis module calculates a flame index Y:
wherein N1 is the number of the pixels in the set { H1}, N2 is the number of the pixels in the set { H2}, and r is a proportional cardinality;
the depth analysis module identifies a flame when the flame index exceeds a threshold;
furthermore, the picture acquired by the video acquisition module is provided with picture attributes, the picture attributes comprise a horizontal angle, a pitching angle, a focal length, a weather condition and shooting time, and the comparison picture group comprises pictures with different attributes;
further, the acquisition control module is provided with a patrol mode and a monitoring mode, in the patrol mode, the video acquisition module shoots videos at different horizontal angles and different pitching angles at a fixed speed, and in the monitoring mode, the video acquisition module shoots videos at a fixed horizontal angle, a fixed pitching angle and a fixed focal length;
furthermore, the system also comprises a contrast replacement module, when the picture information acquired by the video acquisition module is different from the picture information in the contrast picture group due to seasonal variation and plant growth, the contrast replacement module intercepts a picture from the video shot by the video acquisition module to replace the original contrast picture group;
further, after one of the forest early warning stations detects flame information, the central service station sends an instruction to the forest early warning stations around the flame point, so that the central service station is in a monitoring mode to monitor subsequent fire development and upload image data.
The beneficial effects obtained by the invention are as follows:
the system divides the detection of the fire into two levels, the common image contrast and the detailed image analysis are performed, the real-time image and the contrast image are compared through the common image contrast, subtle differences can be found, the contrast image contains images with different attributes, the occurrence rate of false detection is reduced, the detail images of the subtle differences are obtained, whether flame exists in deep analysis or not is determined, the efficiency of detecting the fire can be greatly improved, workers can make processing as early as possible, and the loss caused by the fire is reduced.
Drawings
The invention will be further understood from the following description in conjunction with the accompanying drawings. The components in the figures are not necessarily to scale, emphasis instead being placed upon illustrating the principles of the embodiments. Like reference numerals designate corresponding parts throughout the different views.
Fig. 1 is a schematic view of an overall structural framework.
Fig. 2 is a schematic diagram of a fire detection and early warning process.
Fig. 3 is a schematic diagram of the working flow of the webcam in the patrol mode.
Fig. 4 is a schematic diagram of sampling in patrol mode.
Fig. 5 is a schematic diagram of a suspicious region flow for obtaining a sampled picture.
Detailed Description
In order to make the objects and advantages of the present invention more apparent, the present invention will be further described in detail with reference to the following embodiments; it should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. Other systems, methods, and/or features of the present embodiments will become apparent to those skilled in the art upon review of the following detailed description. It is intended that all such additional systems, methods, features and advantages be included within this description, be within the scope of the invention, and be protected by the accompanying claims. Additional features of the disclosed embodiments are described in, and will be apparent from, the detailed description that follows.
The same or similar reference numerals in the drawings of the embodiments of the present invention correspond to the same or similar components; in the description of the present invention, it should be understood that if there is an orientation or positional relationship indicated by terms such as "upper", "lower", "left", "right", etc., based on the orientation or positional relationship shown in the drawings, it is only for convenience of description and simplification of description, but it is not indicated or implied that the device or component referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore, the terms describing the positional relationship in the drawings are only used for illustrative purposes and are not to be construed as limitations of the present patent, and specific meanings of the terms may be understood by those skilled in the art according to specific situations.
The first embodiment.
With reference to fig. 1 and fig. 2, the present embodiment provides a forest fire detection and early warning system based on computer vision, which includes a plurality of forest early warning stations and a central service station, where the forest early warning stations are configured to collect image data of a front line and perform fire analysis, and the central service station is configured to pool data of all forest early warning stations and visualization of a fire map;
the forest early warning station comprises a video acquisition module, a video storage module, a picture comparison module, an acquisition control module, a depth analysis module and a communication module, wherein the video acquisition module is used for acquiring image data around the forest early warning station, the video storage module is used for storing a comparison picture group, the picture comparison module is used for comparing the comparison picture group with real-time image data to obtain a suspicious region, the acquisition control module controls the video acquisition module to acquire a detail picture of the suspicious region, the depth analysis module performs flame identification on the detail picture, and the communication module transmits the image data to the central service station after the flame is identified;
the picture comparison module selects a comparison picture closest to the real-time image from the comparison picture group according to the pattern, and the calculation mode of the pattern is as follows:
obtaining gray level features g according to pixel points of the picture, and obtaining a gray level feature sequence { g ] from the whole picture in a fixed sequenceiAnd i is an element serial number, and the gray characteristic sequence is compressed according to the following modes:
wherein j is the element serial number of the compressed gray level feature sequence;
continuously repeating the compression process to enable the number of elements of the final gray characteristic sequence to be within a preset range, wherein the gray characteristic sequence is the pattern of the picture;
the depth analysis module retrieves suspected flame pixel points from two continuous detail graphs, the overlapped suspected flame pixel points are set { H1}, the non-overlapped suspected flame pixel points are set { H2}, and the dispersion Q is calculated for the sets { H1} and { H2} respectively:
wherein N is the total number of the pixel points in the set, and (x, y) is the coordinate of the pixel points in the set,
the coordinate average value of all pixel points in the set is obtained;
the depth analysis module calculates a flame index Y:
wherein N1 is the number of the pixels in the set { H1}, N2 is the number of the pixels in the set { H2}, and r is a proportional cardinality;
the depth analysis module identifies a flame when the flame index exceeds a threshold;
the image acquired by the video acquisition module is provided with image attributes, the image attributes comprise a horizontal angle, a pitching angle, a focal length, weather conditions and shooting time, and the comparison image group comprises images with different attributes;
the acquisition control module is provided with a patrol mode and a monitoring mode, wherein in the patrol mode, the video acquisition module shoots videos at different horizontal angles and different pitching angles at a fixed speed, and in the monitoring mode, the video acquisition module shoots videos at a fixed horizontal angle, a fixed pitching angle and a fixed focal length;
the system also comprises a contrast replacement module, wherein when picture information acquired by the video acquisition module is different from picture information in the contrast picture group due to seasonal variation and plant growth, the contrast replacement module intercepts a picture from a video shot by the video acquisition module to replace the original contrast picture group;
and when one of the forest early warning stations detects flame information, the central service station sends an instruction to the forest early warning stations around the flame point, so that the central service station is in a monitoring mode to monitor the subsequent fire development and upload image data.
Example two.
The embodiment includes the whole content in the first embodiment, and provides a forest fire detection and early warning system based on computer vision, which comprises a plurality of forest early warning stations and a central service station distributed at all places of a forest, the forest early warning station comprises a video acquisition module, a video storage module, a picture comparison module, an acquisition control module, a depth analysis module and a communication module, the video acquisition module is arranged in a network camera, the video storage module, the picture comparison module, the acquisition control module, the depth analysis module and the communication module are arranged in a computer, the network camera is used for acquiring peripheral picture data, the computer is used for analyzing the picture data, controlling the shooting state of the network camera and communicating with the central service station, and the central service station is used for inquiring the visualization of a shot video and a fire map;
the network camera has a 360-degree horizontal rotation function, a pitch angle function and a focusing function, the network camera takes pictures at different horizontal angles, pitch angles and focal lengths and stores the pictures as comparison pictures in the video storage module, the attribute of each comparison picture comprises the three shooting states of the network camera, the set of the comparison pictures forms a comparison picture group, the picture content of the comparison picture group comprises all scenes around the network camera, the attribute of the comparison picture also comprises the weather condition and the shooting time, and when the attributes of the current shot picture and the comparison picture group are different and the pictures are different, the current shot picture is added into the comparison picture group;
with reference to fig. 3, the acquisition control module is provided with a patrol mode, in which the webcam horizontally rotates 360 degrees at a fixed elevation angle, then horizontally rotates 360 degrees at a head-up state, and finally horizontally rotates 360 degrees at a fixed depression angle, and the three processes are repeated continuously, the webcam horizontally rotates at a fixed patrol speed v during the horizontal rotation, but the rotation speed of the webcam is 0 when the elevation angle is adjusted, and the acquisition control module is further provided with a fire monitoring mode in which the webcam photographs videos of the same scene in a fixed state;
with reference to fig. 4, the image comparison module intercepts an image from a video shot by a webcam in a patrol mode as a sampling image at a sampling frequency f, wherein the sampling frequency satisfies
The sampling image in each rotation period can not be divided by 360 degrees, so that the sampling image is in different horizontal angle attributes, the sampling image is compared with the image with similar attributes in a comparison image group, if the comparison result shows that a suspicious region is different from the comparison image in the sampling image, the acquisition control module is in a pause patrol mode, the horizontal angle, the pitch angle and the focal length of the network camera are set to be in a state capable of acquiring a detailed image of the suspicious region, and the network camera shoots a detailed image in the state and sends the detailed image to the depth analysis module;
the depth analysis module is internally provided with a flame algorithm used for identifying flames in the detail map, if the flames are detected, the communication module sends a fire alarm to the central service station, meanwhile, video pictures shot by the network camera are continuously sent to the central service station, the central service station sends information to other forest early warning stations nearby a fire point, the forest early warning stations shoot the surrounding situation of the fire point in a fire monitoring mode and continuously send the shot videos to the central service station, workers carry out rescue deployment according to the fire information of the forest early warning stations collected by the central service station, and if the flames are not detected by the depth analysis module, the acquisition control module resets the network camera to a patrol mode.
Example three.
This embodiment includes all the contents in the above embodiments, the horizontal angle attributes of the comparison picture group in this embodiment are 0 °, 60 °, 120 °, 180 °, 240 °, and 300 °, and the horizontal angle attributes of the sampling picture are
The picture comparison module divides the sampling picture into a left part and a right part P1 and P2, and the width ratio of the pictures P1 and P2
Comprises the following steps:
wherein the content of the first and second substances,
and
to compare two adjacent horizontal angle attributes in the picture group and satisfy
When needed, 0 degree is written as 360 degree;
selecting a horizontal angle from the map picture group as
The reference picture of (A) constitutes a set A1, the horizontal angle is selected to be
The comparison pictures in a1 and a2 form a set a2, and it should be noted that the comparison pictures in a1 and a2 have the same pitch angle attribute as the sampled pictures, the picture comparison module compares the picture P1 with the set a1, and the picture P2 with the set a 2;
with reference to fig. 5, the following description is directed to a method for comparing the picture P1 with the set a1, including the following steps;
s1, cutting and reserving the pictures in the set A1 as a part corresponding to the picture P1;
s2, calculating the patterns of the picture P1 and the reference picture in the set A1;
s3, selecting the comparison picture B which is closest to the picture P1 in the set A1;
s4, comparing the picture P1 with the control picture B to obtain a suspicious region;
the method for calculating the pattern of the picture in step S2 is as follows:
converting a picture into a 256-color gray scale map, dividing the gray scale value into 16 levels, wherein each level comprises 16 continuous gray scales, taking a 4 x 4 pixel block as a statistical unit, taking the gray scale with the maximum number of pixel points in the pixel block as a gray scale feature g of the pixel block, and performing statistical processing on the picture according to the sequence of from left to right and then from top to bottom to obtain a gray scale feature sequence { giAnd if the right side or the lower side of the picture is not full of a pixel block, directly omitting pixel points on the right side or the lower side, and compressing the gray feature sequence, wherein a compression formula is as follows:
wherein j is the element serial number of the compressed gray level feature sequence;
continuously repeating the compression process to make the number of elements of the final gray level feature sequence between 64 and 128, wherein the gray level feature sequence is the pattern of the picture;
the method for comparing the two pictures to obtain the suspicious region in step S4 is as follows:
subtracting the gray scale image of the picture P1 from the gray scale image of the comparison picture B to obtain a difference matrix C of m x n, wherein m is the number of pixels in the horizontal direction of the picture P1, n is the number of pixels in the vertical direction of the picture P1, the value of each element in the difference matrix C is an integer and comprises a positive number, a negative number and 0, and the value with the largest number of elements in the matrix C is counted and recorded as the value
And then, counting suspicious point elements, wherein the value a of the suspicious point elements meets the following requirements:
wherein k is a suspicious threshold;
performing impurity removal processing on the suspicious points, specifically, acquiring a 5 x 5 matrix taking each suspicious point as a center, and removing the suspicious points for statistics if the number of the suspicious points in the matrix is less than 10;
partitioning the suspicious points, if two suspicious points are in a 5 x 5 matrix, enabling the two suspicious points to be in a connected state, and taking all the connected suspicious points as a set to obtain a plurality of suspicious point sets;
and finally confirming the suspicious point set, wherein when the number of the suspicious points in the suspicious point set exceeds a threshold value, a circumscribed rectangle of the suspicious point set is used as a suspicious region.
Example four.
The present embodiment includes all the contents in the above embodiments, where the depth analysis module in the present embodiment obtains a group of detail maps from the network camera, and performs flame algorithm and smoke algorithm analysis on the detail maps, where the flame algorithm analysis process is as follows;
the depth analysis module extracts suspected flame pixel points in the detail graph, wherein the suspected flame pixel points meet the following requirements:
wherein R is the red component of the pixel, G is the green component of the pixel, B is the blue component of the pixel, and is the red component threshold of the pixel;
comparing suspected flame pixel points in 2 continuous detail graphs, taking the pixel points with coincident positions as a set { H1}, taking the pixel points with non-coincident positions as a set { H2}, and calculating dispersion Q of the sets { H1} and { H2} respectively:
wherein N is the total number of the pixel points in the set, and (x, y) is the coordinate of each pixel point in the set,
the coordinate average value of all pixel points in the set is obtained;
the dispersion of the set { H1} is represented by Q1, the dispersion of the set { H2} is represented by Q2, and the flame index Y of the continuous 2-piece detail diagram is calculated by the depth analysis module:
wherein N1 is the number of the pixels in the set { H1}, N2 is the number of the pixels in the set { H2}, and r is a proportional cardinality;
the depth analysis module considers a flame to be identified in the detail map when the flame index exceeds a threshold.
Although the invention has been described above with reference to various embodiments, it should be understood that many changes and modifications may be made without departing from the scope of the invention. That is, the methods, systems, and devices discussed above are examples. Various configurations may omit, substitute, or add various procedures or components as appropriate. For example, in alternative configurations, the methods may be performed in an order different than that described, and/or various components may be added, omitted, and/or combined. Moreover, features described with respect to certain configurations may be combined in various other configurations, as different aspects and elements of the configurations may be combined in a similar manner. Further, elements therein may be updated as technology evolves, i.e., many elements are examples and do not limit the scope of the disclosure or claims.
Specific details are given in the description to provide a thorough understanding of the exemplary configurations including implementations. However, configurations may be practiced without these specific details, for example, well-known circuits, processes, algorithms, structures, and techniques have been shown without unnecessary detail in order to avoid obscuring the configurations. This description provides example configurations only, and does not limit the scope, applicability, or configuration of the claims. Rather, the foregoing description of the configurations will provide those skilled in the art with an enabling description for implementing the described techniques. Various changes may be made in the function and arrangement of elements without departing from the spirit or scope of the disclosure.
In conclusion, it is intended that the foregoing detailed description be regarded as illustrative rather than limiting, and that it be understood that these examples are illustrative only and are not intended to limit the scope of the invention. After reading the description of the invention, the skilled person can make various changes or modifications to the invention, and these equivalent changes and modifications also fall into the scope of the invention defined by the claims.
Claims (5)
1. A forest fire detection and early warning system based on computer vision is characterized by comprising a plurality of forest early warning stations and a central service station, wherein the forest early warning stations are used for collecting image data of a front line and carrying out fire analysis, and the central service station is used for overall planning the data of all the forest early warning stations and visualization of a fire map;
the forest early warning station comprises a video acquisition module, a video storage module, a picture comparison module, an acquisition control module, a depth analysis module and a communication module, wherein the video acquisition module is used for acquiring image data around the forest early warning station, the video storage module is used for storing a comparison picture group, the picture comparison module is used for comparing the comparison picture group with real-time image data to obtain a suspicious region, the acquisition control module controls the video acquisition module to acquire a detail picture of the suspicious region, the depth analysis module performs flame identification on the detail picture, and the communication module transmits the image data to the central service station after the flame is identified;
the picture comparison module selects a comparison picture closest to the real-time image from the comparison picture group according to the pattern, and the calculation mode of the pattern is as follows:
obtaining gray level features g according to pixel points of the picture, and obtaining a gray level feature sequence { g ] from the whole picture in a fixed sequenceiAnd i is an element serial number, and the gray characteristic sequence is compressed according to the following modes:
wherein j is the element serial number of the compressed gray level feature sequence;
continuously repeating the compression process to enable the number of elements of the final gray characteristic sequence to be within a preset range, wherein the gray characteristic sequence is the pattern of the picture;
the depth analysis module retrieves suspected flame pixel points from two continuous detail images, wherein the suspected flame pixel points meet the following requirements:
wherein R is the red component of the pixel, G is the green component of the pixel, B is the blue component of the pixel,
a red component threshold for a pixel;
comparing suspected flame pixel points in two continuous detail graphs, taking the pixel points with coincident positions as a set { H1}, taking the pixel points with non-coincident positions as a set { H2}, and calculating dispersion Q of the sets { H1} and { H2} respectively:
wherein N is the total number of the pixel points in the set, and (x, y) is the coordinate of the pixel points in the set,
the coordinate average value of all pixel points in the set is obtained;
the depth analysis module calculates a flame index Y:
wherein N1 is the number of the pixels in the set { H1}, N2 is the number of the pixels in the set { H2}, and r is a proportional cardinality; q1 represents the dispersion of the set { H1}, Q2 represents the dispersion of the set { H2 };
the depth analysis module identifies a flame when the flame index exceeds a threshold.
2. A forest fire detection and pre-warning system based on computer vision as claimed in claim 1, wherein the picture obtained by the video capture module has picture attributes, the picture attributes include horizontal angle, pitch angle, focal length, weather condition and shooting time, and the comparison picture group includes pictures with different attributes.
3. A forest fire detection and early warning system based on computer vision as claimed in claim 2, wherein the collection control module is provided with a patrol mode in which the video collection module shoots videos at different horizontal angles and different elevation angles at a fixed speed, and a monitor mode in which the video collection module shoots videos at a fixed horizontal angle, elevation angle and focal length.
4. A forest fire detection and early warning system based on computer vision as claimed in claim 3, wherein said system further comprises a contrast replacement module, when the picture information obtained by the video capture module is different from the picture information in the contrast group due to seasonal changes and plant growth, said contrast replacement module cuts out the picture from the video captured by said video capture module to replace the original contrast group.
5. A forest fire detection and pre-warning system based on computer vision as claimed in claim 4, characterised in that after one of the forest pre-warning stations detects flame information, the central service station sends instructions to forest pre-warning stations around the flame point to enable them to be in a monitoring mode for monitoring the subsequent development of fire and uploading image data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110770087.0A CN113298048B (en) | 2021-07-08 | 2021-07-08 | Forest fire detection and early warning system based on computer vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110770087.0A CN113298048B (en) | 2021-07-08 | 2021-07-08 | Forest fire detection and early warning system based on computer vision |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113298048A CN113298048A (en) | 2021-08-24 |
| CN113298048B true CN113298048B (en) | 2021-11-02 |
Family
ID=77330598
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110770087.0A Active CN113298048B (en) | 2021-07-08 | 2021-07-08 | Forest fire detection and early warning system based on computer vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113298048B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103605958A (en) * | 2013-11-12 | 2014-02-26 | 北京工业大学 | Living body human face detection method based on gray scale symbiosis matrixes and wavelet analysis |
| CN106373320A (en) * | 2016-08-22 | 2017-02-01 | 中国人民解放军海军工程大学 | Fire identification method based on flame color dispersion and continuous frame image similarity |
| CN106778582A (en) * | 2016-12-07 | 2017-05-31 | 哈尔滨工业大学 | Flame/smog recognition methods after forest map picture cutting based on RGB reconstruct |
| CN106934404A (en) * | 2017-03-10 | 2017-07-07 | 深圳市瀚晖威视科技有限公司 | A kind of image flame identifying system based on CNN convolutional neural networks |
| CN108877127A (en) * | 2018-04-27 | 2018-11-23 | 西安科技大学 | Forest fire detection system and method based on image procossing |
| CN109165577A (en) * | 2018-08-07 | 2019-01-08 | 东北大学 | A kind of early stage forest fire detection method based on video image |
| CN111582698A (en) * | 2020-04-29 | 2020-08-25 | 国电科学技术研究院有限公司 | Combustion stability evaluation index calculation method based on hearth outlet temperature |
| CN111914818A (en) * | 2020-09-21 | 2020-11-10 | 北京林业大学 | Forest fire smoke root node detection method based on multi-frame discrete confidence |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102142085B (en) * | 2011-05-11 | 2012-11-21 | 武汉大学 | Robust tracking method for moving flame target in forest region monitoring video |
| CN111047818A (en) * | 2019-11-01 | 2020-04-21 | 浙江省林业技术推广总站(浙江省林业信息宣传中心) | Forest fire early warning system based on video image |
-
2021
- 2021-07-08 CN CN202110770087.0A patent/CN113298048B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103605958A (en) * | 2013-11-12 | 2014-02-26 | 北京工业大学 | Living body human face detection method based on gray scale symbiosis matrixes and wavelet analysis |
| CN106373320A (en) * | 2016-08-22 | 2017-02-01 | 中国人民解放军海军工程大学 | Fire identification method based on flame color dispersion and continuous frame image similarity |
| CN106778582A (en) * | 2016-12-07 | 2017-05-31 | 哈尔滨工业大学 | Flame/smog recognition methods after forest map picture cutting based on RGB reconstruct |
| CN106934404A (en) * | 2017-03-10 | 2017-07-07 | 深圳市瀚晖威视科技有限公司 | A kind of image flame identifying system based on CNN convolutional neural networks |
| CN108877127A (en) * | 2018-04-27 | 2018-11-23 | 西安科技大学 | Forest fire detection system and method based on image procossing |
| CN109165577A (en) * | 2018-08-07 | 2019-01-08 | 东北大学 | A kind of early stage forest fire detection method based on video image |
| CN111582698A (en) * | 2020-04-29 | 2020-08-25 | 国电科学技术研究院有限公司 | Combustion stability evaluation index calculation method based on hearth outlet temperature |
| CN111914818A (en) * | 2020-09-21 | 2020-11-10 | 北京林业大学 | Forest fire smoke root node detection method based on multi-frame discrete confidence |
Non-Patent Citations (1)
| Title |
|---|
| A new fire detection method using a multi-expert system based on color dispersion, similarity and centroid motion in indoor environment;Teng Wang;《IEEE/CAA JOURNAL OF AUTOMATICA SINICA》;20190619;第7卷(第1期);263-275 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113298048A (en) | 2021-08-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107911653B (en) | Intelligent video monitoring module, system, method and storage medium for residence | |
| CN108389359B (en) | Deep learning-based urban fire alarm method | |
| CN108319926A (en) | A kind of the safety cap wearing detecting system and detection method of building-site | |
| CN111932709A (en) | Method for realizing violation safety supervision of inspection operation of gas station based on AI identification | |
| KR102149832B1 (en) | Automated Violence Detecting System based on Deep Learning | |
| CN103106766A (en) | Forest fire identification method and forest fire identification system | |
| KR102407327B1 (en) | Apparatus for Monitoring Fire And System having the same | |
| CN114821406A (en) | Method and system for judging electric power operation field violation behaviors | |
| CN111899452A (en) | Forest fire prevention early warning system based on edge calculation | |
| CN112084963B (en) | Monitoring early warning method, system and storage medium | |
| CN110473375A (en) | Monitoring method, device, equipment and the system of forest fire | |
| CN111178424A (en) | Petrochemical production site safety compliance real-time detection system and method | |
| CN116743970B (en) | Intelligent management platform with video AI early warning analysis | |
| CN116189103B (en) | Equipment monitoring automatic control method and control system based on visual image analysis | |
| JP3486229B2 (en) | Image change detection device | |
| CN113298048B (en) | Forest fire detection and early warning system based on computer vision | |
| KR101270718B1 (en) | Video processing apparatus and method for detecting fire from video | |
| CN117037065A (en) | Flame smoke concentration detection method, device, computer equipment and storage medium | |
| CN111582183A (en) | Mask identification method and system in public place | |
| CN116682162A (en) | Robot detection algorithm based on real-time video stream | |
| CN114612771A (en) | Fire source monitoring method and system based on neural network | |
| CN113627321A (en) | Image identification method and device based on artificial intelligence and computer equipment | |
| JP2005166054A (en) | Moving video processing system, and video monitoring system | |
| JP2022120480A (en) | Analysis system, analysis device, analysis method, and program | |
| KR20240040530A (en) | System for infectious disease prevention based on deep learning |
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 |