CN118280054B - Forest fire monitoring and early warning system based on unmanned aerial vehicle image analysis - Google Patents
Forest fire monitoring and early warning system based on unmanned aerial vehicle image analysis Download PDFInfo
- Publication number
- CN118280054B CN118280054B CN202410712138.8A CN202410712138A CN118280054B CN 118280054 B CN118280054 B CN 118280054B CN 202410712138 A CN202410712138 A CN 202410712138A CN 118280054 B CN118280054 B CN 118280054B
- Authority
- CN
- China
- Prior art keywords
- fire
- area
- monitoring
- value
- smoke
- 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
- 238000012544 monitoring process Methods 0.000 claims abstract description 136
- 239000000779 smoke Substances 0.000 claims abstract description 84
- 238000002485 combustion reaction Methods 0.000 claims abstract description 68
- 238000011156 evaluation Methods 0.000 claims abstract description 64
- 238000004458 analytical method Methods 0.000 claims abstract description 51
- 238000010191 image analysis Methods 0.000 claims abstract description 11
- 238000000034 method Methods 0.000 claims description 28
- 230000002159 abnormal effect Effects 0.000 claims description 22
- 238000001931 thermography Methods 0.000 claims description 22
- 230000008569 process Effects 0.000 claims description 21
- 238000009792 diffusion process Methods 0.000 claims description 18
- 238000001514 detection method Methods 0.000 claims description 17
- 238000010606 normalization Methods 0.000 claims description 12
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 claims description 9
- 229910052760 oxygen Inorganic materials 0.000 claims description 9
- 239000001301 oxygen Substances 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 9
- 230000000694 effects Effects 0.000 claims description 8
- 239000000126 substance Substances 0.000 claims description 4
- 101100518501 Mus musculus Spp1 gene Proteins 0.000 claims description 3
- 230000004044 response Effects 0.000 abstract description 15
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 description 2
- RAHZWNYVWXNFOC-UHFFFAOYSA-N Sulphur dioxide Chemical compound O=S=O RAHZWNYVWXNFOC-UHFFFAOYSA-N 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- UGFAIRIUMAVXCW-UHFFFAOYSA-N Carbon monoxide Chemical compound [O+]#[C-] UGFAIRIUMAVXCW-UHFFFAOYSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 229910002092 carbon dioxide Inorganic materials 0.000 description 1
- 239000001569 carbon dioxide Substances 0.000 description 1
- 229910002091 carbon monoxide Inorganic materials 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 108700041286 delta Proteins 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B17/00—Fire alarms; Alarms responsive to explosion
- G08B17/005—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
- G08B17/06—Electric actuation of the alarm, e.g. using a thermally-operated switch
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B17/00—Fire alarms; Alarms responsive to explosion
- G08B17/10—Actuation by presence of smoke or gases, e.g. automatic alarm devices for analysing flowing fluid materials by the use of optical means
-
- 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
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B31/00—Predictive alarm systems characterised by extrapolation or other computation using updated historic data
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A40/00—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production
- Y02A40/10—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production in agriculture
- Y02A40/28—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production in agriculture specially adapted for farming
Landscapes
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Multimedia (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biodiversity & Conservation Biology (AREA)
- Alarm Systems (AREA)
Abstract
The invention relates to the technical field of fire monitoring and early warning, in particular to a forest fire monitoring and early warning system based on unmanned aerial vehicle image analysis, which comprises an unmanned aerial vehicle throwing module, an identification locking module, a fire monitoring module, a fire analysis module, an early warning terminal and a cloud database; according to the invention, accurate region division is carried out on the target monitoring forest, then the corresponding unmanned aerial vehicle monitoring mode is intelligently matched according to the state information of each region, the fire is rapidly locked, the fire image and the influence factor information are obtained in real time, meanwhile, the combustion state, the smoke state and the influence factor state of the fire region are analyzed, the evaluation index is generated, the fire state evaluation coefficient is comprehensively obtained, the fire trend grade of the fire region is obtained, and timely early warning display is carried out based on the fire trend grade, so that comprehensive, real-time and accurate monitoring and early warning of forest fires are realized, and intelligent decision support is provided for rapid response of forest fires.
Description
Technical Field
The invention relates to the technical field of fire monitoring and early warning, in particular to a forest fire monitoring and early warning system based on unmanned aerial vehicle image analysis.
Background
The forest is like a green daemon of the earth, carries a erector and calms, is vital to maintaining ecological balance and guaranteeing human welfare, however, frequent forest fires form serious threat to the precious green home, and when the traditional forest fire monitoring means deal with a wide forest area, the traditional forest fire monitoring means are limited by the coverage of manpower and ground equipment, so that comprehensive and real-time monitoring is difficult to realize, and the monitoring method mainly depends on a patrol staff to carry out fixed-point patrol according to a preset time table, but the method exposes obvious limitations in actual operation;
Firstly, because patrol personnel need to patrol according to a preset time schedule, the forest area cannot be continuously and uninterruptedly monitored, and in the early stage of a fire, only very small fire sources are often needed, but because the patrol personnel cannot be in place at any time, the fire sources are often difficult to find in time, so that the optimal fire extinguishing time is delayed;
secondly, the judgment of the fire is mainly dependent on personal experience and intuition of the patrol personnel, and although the patrol personnel possibly has rich experience, in a complex forest environment, the development of the fire is often influenced by various factors, and the fire judgment is simply dependent on personal experience and intuition, so that the development trend of the fire is often not accurately predicted, and therefore, effective preparation cannot be provided for rescue work.
Disclosure of Invention
The invention aims to solve the problem that the traditional forest fire monitoring means are limited by the coverage range of manpower and ground equipment when dealing with a wide forest area, so that comprehensive and real-time monitoring is difficult to realize, and provides a forest fire monitoring and early warning system based on unmanned aerial vehicle image analysis.
The aim of the invention can be achieved by the following technical scheme:
Forest fire monitoring and early warning system based on unmanned aerial vehicle image analysis includes:
the unmanned aerial vehicle throwing module is used for throwing and distributing unmanned aerial vehicles to the target monitoring forest, and the specific analysis process is as follows: dividing a target monitoring forest according to a preset area to obtain each subarea of the target monitoring forest, carrying out monitoring analysis on area state information of each subarea of the target monitoring forest to obtain area values of each subarea, carrying out comparison matching analysis on the area values of each subarea and an area state judgment table stored in a cloud database to obtain area state grades of each subarea, and simultaneously matching the area state grades with unmanned aerial vehicle monitoring modes corresponding to the area state grades to obtain unmanned aerial vehicle monitoring modes of each subarea;
The identification locking module is used for monitoring abnormal states of fire corresponding to each subarea, so that identification locking analysis is carried out on the fire areas to obtain fire identification locking signals, real-time tracking shooting is carried out on the fire states corresponding to the fire areas according to the fire identification locking signals, and meanwhile, real-time acquisition is carried out on influence factor information corresponding to different heights of the fire areas to obtain a fire image set and influence factor information corresponding to the fire areas;
The fire monitoring module is used for analyzing the combustion state of the fire area through the combustion monitoring unit to obtain a combustion evaluation index, analyzing the smoke state of the fire area through the smoke monitoring unit to obtain a smoke evaluation index, and simultaneously analyzing the influence factor state of the fire area through the influence factor monitoring unit to obtain an influence factor evaluation index, and further transmitting the obtained combustion evaluation index, the smoke evaluation index and the influence factor evaluation index to the fire analysis module;
The fire analysis module is used for receiving the combustion evaluation index, the smoke evaluation index and the influence factor evaluation index, so as to predict and analyze the fire state of the fire area and obtain the fire trend grade of the fire area;
And the early warning terminal is used for carrying out corresponding early warning display on the fire trend grade based on the fire area.
Further, the monitoring analysis is performed on the area state information of each subarea of the target monitoring forest, and the specific analysis process is as follows:
Acquiring the regional state information of each subregion of the target monitoring forest to obtain the regional state information of each subregion of the target monitoring forest, wherein the regional state information comprises a personnel activity value, a vegetation growth value and a coverage value, and is respectively calibrated to be Q1, Q2 and Q3 according to the formula: obtaining a region value qyu of each sub-region, wherein gamma 1, gamma 2 and gamma 3 respectively represent the personnel activity value, the vegetation growth value and the coverage value weight coefficient.
Further, the identification and locking analysis is performed on the fire area, and the specific analysis process is as follows:
Acquiring region thermal imaging images corresponding to all the subregions through thermal imaging cameras carried by the unmanned aerial vehicle, obtaining region thermal imaging images corresponding to all the subregions, arranging detection points, and extracting chromaticity values of all the detection points from the region thermal imaging images corresponding to all the subregions;
Matching the chromaticity value of each detection point in the region thermal imaging image corresponding to each subarea with a reference chromaticity threshold value of the thermal image corresponding to the preset flame, if the chromaticity value of a certain detection point is successfully matched with the reference chromaticity threshold value of the thermal image corresponding to the preset flame, marking the detection point as a flame point, integrating the flame points in the region thermal imaging image corresponding to each subarea to obtain the flame region area, comparing and analyzing the flame region area with the preset flame region area threshold value, and if the flame region area is larger than the preset flame region area threshold value, judging the subarea as a fire region.
Further, the combustion state of the fire area is analyzed, and the specific analysis process is as follows:
Acquiring a fire image set corresponding to a fire region within a period of time, identifying the direction of smoke diffusion and the direction of flame combustion diffusion from the fire image set, judging as downwind if the direction of smoke diffusion and the direction of flame combustion diffusion are the same, judging as upwind if the direction of smoke diffusion and the direction of flame combustion diffusion are opposite, and extracting a current wind speed mark as a wind shadow value Pm from influence factor information corresponding to the fire region;
Marking a fire image set corresponding to a fire area in a period of time as U, wherein U= {1,2,3,4 … n }, n represents the total number of fire images corresponding to the fire area, each fire image corresponding to the fire area is marked as a monitoring point, the flame height and the combustion area of each monitoring point in a period of time are identified, each monitoring point in a period of time is taken as an abscissa, the flame height and the combustion area are taken as an ordinate, a two-dimensional flame height dynamic coordinate system and a two-dimensional combustion area dynamic coordinate system are established according to the U= {1,2,3,4 … n }, and the flame height and the combustion area of each monitoring point in a period of time are respectively drawn on the two-dimensional flame height dynamic coordinate system and the two-dimensional combustion area dynamic coordinate system in a description manner;
Acquiring the slope between the flame height of each monitoring point and the origin point on a two-dimensional flame height dynamic coordinate system, obtaining the flame height slope value of each monitoring point, performing difference calculation on the flame height slope values of adjacent monitoring points to obtain a flame Gao Xielv difference value, and recording the flame Gao Xielv difference value as hg U according to the formula: obtaining a fire height fluctuation value theta 1, wherein eta 1 and eta 2 are expressed as preset weight coefficients;
Drawing a reference line parallel to the abscissa on a two-dimensional combustion area dynamic coordinate system by using a preset combustion area threshold value, comparing and analyzing the combustion area of each monitoring point with the reference line, extracting the distance length between the combustion area description point of each monitoring point and the reference line, and marking the distance length as a combustion face distance value rs U according to the formula: Obtaining a spread fluctuation value theta 2, wherein eta 3 and eta 4 are expressed as preset weight coefficients;
extracting the numerical values of the fire height fluctuation value theta 1 and the spread fluctuation value theta 2, and carrying out normalization processing according to the formula: a combustion evaluation index rsz is obtained in which μ1 and μ2 represent the weight coefficients of the fire height fluctuation value and the spread fluctuation value, respectively.
Further, the smoke state of the fire area is analyzed, and the specific analysis process is as follows:
Comparing and analyzing the smoke gray value with a preset reference interval by identifying the smoke gray value of each monitoring point in a period of time, judging abnormal smoke when the smoke gray value is out of the preset reference interval, otherwise judging normal smoke, counting the number of times of judging the abnormal smoke, and performing duty ratio analysis on the number of times of judging the abnormal smoke and the total number of times of judging to obtain a smoke abnormal color value ys;
acquiring the smoke concentration of each monitoring point in a period of time, extracting the concentration of harmful substances from the smoke concentration, and carrying out weighted addition to obtain a smoke concentration value yn;
extracting the values of the smoke abnormal color value ys, the smoke damage concentration value yn and the wind shadow value Pm, carrying out normalization treatment, and according to the formula: obtaining a smoke evaluation index ywz; wherein, mu 3, mu 4 and mu 5 respectively represent the weight coefficients of the smoke abnormal color value, the smoke intensity value and the wind shadow value.
Further, the analysis of the state of the influencing factors of the fire area is performed by the following specific analysis process:
The temperature, the humidity and the oxygen content in the influence factor information corresponding to each height of the fire area are obtained, and are respectively calibrated to wd f、sdf and hk f, and the numerical values of the three are extracted for normalization processing according to the formula: An influence factor evaluation index yxz is obtained, in which wd f *、sdf * and hk f * represent a reference temperature, a reference humidity, and a reference oxygen content, respectively, f represents the number of each height, and f=1, 2,3, … v, v represents the total number of each height number, and μ6, μ7, and μ8 represent weight coefficients of the degree of temperature deviation, the degree of humidity deviation, and the degree of oxygen content deviation, respectively.
Further, the prediction analysis is performed on the fire state of the fire area, and the specific analysis process is as follows:
and extracting the values of the combustion evaluation index, the smoke evaluation index and the influence factor evaluation index of the fire area, carrying out normalization processing to obtain fire state evaluation coefficients, carrying out matching analysis on the fire state evaluation coefficients of the fire area and a fire trend state table stored in a cloud database, and enabling each fire state evaluation coefficient of the fire area to correspond to one fire trend grade, thereby obtaining the fire trend grade of the fire area.
Compared with the prior art, the technical scheme provided by the invention has the following beneficial effects:
1. According to the method, the target monitoring forest is divided into a plurality of subareas with equal areas on average, the area state information of each subarea is analyzed to obtain the area value of each subarea, the area state grade of each subarea is determined according to the area value of each subarea and the area state judging table stored in the cloud database in a comparison and matching mode of corresponding unmanned aerial vehicles, so that the forest is finely managed, each subarea is ensured to be properly monitored, meanwhile, the differential monitoring of subareas with different states is realized, the monitoring efficiency and accuracy are improved, and powerful data support is provided for the monitoring analysis of the fire state of the target forest;
2. According to the invention, through the thermal imaging and the high-definition camera carried by the unmanned aerial vehicle, fire identification can be efficiently carried out on each subarea, through comparison of the detection point chromaticity value on the thermal imaging image and the preset flame chromaticity threshold value, flame points can be accurately marked and the area of the flame area can be calculated, once the area of the flame area exceeds the threshold value, the fire area is immediately locked and a fire identification signal is generated, the high-definition camera carries out real-time tracking shooting on the fire area, and meanwhile, a sensor collects influence factor information, so that not only is the accuracy and efficiency of fire identification improved, but also real-time monitoring and timely collection of influence factors are carried out on the fire area, comprehensive and timely information support is provided for fire response, and the remarkable effects of quick response and effective response are achieved;
3. According to the invention, the combustion state, the smoke state and the influence factor state of the fire area are analyzed to obtain the corresponding evaluation index, the obtained evaluation index is comprehensively analyzed to obtain the fire state evaluation coefficient, the fire state evaluation coefficient is matched with the fire trend state table in the cloud database to obtain the fire trend grade of the fire area, and the corresponding early warning display is carried out based on the fire trend grade, so that the automatic data analysis and processing are realized, the intelligent decision support is provided for the fire response, the fire state can be rapidly and accurately evaluated, the corresponding early warning and countermeasure can be formulated according to the fire trend grade, and the efficiency and accuracy of the fire response are improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below. It is evident that the drawings in the following description are only some embodiments of the present invention and that other drawings may be obtained from these drawings without inventive effort for a person of ordinary skill in the art.
Fig. 1 is an overall block diagram of the module of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention more clear, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. It will be apparent that the described embodiments are some, but not all, embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
As shown in fig. 1, a forest fire monitoring and early warning system based on unmanned aerial vehicle image analysis includes: the forest fire monitoring platform is internally provided with an unmanned aerial vehicle throwing module, an identification locking module, a fire monitoring module, a fire analysis module, an early warning terminal and a cloud database, wherein the fire monitoring module comprises a combustion monitoring unit, a smoke monitoring unit and an influence factor monitoring unit;
the unmanned aerial vehicle throwing module is used for throwing and distributing unmanned aerial vehicles to the target monitoring forest, and the specific analysis process is as follows:
Dividing a target monitoring forest according to a preset area, wherein the specific dividing mode is as follows: the method comprises the steps of obtaining the area of a target monitoring forest, dividing the target monitoring forest into a plurality of sub-areas with equal areas according to the preset area of the area, and obtaining the sub-areas of the target monitoring forest;
Acquiring the regional state information of each subregion of the target monitoring forest to obtain the regional state information of each subregion of the target monitoring forest, wherein the regional state information comprises a personnel activity value, a vegetation growth value and a coverage value, and is respectively calibrated to be Q1, Q2 and Q3 according to the formula: Obtaining a region value qyu of each sub-region, wherein gamma 1, gamma 2 and gamma 3 respectively represent a personnel activity value, a vegetation growth value and a coverage value weight coefficient, and gamma 2 is larger than gamma 1 and larger than gamma 3;
The method comprises the steps of obtaining a vegetation image of each subarea of a target monitoring forest by shooting the subareas of the target monitoring forest through a high-definition camera carried by an unmanned aerial vehicle, dividing the vegetation image of each subarea of the target monitoring forest into a plurality of areas, identifying color characteristics of each area, extracting green saturation from the vegetation image, marking the area with the green saturation being larger than a preset green saturation threshold as a bright area, marking the area with the green saturation being smaller than or equal to the preset green saturation threshold as a dark area, counting the total number of the bright area and the total number of the dark area respectively, dividing the total number of the bright area by the total number of the dark area to obtain a vegetation color intensity value, extracting a numerical value corresponding to the vegetation color intensity value, and adding the numerical value of the total number of the bright area and the dark area to obtain a vegetation growth coefficient; the coverage value represents the ratio of the vegetation coverage surface area of each subarea of the target monitoring forest to the total area;
Comparing and matching the area value of each subarea with an area state judging table stored in a cloud database to obtain an area state grade of each subarea, matching the obtained area value of each subarea with an unmanned aerial vehicle monitoring mode corresponding to the area state grade to obtain the unmanned aerial vehicle monitoring mode of each subarea, and carrying out unmanned aerial vehicle throwing layout on each subarea, wherein the unmanned aerial vehicle monitoring mode adopts a timing monitoring mode, i.e. monitoring with the duration of G minutes every T minutes in each day, for example, if a subarea is A grade, the matched unmanned aerial vehicle monitoring mode is monitoring with the duration of 30 minutes every 5 minutes in each day, and if a subarea is B grade, the matched unmanned aerial vehicle monitoring mode is monitoring with the duration of 60 minutes every 10 minutes in each day;
in a specific embodiment, the method comprises the steps of dividing a target monitoring forest into a plurality of subareas with equal areas, analyzing the area state information of each subarea to obtain the area value of each subarea, and carrying out comparison matching according to the area value of each subarea and an area state judging table stored in a cloud database so as to determine the area state grade of each subarea and match a corresponding unmanned aerial vehicle monitoring mode, so that the forest is finely managed, each subarea is ensured to be properly monitored, meanwhile, the differential monitoring of subareas with different states is realized, the monitoring efficiency and accuracy are improved, and powerful data support is provided for the monitoring analysis of the fire state of the target forest;
the identification locking module is used for monitoring abnormal states of fire corresponding to each subarea, so that the fire areas are identified, locked and analyzed, and the specific analysis process is as follows:
Acquiring region thermal imaging images corresponding to all the subregions through thermal imaging cameras carried by the unmanned aerial vehicle, obtaining region thermal imaging images corresponding to all the subregions, arranging detection points, and extracting chromaticity values of all the detection points from the region thermal imaging images corresponding to all the subregions;
Matching the chromaticity value of each detection point in the region thermal imaging image corresponding to each subarea with a reference chromaticity threshold value of the thermal image corresponding to the preset flame, if the chromaticity value of a certain detection point is successfully matched with the reference chromaticity threshold value of the thermal image corresponding to the preset flame, recording the detection point as a flame point, integrating the flame points in the region thermal imaging image corresponding to each subarea to obtain the flame region area, comparing and analyzing the flame region area with the preset flame region area threshold value, judging the subarea as a fire region if the flame region area is larger than the preset flame region area threshold value, generating a fire identification locking signal, carrying out real-time tracking shooting on the fire state corresponding to the fire region through a high-definition camera carried by an unmanned aerial vehicle according to the fire identification locking signal, transmitting a fire image set corresponding to the fire region subjected to the real-time tracking shooting to a fire monitoring module, and simultaneously carrying out real-time acquisition on influence factor information corresponding to different degrees of the fire region through a sensor carried by the unmanned aerial vehicle and transmitting the acquired influence factor information to the fire monitoring module in real time;
In a specific embodiment, the fire identification can be efficiently carried out on each subarea by using the thermal imaging carried by the unmanned aerial vehicle and the high-definition camera, the flame point can be accurately marked and the area of the flame area can be calculated by comparing the chromaticity value of the detection point on the thermal imaging image with the preset flame chromaticity threshold value, once the area of the flame area exceeds the threshold value, the fire area is immediately locked and the fire identification signal is generated, the high-definition camera carries out real-time tracking shooting on the fire area, and meanwhile, the sensor collects the influence factor information, so that the accuracy and efficiency of the fire identification are improved, the real-time monitoring on the fire area and the timely collection of the influence factors are realized, the comprehensive and timely information support is provided for the fire response, and the remarkable effects of quick response and effective response are realized.
The fire monitoring module comprises a combustion monitoring unit, a smoke monitoring unit and an influence factor monitoring unit;
The combustion monitoring unit is used for analyzing the combustion state of the fire area, and the specific analysis process is as follows:
acquiring a fire image set corresponding to a fire region within a period of time, and identifying the smoke diffusion direction and the flame combustion diffusion direction from the fire image set, if the smoke diffusion direction is the same as the flame combustion diffusion direction, judging as downwind, if the smoke diffusion direction is opposite to the flame combustion diffusion direction, judging as upwind, and extracting a current wind speed mark from influence factor information corresponding to the fire region as a wind shadow value Pm, wherein m=1, 2, the current wind speed mark is represented as downwind when m=1, the wind shadow value is represented as P1, the wind shadow value is represented as upwind when m=2, and the wind shadow value is represented as P2;
Marking a fire image set corresponding to a fire area in a period of time as U, wherein U= {1,2,3,4 … n }, n represents the total number of fire images corresponding to the fire area, each fire image corresponding to the fire area is marked as a monitoring point, the flame height and the combustion area of each monitoring point in a period of time are identified, each monitoring point in a period of time is taken as an abscissa, the flame height and the combustion area are taken as an ordinate, a two-dimensional flame height dynamic coordinate system and a two-dimensional combustion area dynamic coordinate system are established according to the U= {1,2,3,4 … n }, and the flame height and the combustion area of each monitoring point in a period of time are respectively drawn on the two-dimensional flame height dynamic coordinate system and the two-dimensional combustion area dynamic coordinate system in a description manner;
Acquiring the slope between the flame height of each monitoring point and the origin point on a two-dimensional flame height dynamic coordinate system, obtaining the flame height slope value of each monitoring point, performing difference calculation on the flame height slope values of adjacent monitoring points to obtain a flame Gao Xielv difference value, and recording the flame Gao Xielv difference value as hg U according to the formula: obtaining a fire height fluctuation value theta 1, wherein eta 1 and eta 2 are expressed as preset weight coefficients;
Drawing a reference line parallel to the abscissa on a two-dimensional combustion area dynamic coordinate system by using a preset combustion area threshold value, comparing and analyzing the combustion area of each monitoring point with the reference line, extracting the distance length between the combustion area description point of each monitoring point and the reference line, and marking the distance length as a combustion face distance value rs U according to the formula: Obtaining a spread fluctuation value theta 2, wherein eta 3 and eta 4 are expressed as preset weight coefficients;
extracting the numerical values of the fire height fluctuation value theta 1 and the spread fluctuation value theta 2, and carrying out normalization processing according to the formula: obtaining a combustion evaluation index rsz, wherein mu 1 and mu 2 respectively represent weight coefficients of a fire height fluctuation value and a spread fluctuation value, and mu 1 is more than mu 2;
In a specific embodiment, the invention comprehensively evaluates the combustion state by judging the wind direction in real time, dynamically monitoring the flame and the combustion area, quantitatively evaluating the flame height change and the combustion area expansion, realizes comprehensive, real-time and accurate evaluation of the combustion state, and provides more comprehensive, accurate and intelligent information support for fire response.
The smoke monitoring unit is used for analyzing the smoke state of the fire area, and the specific analysis process is as follows:
Comparing and analyzing the smoke gray value with a preset reference interval by identifying the smoke gray value of each monitoring point in a period of time, judging abnormal smoke when the smoke gray value is out of the preset reference interval, otherwise judging normal smoke, counting the number of times of judging the abnormal smoke, and performing duty ratio analysis on the number of times of judging the abnormal smoke and the total number of times of judging to obtain a smoke abnormal color value ys;
Acquiring the smoke concentration of each monitoring point in a period of time, extracting the concentration of harmful substances (carbon monoxide, carbon dioxide, sulfur dioxide and particulate matters) from the smoke concentration, and carrying out weighted addition to obtain a smoke concentration value yn;
extracting the values of the smoke abnormal color value ys, the smoke damage concentration value yn and the wind shadow value Pm, carrying out normalization treatment, and according to the formula: Obtaining a smoke evaluation index ywz; wherein, mu 3, mu 4 and mu 5 respectively represent the weight coefficients of the smoke abnormal color value, the smoke concentration value and the wind shadow value, and mu 3 is more than mu 2 and more than mu 1;
in a specific embodiment, the method quantitatively evaluates abnormal smoke by identifying the smoke state in real time, analyzes the concentration of harmful substances in the smoke, and comprehensively analyzes to obtain a smoke evaluation index, so that the smoke state in a fire area is comprehensively and accurately evaluated, and important technical support and decision reference are provided for fire response;
the influence factor monitoring unit is used for analyzing the state of the influence factor of the fire area, and the specific analysis process is as follows:
The temperature, the humidity and the oxygen content in the influence factor information corresponding to each height of the fire area are obtained, and are respectively calibrated to wd f、sdf and hk f, and the numerical values of the three are extracted for normalization processing according to the formula: An influence factor evaluation index yxz is obtained, in which wd f *、sdf * and hk f * represent a reference temperature, a reference humidity, and a reference oxygen content, respectively, f represents the number of each height, and f=1, 2,3, … v, v represents the total number of each height number, μ6, μ7, and μ8 represent weight coefficients of the degree of temperature deviation, the degree of humidity deviation, and the degree of oxygen content deviation, respectively, and μ6 > μ8 > μ7;
In a specific embodiment, the invention realizes more comprehensive understanding of the environmental state of the fire area by comprehensively considering the influence factor information corresponding to each height of the fire area and carrying out deep analysis and accurate evaluation, and provides more accurate and scientific decision support for fire response;
Transmitting the obtained combustion evaluation index, smoke evaluation index and influence factor evaluation index to a fire analysis module;
The fire analysis module is used for receiving the combustion evaluation index, the smoke evaluation index and the influence factor evaluation index, so as to predict and analyze the fire state of the fire area, and the specific analysis process is as follows:
extracting the values of the combustion evaluation index rsz, the smoke evaluation index ywz and the influence factor evaluation index yxz of the fire area, and carrying out normalization treatment according to the formula: obtaining a fire state evaluation coefficient PGZ, wherein e represents a natural constant, delta 1, delta 2 and delta 3 are respectively represented as set correction factors, and the correction factors are used for correcting deviation of various parameters in the formula calculation process, so that more accurate parameter data are calculated;
Matching and analyzing the fire state evaluation coefficients of the fire areas with a fire trend state table stored in a cloud database, wherein each fire state evaluation coefficient of the fire areas corresponds to one fire trend grade, and thus the fire trend grade of the fire areas is obtained;
the early warning terminal is used for carrying out corresponding early warning display on the fire trend grade of the fire area;
In a specific embodiment, the fire state evaluation coefficient is obtained by comprehensively analyzing the combustion evaluation index, the smoke evaluation index and the influence factor evaluation index of the fire region, the fire state evaluation coefficient is obtained by carrying out matching analysis on the fire state evaluation coefficient and the fire trend state table in the cloud database, corresponding early warning display is carried out on the basis of the fire trend level, intelligent decision support is provided for fire response, the fire state can be rapidly and accurately evaluated, corresponding early warning and response measures can be formulated according to the fire trend level, and the efficiency and accuracy of the fire response are improved;
the cloud database is used for storing an area state judging table and a fire trend state table.
The preferred embodiments of the invention disclosed above are intended only to assist in the explanation of the invention. The preferred embodiments are not intended to be exhaustive or to limit the invention to the precise form disclosed. Obviously, many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the invention and the practical application, to thereby enable others skilled in the art to best understand and utilize the invention. The invention is limited only by the claims and the full scope and equivalents thereof.
Claims (5)
1. Forest fire monitoring and early warning system based on unmanned aerial vehicle image analysis, its characterized in that includes:
the unmanned aerial vehicle throwing module is used for throwing and distributing unmanned aerial vehicles to the target monitoring forest, and the specific analysis process is as follows: dividing a target monitoring forest according to a preset area to obtain each subarea of the target monitoring forest, carrying out monitoring analysis on area state information of each subarea of the target monitoring forest to obtain area values of each subarea, carrying out comparison matching analysis on the area values of each subarea and an area state judgment table stored in a cloud database to obtain area state grades of each subarea, and simultaneously matching the area state grades with unmanned aerial vehicle monitoring modes corresponding to the area state grades to obtain unmanned aerial vehicle monitoring modes of each subarea;
The identification locking module is used for monitoring abnormal states of fire corresponding to each subarea, so that identification locking analysis is carried out on the fire areas to obtain fire identification locking signals, real-time tracking shooting is carried out on the fire states corresponding to the fire areas according to the fire identification locking signals, and meanwhile, real-time acquisition is carried out on influence factor information corresponding to different heights of the fire areas to obtain a fire image set and influence factor information corresponding to the fire areas;
The fire monitoring module is used for analyzing the combustion state of the fire area through the combustion monitoring unit, and the specific analysis process is as follows: acquiring a fire image set corresponding to a fire region within a period of time, identifying the direction of smoke diffusion and the direction of flame combustion diffusion from the fire image set, judging as downwind if the direction of smoke diffusion and the direction of flame combustion diffusion are the same, judging as upwind if the direction of smoke diffusion and the direction of flame combustion diffusion are opposite, and extracting a current wind speed mark as a wind shadow value Pm from influence factor information corresponding to the fire region;
Marking a fire image set corresponding to a fire area in a period of time as U, wherein U= {1,2,3,4 … n }, n represents the total number of fire images corresponding to the fire area, each fire image corresponding to the fire area is marked as a monitoring point, the flame height and the combustion area of each monitoring point in a period of time are identified, each monitoring point in a period of time is taken as an abscissa, the flame height and the combustion area are taken as an ordinate, a two-dimensional flame height dynamic coordinate system and a two-dimensional combustion area dynamic coordinate system are established according to the U= {1,2,3,4 … n }, and the flame height and the combustion area of each monitoring point in a period of time are respectively drawn on the two-dimensional flame height dynamic coordinate system and the two-dimensional combustion area dynamic coordinate system in a description manner;
Acquiring the slope between the flame height of each monitoring point and the origin point on a two-dimensional flame height dynamic coordinate system, obtaining the flame height slope value of each monitoring point, performing difference calculation on the flame height slope values of adjacent monitoring points to obtain a flame Gao Xielv difference value, and recording the flame Gao Xielv difference value as hg U according to the formula: obtaining a fire height fluctuation value theta 1, wherein eta 1 and eta 2 are expressed as preset weight coefficients;
Drawing a reference line parallel to the abscissa on a two-dimensional combustion area dynamic coordinate system by using a preset combustion area threshold value, comparing and analyzing the combustion area of each monitoring point with the reference line, extracting the distance length between the combustion area description point of each monitoring point and the reference line, and marking the distance length as a combustion face distance value rs U according to the formula: Obtaining a spread fluctuation value theta 2, wherein eta 3 and eta 4 are expressed as preset weight coefficients;
extracting the numerical values of the fire height fluctuation value theta 1 and the spread fluctuation value theta 2, and carrying out normalization processing according to the formula: Obtaining a combustion evaluation index rsz, wherein mu 1 and mu 2 respectively represent weight coefficients of a fire height fluctuation value and a spread fluctuation value;
and the smoke state of the fire area is analyzed by the smoke monitoring unit, and the specific analysis process is as follows: comparing and analyzing the smoke gray value with a preset reference interval by identifying the smoke gray value of each monitoring point in a period of time, judging abnormal smoke when the smoke gray value is out of the preset reference interval, otherwise judging normal smoke, counting the number of times of judging the abnormal smoke, and performing duty ratio analysis on the number of times of judging the abnormal smoke and the total number of times of judging to obtain a smoke abnormal color value ys;
acquiring the smoke concentration of each monitoring point in a period of time, extracting the concentration of harmful substances from the smoke concentration, and carrying out weighted addition to obtain a smoke concentration value yn;
extracting the values of the smoke abnormal color value ys, the smoke damage concentration value yn and the wind shadow value Pm, carrying out normalization treatment, and according to the formula: Obtaining a smoke evaluation index ywz; wherein, mu 3, mu 4 and mu 5 respectively represent the weight coefficients of the smoke abnormal color value, the smoke intensity value and the wind shadow value;
Meanwhile, the influence factor monitoring unit analyzes the influence factor state of the fire area to obtain an influence factor evaluation index, and then the obtained combustion evaluation index, smoke evaluation index and influence factor evaluation index are sent to the fire analysis module;
The fire analysis module is used for receiving the combustion evaluation index, the smoke evaluation index and the influence factor evaluation index, so as to predict and analyze the fire state of the fire area and obtain the fire trend grade of the fire area;
And the early warning terminal is used for carrying out corresponding early warning display on the fire trend grade based on the fire area.
2. The forest fire monitoring and early warning system based on unmanned aerial vehicle image analysis according to claim 1, wherein the monitoring and analysis are performed on the area state information of each subarea of the target monitoring forest, and the specific analysis process is as follows:
Acquiring the regional state information of each subregion of the target monitoring forest to obtain the regional state information of each subregion of the target monitoring forest, wherein the regional state information comprises a personnel activity value, a vegetation growth value and a coverage value, and is respectively calibrated to be Q1, Q2 and Q3 according to the formula: obtaining a region value qyu of each sub-region, wherein gamma 1, gamma 2 and gamma 3 respectively represent the personnel activity value, the vegetation growth value and the coverage value weight coefficient.
3. The forest fire monitoring and early warning system based on unmanned aerial vehicle image analysis according to claim 1, wherein the identification and locking analysis is performed on the fire area, and the specific analysis process is as follows:
Acquiring region thermal imaging images corresponding to all the subregions through thermal imaging cameras carried by the unmanned aerial vehicle, obtaining region thermal imaging images corresponding to all the subregions, arranging detection points, and extracting chromaticity values of all the detection points from the region thermal imaging images corresponding to all the subregions;
Matching the chromaticity value of each detection point in the region thermal imaging image corresponding to each subarea with a reference chromaticity threshold value of the thermal image corresponding to the preset flame, if the chromaticity value of a certain detection point is successfully matched with the reference chromaticity threshold value of the thermal image corresponding to the preset flame, marking the detection point as a flame point, integrating the flame points in the region thermal imaging image corresponding to each subarea to obtain the flame region area, comparing and analyzing the flame region area with the preset flame region area threshold value, and if the flame region area is larger than the preset flame region area threshold value, judging the subarea as a fire region.
4. The forest fire monitoring and early warning system based on unmanned aerial vehicle image analysis according to claim 1, wherein the analysis of the state of the influencing factors of the fire area is performed by the following specific analysis process:
The temperature, the humidity and the oxygen content in the influence factor information corresponding to each height of the fire area are obtained, and are respectively calibrated to wd f、sdf and hk f, and the numerical values of the three are extracted for normalization processing according to the formula: An influence factor evaluation index yxz is obtained, in which wd f *、sdf * and hk f * represent a reference temperature, a reference humidity, and a reference oxygen content, respectively, f represents the number of each height, and f=1, 2,3, … v, v represents the total number of each height number, and μ6, μ7, and μ8 represent weight coefficients of the degree of temperature deviation, the degree of humidity deviation, and the degree of oxygen content deviation, respectively.
5. The forest fire monitoring and early warning system based on unmanned aerial vehicle image analysis according to claim 1, wherein the prediction analysis is performed on the fire state of the fire area, and the specific analysis process is as follows:
and extracting the values of the combustion evaluation index, the smoke evaluation index and the influence factor evaluation index of the fire area, carrying out normalization processing to obtain fire state evaluation coefficients, carrying out matching analysis on the fire state evaluation coefficients of the fire area and a fire trend state table stored in a cloud database, and enabling each fire state evaluation coefficient of the fire area to correspond to one fire trend grade, thereby obtaining the fire trend grade of the fire area.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410712138.8A CN118280054B (en) | 2024-06-04 | 2024-06-04 | Forest fire monitoring and early warning system based on unmanned aerial vehicle image analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410712138.8A CN118280054B (en) | 2024-06-04 | 2024-06-04 | Forest fire monitoring and early warning system based on unmanned aerial vehicle image analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118280054A CN118280054A (en) | 2024-07-02 |
| CN118280054B true CN118280054B (en) | 2024-07-30 |
Family
ID=91644381
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410712138.8A Active CN118280054B (en) | 2024-06-04 | 2024-06-04 | Forest fire monitoring and early warning system based on unmanned aerial vehicle image analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118280054B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118426493B (en) * | 2024-07-05 | 2024-08-30 | 山东字节信息科技有限公司 | Unmanned aerial vehicle inspection system and method based on cloud platform |
| CN119131976A (en) * | 2024-09-19 | 2024-12-13 | 浙江省森林资源监测中心(浙江省林业调查规划设计院) | A forest fire monitoring system and network construction method based on the Internet of Things |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115131937A (en) * | 2022-06-27 | 2022-09-30 | 南京林业大学 | Forest fire early warning method based on sensor and deep learning |
| CN116778192A (en) * | 2023-05-25 | 2023-09-19 | 淮北矿业(集团)有限责任公司物业分公司 | A fire safety early warning system based on air-ground equipment coordination |
| CN117475570A (en) * | 2023-11-08 | 2024-01-30 | 四川欧航科技有限责任公司 | Radar monitoring intelligent early warning system suitable for forest fire prevention |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110047241A (en) * | 2019-04-27 | 2019-07-23 | 刘秀萍 | A kind of forest fire unmanned plane cruise monitoring system |
| CN111951510A (en) * | 2020-08-24 | 2020-11-17 | 广州立信电子科技有限公司 | Forestry fire prevention intelligence patrols and examines monitoring early warning system based on big data |
| US11295131B1 (en) * | 2021-06-15 | 2022-04-05 | Knoetik Solutions, Inc. | Smoke and fire recognition, fire forecasting, and monitoring |
| CN116386252B (en) * | 2023-04-11 | 2024-04-02 | 东北林业大学 | A combustible monitoring method based on forest area division |
| CN117291430B (en) * | 2023-11-24 | 2024-03-19 | 江苏海内软件科技有限公司 | Safety production detection method and device based on machine vision |
-
2024
- 2024-06-04 CN CN202410712138.8A patent/CN118280054B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115131937A (en) * | 2022-06-27 | 2022-09-30 | 南京林业大学 | Forest fire early warning method based on sensor and deep learning |
| CN116778192A (en) * | 2023-05-25 | 2023-09-19 | 淮北矿业(集团)有限责任公司物业分公司 | A fire safety early warning system based on air-ground equipment coordination |
| CN117475570A (en) * | 2023-11-08 | 2024-01-30 | 四川欧航科技有限责任公司 | Radar monitoring intelligent early warning system suitable for forest fire prevention |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118280054A (en) | 2024-07-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN118280054B (en) | Forest fire monitoring and early warning system based on unmanned aerial vehicle image analysis | |
| EP1628260B1 (en) | Method and system for automatic forest fire recognition | |
| CN111930049A (en) | Forestry forest fire prevention safety intelligent monitoring management system based on thing networking | |
| CN105915840A (en) | Factory smoke emission automatic monitoring method based on video signal | |
| CN109543631A (en) | A kind of fire image detection alarm method based on machine learning | |
| CN109406412A (en) | A kind of plant health method for monitoring state and device | |
| CN113487166A (en) | Chemical fiber floating filament quality detection method and system based on convolutional neural network | |
| CN118314576A (en) | Seedling condition monitoring system and method based on Internet of things | |
| CN118092326A (en) | Digital twinning-based factory emergency management system | |
| CN119296239A (en) | Anti-theft monitoring and early warning method based on video remote monitoring | |
| CN113267601B (en) | Industrial production environment remote real-time monitoring cloud platform based on machine vision and data analysis | |
| CN118075511B (en) | Intelligent camera monitoring system based on Internet of things | |
| CN119007099A (en) | Method and system for monitoring fish floating head phenomenon based on machine vision edge AI identification | |
| CN117372224B (en) | Greening ecological restoration system | |
| CN118509557A (en) | Land supervision and inspection method based on unmanned aerial vehicle cluster | |
| CN115063427B (en) | Pollutant discharge monitoring image processing method for novel ship | |
| CN117581777A (en) | Intelligent irrigation system and computer equipment based on internet of things technology | |
| CN117440018A (en) | Network camera equipment fortune pipe platform based on thing networking | |
| CN117240997A (en) | AI (advanced technology attachment) -identification-based inspection method and system | |
| CN113052959B (en) | Carbon emission-based sea island vegetation resource releasing method and device | |
| CN117035395A (en) | Operation and maintenance management method for industrial identification analysis application | |
| Putra et al. | Vision-based object detection for efficient monitoring in smart hydroponic systems | |
| CN118823633B (en) | Image recognition method and system in the field of industrial safety based on AI big model | |
| CN117932278B (en) | Smart city environment-friendly monitoring system and method | |
| CN110532893A (en) | Icon detection method in the competing small map image of electricity |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 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 |