METHOD AND SYSTEM FOR MAPPING A FOREST FIRE RISK
The process and system for mapping forest fire risk by detecting fuel types and determining parametric values for wind, water content, and topography produce precise and scalable risk maps, addressing the imprecision of existing maps and enhancing fire management.
Patent Information
- Application Number
- FR2023004053
- Authority / Receiving Office
- FR · FR
- Patent Type
- Utility models
- Current Assignee / Owner
- Filing Date
- 2023-04-21
- Publication Date
- 2025-04-18
- Estimated Expiration
- 2033-04-21
Smart Images
Figure 00000000_0000_ABST
Abstract
Description
Title of the invention: METHOD AND SYSTEM FOR MAPPING A FOREST FIRE RISK Technical field of the invention
[0001] The present invention relates to a method for mapping a forest fire risk and a system for mapping a forest fire risk. It applies, in particular, to the field of anticipation and combating forest fires of natural or human origin. State of the art
[0002] The approaches described in this section are approaches that could be pursued, but not necessarily approaches that have been conceived or pursued previously. Therefore, unless otherwise indicated, it should not be assumed that any of the approaches described in this section constitute prior art solely because of its inclusion in this section.
[0003] Forest fires are particularly difficult to anticipate, which requires constant vigilance from the emergency services and authorities, particularly during the summer periods when these fires are most likely to start and spread.
[0004] Today, so-called "forest fire hazard" maps are developed according to different methodologies or doctrines. However, these maps are imprecise and rarely updated, poorly reflecting the observed reality on the ground. Brief description of the figures
[0005] Other advantages, aims and particular characteristics of the invention will emerge from the following non-limiting description of at least one particular embodiment of the method and device which are the subject of the present invention, with reference to the appended drawings, in which: [Fig.l] represents, schematically and in the form of a flowchart, a particular succession of steps of the method which is the subject of the present invention, [Fig.2] schematically represents a particular embodiment of the device which is the subject of the present invention and [Fig.3] schematically represents an example of a map produced by implementing the method and / or device which is the subject of the present invention. Detailed description of the invention
[0006] The present description is given without limitation, each characteristic of an embodiment being able to be combined with any other characteristic of any other embodiment in an advantageous manner.
[0007] It should be noted from now on that the figures are not to scale.
[0008] As understood from the present description, various inventive concepts may be implemented by one or more methods or devices described below, several examples of which are provided herein. The actions or steps performed in carrying out the method or device may be ordered in any suitable manner. Accordingly, it is possible to construct embodiments in which the actions or steps are performed in a different order than illustrated, which may include performing certain acts simultaneously, even if they are presented as sequential acts in the illustrated embodiments.
[0009] The expression "and / or", as used herein and in the claims, is to be understood as meaning "either or both" of the elements so conjoined, i.e., elements which are present conjunctively in some cases and disjunctively in other cases. Multiple elements listed with "and / or" are to be interpreted in the same way, i.e., "one or more" of the elements so conjoined. Other elements may optionally be present, other than the elements specifically identified by the "and / or" clause, whether or not they are related to these specifically identified elements.Thus, by way of non-limiting example, a reference to "A and / or B", when used in conjunction with open language such as "comprising" may refer, in one embodiment, to A only (optionally including elements other than B); in another embodiment, to B only (optionally including elements other than A); in yet another embodiment, to both A and B (optionally including other elements); etc.
[0010] As used herein in the description and claims, "or" is to be understood inclusively.
[0011] As used in this specification and in the claims, the expression "at least one", with reference to a list of one or more elements, is to be understood to mean at least one element selected from one or more elements in the list of elements, but not necessarily including at least one of each element specifically listed in the list of elements and not excluding any combination of elements in the list of elements. This definition also allows for the optional presence of elements other than the specifically identified elements in the list of elements to which the expression "at least one" refers, whether or not related to those specifically identified elements.Thus, by way of non-limiting example, "at least one of A and B" (or, equivalently, "at least one of A or B", or, equivalently, "at least one of A and / or B") may refer, in one embodiment, to at least one, possibly including more than one, A, without B present (and possibly including elements other than B); in another embodiment. embodiment, at least one, optionally including more than one, B, without A present (and optionally including elements other than A); in yet another embodiment, at least one, optionally including more than one, A, and at least one, optionally including more than one, B (and optionally including other elements); etc.
[0012] In the claims, as well as in the description below, all transitional expressions such as "comprising", "including", "carrying", "having", "containing", "involving", "holding", "composed of", and the like, are to be understood as being open, i.e., as meaning including but not limited to. Only the transitional expressions "consisting of" and "consisting essentially of" are to be understood as closed or semi-closed transitional expressions, respectively.
[0013] In the present description, an "input means" refers to any device allowing the transmission of information to a computer system. Such an input means is, for example, a keyboard, a mouse and / or a touch screen adapted to interact with a computer system so as to collect user input. In variants, the input means is of a logical nature, such as a network port of a computer system configured to receive an electronically transmitted input command. Such an input means may be associated with a graphical user interface (GUI) presented to a user or an application programming interface (API). In other variants, the input means may be a sensor configured to measure a specified physical parameter relevant to the intended use case.
[0014] A "computing device" is any electronic computing device, unitary or distributed, capable of receiving digital inputs and providing digital outputs through and to any kind of digital interface. Typically, a computing device refers to either a computer running software with access to data storage, or a client-server architecture in which the data and / or at least part of the calculations are performed on the server side while the client side serves as an interface.
[0015] A "digital identifier" is any information that uniquely represents a physical or logical object. Such a digital identifier is, for example, an identifier in a database.
[0016] [Fig. 1], which is not to scale, shows a schematic view of an embodiment of the method 100 which is the subject of the present invention. This method 100 for mapping a forest fire risk comprises: - a step 105 of detecting at least one type of fuel in a geographical area to be mapped, - a step 110 of determining, by a calculation device, a para- value metric representative of the presence of at least one fuel detected in at least one photograph of the area, - a step 115 of determining, by a calculation device, a parametric value representative of wind movements in the zone, - a step 120 of determining, by a calculation device, a parametric value representative of a water content of the zone, - a step 125 of calculation, by a calculation device, of a parametric value representative of the spread of a fire in the zone, - a step 130 of estimating, by a calculation device, a parametric value representative of a probability of occurrence of a fire in the zone, - a step 133 of determining, by a calculation device, a value representative of the topography of the zone, - a step 135 of calculation, by a calculation device, of at least one value representative of a risk of forest fire in the area according to: - the parametric value representative of the presence of at least one detected fuel, - the parametric value representative of wind movements, - the parametric value representative of a water content, - the parametric value representative of a fire propagation, - the parametric value representative of the topography and - the parametric value representative of a probability of occurrence of a fire, - a step 140 of association, by a calculation device, of at least one value representative of a risk of forest fire with geographical coordinates and - a step 145 of supplying, on an interface, a map representative of at least one value representative of a risk of forest fire produced according to the result of the association step.
[0017] The detection step 105 is carried out by a detection means 205, as shown in [Fig.2]. The nature of the detection means 205 depends on the methods of implementing the detection step 105.
[0018] In variants, the detection step 105 is carried out by the study of an area by an operator and by the entry of digital identifiers representative of plants and / or fuels present in an area. The detection means 205 then corresponds, for example, to computer software executed by a calculation device configured to display an interface for entry of digital identifiers of plants by an operator.
[0019] In variants, during the detection step 105, a detected fuel is approximated to a fuel typological model, among a limited selection of fuel models in a typology, as a function of values of physical quantities of a fuel.
[0020] Such physical quantities are, for example: - a quantity of combustible biomass of dead elements of less than 0.64 cm (in tonnes / ha), - a quantity of combustible biomass of dead elements of 0.64 to 2.54 cm (in tonnes / ha), - a quantity of combustible biomass of dead elements of more than 2.54 cm (in tonnes / ha), - a quantity of biomass of living herbaceous plants (in tonnes / ha), - a quantity of biomass of living shrub elements (in tonnes / ha), - a depth of the fuel bed (in metres), - a surface / volume ratio of dead elements of less than 0.64 cm (in m2 / m3), - a surface / volume ratio of dead elements of 0.64 to 2.54 cm (in m2 / m3), - a surface / volume ratio of dead elements of more than 2.54 cm (in m2 / m3), - a surface / volume ratio of living herbaceous plants (in m2 / m3), - an extinction humidity rate of the dead fuel (in %), - a calorific value (kJ / kg), - a wind adjustment factor, - a quantity of surface combustible biomass (in tonnes / ha), - 2 corrective parameters of the model (a and b), - a value representative of the Byram constant.
[0021] The first 6 parameters are measured in the field on the model samples. The next 7 are taken from bibliographic data. The combustible biomass is formed by the sum of the first 4 parameters. The corrective parameters are obtained by adjusting the model. The Byram constant is calculated as follows (Calorific value x (biomass of dead elements less than 2.54 cm and herbaceous plants x 4.88)).
[0022] Such a fuel typology can be developed on the basis of a method developed by Cemagref (or center for the study of agricultural machinery and rural engineering of water and forests) as described in the document "Development of a fuel typology for Lower Provence limestone", Chandioux, Lampin-Maillet, Jappiot, Forêt méditerranéenne, 2009. This typology is a variation of the method that led to the creation of the USDA fuel typology ("How to predict the spread and intensity of forest Range Fire", Rothermel, USDA FSIRS, 1983).
[0023] Such a typology can be defined, for example, by the 17 input parameters of the model named "r.ros" (as implemented by the Geography Resources Analysis Support System, or GRASS).
[0024] Such a typology allows, for example, the classification of fuels among a list of 71 types of fuels. Such a list is compatible, for example, with the typology defined by the FireEUrisk project.
[0025] Thus, in optional embodiments, the detection step 105 comprises: - a step 106 of collecting plant samples in a geographical area to be mapped, - a step 107 of measuring values of predetermined physical quantities as a function of the collected samples and - a step 108 of determining parametric values of a fuel typological representation model as a function of the values of determined physical quantities.
[0026] The collection step 106 is carried out, for example, manually by an operator, via a specialized tool or automatically.
[0027] Step 107 of measuring values of physical quantities depends on the physical quantity measured.
[0028] For example, the biomass quantity values for each fuel class are obtained by field measurement of the biovolumes of the litter, herbaceous and shrub layers (expressed in mVha from a height and a percentage of coverage). This biovolume is then multiplied by typical biomasses of complete vegetation from a database compiled from the bibliography (Cemagref 2009) and additional measurements, expressed in tonnes / m3.
[0029] Step 108 of determining parametric values of a fuel typological representation model is carried out, for example, by implementing a calculation device associated with an input means allowing the input of the measured values. Such a calculation device then implements an algorithmic succession of mathematical and / or statistical operations to obtain a value for at least one parameter as a function of at least one measured value.
[0030] In variants, the collection 106, measurement 107 and determination 108 steps are replaced by: - a step (not shown) of selecting a digital identifier representative of a fuel and - a step (not shown) of selecting a digital identifier representative of a type based on the selected fuel identifier.
[0031] In variants, the step of selecting a digital identifier representative of a fuel is replaced by a step (not shown) of selecting attributes (type of foliage, for example), the method 200 further comprising a step (not shown) of collecting, in a database, data representative of values of physical quantities, the step 108 of determining parametric values being carried out as a function of the values thus collected.
[0032] In optional embodiments, the fuel typological representation model is adapted to the representation of fuel types of the Mediterranean basin.
[0033] Step 110 of determining a parametric value representative of the presence of at least one fuel detected in at least one photograph of the area is carried out by a determination means 210 as represented in [Fig.l]. The nature of the determination means 210 depends on the methods of implementing the determination step 110.
[0034] In variants, the determination step 110 is carried out by implementing an input means, such as a user interface, associated with software executed by a computing device, allowing a user to select coordinates in a photograph of the area, corresponding to geographic coordinates, and to associate with these coordinates a digital fuel identifier. In advanced modes, parameters representative of a quantity or concentration of the fuel can also be entered.
[0035] In variants, the determination step 110 implements image processing software executed by a computing device. During this determination step 110, an image processing algorithm, which may implement a trained machine learning model, detects the presence of fuels.
[0036] For example, such a determination step 110 is implemented, over the entire area concerned by the mapping, a set of learning data is constituted by photo-interpretation, in the form of polygons at a rate of several tens to several hundreds of reference polygons for each type of fuel. These learning data are then associated with the data from the multi-temporal satellite imagery by the use of suitable algorithms.
[0037] In optional embodiments, step 110 of determining a parametric value representative of the presence of at least one detected fuel comprises a step 111 of processing, by a calculation device, at least one satellite photograph of the zone.
[0038] In optional embodiments, step 111 of processing at least one satellite photograph of the area implements a trained machine learning model.
[0039] Such a trained machine learning model can be obtained by implementing a machine learning algorithm suitable for classifying patterns in images. Such an algorithm can implement, for example, an artificial neural network.
[0040] Such a satellite photograph may be obtained at regular or irregular intervals. Such a satellite photograph is updated, for example, once a year, month or week.
[0041] Step 115 of determining a parametric value representative of wind movements in the area is carried out by a determination means 215 as shown in [Fig.l]. The nature of the determination means 215 depends on the methods of implementing the determination step 115.
[0042] In variants, step 115 of determining a parametric value representative of wind movements is carried out by removing a set of values representative of wind movement from a database.
[0043] In variants, the determination step 115 implements an input means, such as a user interface, allowing the input of a unidirectional direction value and a wind intensity value.
[0044] In variants, the wind movement step is 10 meters.
[0045] For example, the determination step 115 implements a model for determining the speed and direction of the wind by wind flow, the input data of which are: - the speed and direction of the wind at a height of 10 meters at at least one point in the territory, - the wind speed and direction at several points in the territory allowing the algorithm to model the values between the different reference points and - a digital terrain model allowing the software to calculate the flow of the surface layer's liminal flux.
[0046] In particular embodiments, step 115 of determining wind movements is carried out as a function of at least one topography value determined during step 133 of determination.
[0047] Step 120 of determining a parametric value representative of a water content of the zone is carried out by a determination means 220 as represented in [Fig.l]. The nature of the determination means 220 depends on the methods of implementing the determination step 120.
[0048] In variants, a database of water contents associated with the different fuels is implemented and consulted to extract therefrom, for at least one fuel detected during the detection step 105, a water content value.
[0049] In variants, a soil water content database is implemented and consulted to extract, for geographic coordinates corresponding to the analyzed area, a soil water content value.
[0050] In optional embodiments, step 120 of determining a parametric value representative of a water content comprises: - a step 121 of determining, by a calculation device, an index representative of the water content of the vegetation on a map of the area according to at least three classes, - a step 122 of locating, by a calculation device, each class at a minimum humidity value from the fuel cut-off network and - a step 123 of generating, by a calculation device, a map of the minimum water content in at least three classes.
[0051] Step 125 of calculating a parametric value representative of a propagation of a fire in the zone is carried out by a calculation means 225 as represented in [Fig.l]. The nature of the calculation means 225 depends on the methods of implementing the calculation step 125.
[0052] In variants, the calculation step 125 is carried out by implementing a means of input, by a user, of a value representative of a speed of propagation of the fire.
[0053] In variants, the calculation step 125 implements a parametric (or index) model or a numerical model. The references linked to parametric models in France are linked to the work of D. Alexandrian around 1982. Parametric models give relative values and are constructed by simple equations (additions and multiplications) implementing parameters derived from expertise (flammability index, combustibility index). Research has focused more, for about twenty years, on the development of numerical models. Numerical models give absolute values expressing physical values such as the intensity of the fire in kW / m and are constructed on equations describing physical phenomena. Such a numerical model is, for example, a 3D model coupling fire and atmosphere making it possible to produce realistic fire propagation values (in particular their unstable and explosive nature).Such a numerical model associates, for example, the intensity of a fire with different propagation parameters (slope, wind, water content) for a range of fuel types.
[0054] In embodiments, step 125 of calculating a parametric value representative of a propagation of a fire in the zone comprises: - a step 126 of determining, by a calculation device, a speed of propagation of the fire, - a step 127 of determining, by a calculation device, an intensity of the fire, - a step 128 of determining, by a calculation device, a value representative of a topography in a photograph of the area and - a step 129 of determining, by a calculation device, a value representative of a wind speed and direction in a photograph of the area.
[0055] Step 125 of calculating a parametric value representative of a fire propagation is then carried out as a function of the determined propagation speed, the determined fire intensity, at least one determined topography value and at least one minus a wind speed and direction value.
[0056] Step 126 of determining a propagation speed is carried out, for example, by computer software executed by a computing device and configured to operate a mathematical model of propagation of a fire as a function of a finite number of parameters.
[0057] Step 127 of determining an intensity is carried out, for example, by computer software executed by a computing device and configured to operate a mathematical model of propagation of a fire as a function of a finite number of parameters.
[0058] Step 128 of determining a topography is carried out, for example, by computer software executed by a computing device and configured to carry out image processing on an image of the area to determine parameter values, such as slope values for example.
[0059] Step 129 of determining a wind speed and direction is carried out, for example, by computer software executed by a computing device and configured to carry out image processing on an image of the area to determine parameter values.
[0060] Step 130 of estimating a parametric value representative of a propagation of a fire in the zone is carried out by an estimation means 230 as represented in [Fig.l]. The nature of the estimation means 230 depends on the methods of implementing the estimation step 130.
[0061] In variants, the estimation step 130 is carried out by consulting, by a calculation device, a database representative of past fire outbreaks and by calculating a probability of outbreak based on a number of outbreaks observed in the area or in an area having similar characteristics, for example. In particular, the historical data from the Promethee database dedicated to forest fires and their spatial modeling by adapting the AÏOLI model, as described by Marielle Jappiot in 2006, can be implemented.
[0062] In variants, the estimation step 130 is carried out by consulting, by a calculation device, a database representative of fire contours and over a duration allowing the average fire to reach the surface of the reference fire.
[0063] Such a database of past fires contains, for example, at least information, for each forest fire on the date and time of start, on the location of the start, on the cause of the start of the fire and on the total area burned.
[0064] Such a database can be used: - to make a link between the causes and locations of fire outbreaks and the geographical components of the territory (for example, 2% of fire outbreaks are linked to accidents on the electricity network, we can allocate 2% of fire outbreaks to a 10-meter buffer zone on either side of the power lines) and - to determine the surface area of the reference fires, defined as the last decile of the surface area sample.
[0065] This surface of the reference fires is used to parameterize the fire duration in the model, the median surface obtained having to correspond to the surface of the reference fires.
[0066] In embodiments, step 130 of estimating a parametric value representative of a probability of occurrence of a fire comprises: - a step 131 of determining, by a calculation device, a probable occurrence value based on historical data for the area and - a step 132 of modeling, by a calculation device, a reference fire contour based on data representative of other zones.
[0067] Such a probability of occurrence can thus be obtained as follows: - fire contours are obtained, for example by using predetermined isochronous lines or by using a fire spread modeling tool such as "r.spread", the input parameters of this model are, for a geographical area and for each point in this area, the fuel type, the slope and orientation, the wind speed and direction, the water contents of the dead and living fuel and the output data are fire spread isochronous lines - these contours are obtained for each fire starting point studied, - the modeled hatching points are distributed over the territory studied, at a rate of several hundred to several tens of thousands and according to a random distribution guided by the probability of hatching, - this results in a multiplicity of simulated fires across the entire territory and - the probability of occurrence is defined, for each point in the territory, as the number of simulated fires having affected this point, divided by the total number of simulated fires on a circle with a radius equal to the maximum length of the simulated fires.
[0068] Step 133 of determining a value representative of the topography of the area can be carried out by consulting a database of topographic surveys for example.
[0069] Step 135 of calculating a parametric value representative of a propagation of a fire in the zone is carried out by a calculation means 235 as represented in [Fig.l]. The nature of the calculation means 235 depends on the methods of implementing the calculation step 135.
[0070] The forest fire hazard can be defined as a product of the fire intensity and the probability of occurrence. The intensity is represented in classes corresponding to the fire intensity scale as defined by IRSTEA (Lampin 2003). The occurrence can be divided into classes by distributing the sample into quartiles. The hazard is derived from a matrix crossing the intensity and occurrence classes. This matrix can be adapted to each territory.
[0071] The calculation step 135 implements, for example, computer software executed by a calculation device.
[0072] The step 140 of associating at least one value representative of a forest fire risk with geographical coordinates is carried out by an association means 240 as represented in [Fig.l]. The nature of the association means 240 depends on the methods of implementing the association step 140.
[0073] In variants, the association step 140 implements, for example, computer software executed by a computing device configured to, during the calculation step 135, associate a fire risk value with a positioning value in the area. Such positioning may correspond to coordinates in a reference frame relating to the area only or to coordinates in a planetary reference frame.
[0074] The supply step 145 is carried out by a supply means 145 as shown in [Fig.l]. The nature of the supply means 245 depends on the methods of implementing the supply step 145.
[0075] In variants, the provisioning step 145 implements a user interface, as shown in [Fig.3].
[0076] [Fig. 2] shows schematically a particular embodiment of the system 200 which is the subject of the present invention. This system 200 for mapping a risk of forest fire comprises: - a means 205 for detecting at least one type of fuel in a geographical area to be mapped, - a means 210 for determining, by a calculation device, a parametric value representative of the presence of at least one fuel detected in at least one photograph of the zone, - a means 215 of determining, by a calculation device, a parametric value representative of wind movements in the area, - a means 220 of determining, by a calculation device, a parametric value representative of a water content of the zone, - a means 225 of calculating, by a calculation device, a parametric value representative of the spread of a fire in the zone, - a means 230 of estimating, by a calculation device, a parametric value representative of a probability of occurrence of a fire in the zone, - a step 233 of determining, by a calculation device, a value representative of the topography of the zone, - a means 235 of calculating, by a calculating device, at least one value representative of a risk of forest fire in the area according to: - the parametric value representative of the presence of at least one detected fuel, - the parametric value representative of wind movements, - the parametric value representative of a water content, - the parametric value representative of the topography, - the parametric value representative of the spread of a fire and - the parametric value representing a probability of occurrence of a fire, - a means 240 of associating, by a calculation device, at least one value representative of a risk of forest fire with geographical coordinates and - a means 245 for providing, on an interface, a map representing at least one value representative of a forest fire risk produced according to the result of the association step.
[0077] Particular embodiments of the characteristic means of the system 200 which is the subject of the present invention are described with reference to [Fig.l]. Statement of the invention
[0078] The present invention aims to remedy all or part of these drawbacks.
[0079] To this end, according to a first aspect, the present invention aims at a method for mapping a forest fire risk, which comprises: - a step of detecting at least one type of fuel in a geographical area to be mapped, - a step of determining, by a calculation device, a parametric value representative of the presence of at least one fuel detected in at least one photograph of the area, - a step of determining, by a calculation device, a parametric value representative of wind movements in the area, - a step of determining, by a calculation device, a parametric value representative of the water content of the zone, - a calculation step, by a calculation device, of a parametric value representative of the spread of a fire in the area, - a step of estimating, by a calculation device, a parametric value representative of a probability of occurrence of a fire in the zone, - a step of determining, by a calculation device, a value representative of the topography of the area, - a calculation step, by a calculation device, of at least one value representative of a risk of forest fire in the area depending on: - the parametric value representative of the presence of at least one detected fuel, - the parametric value representative of wind movements, - the parametric value representative of water content, - the parametric value representative of topography, - the parametric value representative of fire propagation and - the parametric value representative of a probability of occurrence of a fire, - a step of association, by a calculation device, of at least one value representative of a risk of forest fire with geographical coordinates and - a step of providing, on an interface, a map representing at least one value representative of a forest fire risk produced based on the result of the association step.
[0080] Thanks to these provisions, an assembly of tools and models, with the aim of producing precise, reliable and scalable forest fire hazard maps, is obtained. These maps make it possible to locate the places most likely to cause a forest fire to start and the probable spread of such a fire based on environmental, climatic and / or topographical characteristics of a given area. Such knowledge allows for effective and preventive deployment of emergency services. A map thus obtained has, for example, a scale close to 1 / 10000. A map thus obtained complies with the following doctrinal documents: - Natural risk prevention plans, Forest fire risks, Methodological guide, MEDD 2002, - Technical note of July 29, 2015, Taking into account the risk of forest fires in prevention and land use planning documents, MEDDE 2015 and - Taking into account the risk of forest fires in urban planning, elements for drafting the State's Notification, CEREMA 2018
[0081] In optional embodiments, the detection step comprises: - a stage of collecting plant samples in a geographical area to be mapped, - a step of measuring values of predetermined physical quantities based on the samples collected and - a step of determining parametric values of a typological representation model of fuels based on the values of determined physical quantities.
[0082] These embodiments improve the reliability of the forest fire start and spread model with respect to the terrain represented by an area photograph.
[0083] In optional embodiments, the fuel typology representation model is adapted to the representation of fuel types of the Mediterranean basin.
[0084] In optional embodiments, the step of determining a value pa The measurement method representative of the presence of at least one detected fuel comprises a step of processing, by a calculation device, at least one satellite photograph of the area.
[0085] These embodiments allow frequent updating of the obtained forest fire risk map.
[0086] In optional embodiments, the step of processing at least one satellite photograph of the area implements a trained machine learning model.
[0087] These embodiments make it possible to automatically determine, from the implementation of a machine learning algorithm on a training data set, a fuel type. Such embodiments may have a better ability to detect a fuel type on an image than a traditional image processing algorithm.
[0088] In optional embodiments, the step of determining wind movements is carried out as a function of at least one topography value determined during the determining step.
[0089] In optional embodiments, determining a parametric value representative of a water content comprises: - a step of determining, by a calculation device, an index representative of the water content of the vegetation on a map of the area according to at least three classes, - a location step, by a calculation device, of each class at a minimum humidity value from the fuel cut-off network and - a step of generating, by a calculation device, a map of the minimum water content in at least three classes.
[0090] These embodiments make it possible to optimize the modeling of the water content in a considered geographical area.
[0091] In optional embodiments, the step of calculating a parametric value representative of a propagation of a fire in the zone comprises: - a step of determining, by a calculation device, a speed of propagation of the fire, - a step of determining, by a calculation device, the intensity of the fire, - a step of determining, by a calculation device, a value representative of a topography in a photograph of the area and
[0092] the step of calculating a parametric value representative of a propagation of a fire being carried out as a function of the determined propagation speed, the determined intensity of the fire and at least one determined topography value.
[0093] These embodiments make it possible to optimize the modeling of the spread of fire in a considered geographical area.
[0094] In optional embodiments, the step of estimating a parametric value representative of a probability of occurrence of a fire comprises: - a step of determining, by a calculation device, a probable occurrence value based on historical data for the area and - a modeling step, using a calculation device, of a reference fire contour based on data representative of other zones.
[0095] These embodiments make it possible to optimize the modeling of a probability of occurrence of a fire in a considered geographical area.
[0096] According to a second aspect, the present invention relates to a system for mapping a forest fire risk, which comprises: - a means of detecting at least one type of fuel in a geographical area to be mapped, - a means of determining, by a calculation device, a parametric value representative of the presence of at least one fuel detected in at least one photograph of the area, - a means of determining, by a calculation device, a parametric value representative of wind movements in the area, - a means of determining, by a calculation device, a parametric value representative of the water content of the zone, - a means of calculating, by a calculation device, a parametric value representative of the spread of a fire in the area, - a means of estimating, by a calculation device, a parametric value representative of a probability of occurrence of a fire in the area, - a means of determining, by a calculation device, a value representative of the topography of the area, - a means of calculating, by a calculation device, at least one value representative of a risk of forest fire in the area based on: - the parametric value representative of the presence of at least one detected fuel, - the parametric value representative of wind movements, - the parametric value representative of a water content, - the parametric value representative of the topography, - the parametric value representative of the spread of a fire and - the parametric value representing a probability of occurrence of a fire, - a means of associating, by a calculation device, at least one value representative of a forest fire risk with geographical coordinates and - a means of providing, on an interface, a map representing at least one representative value of a forest fire risk produced based on the result of the association step.
Claims
Claims
1. Method (100) for mapping a forest fire risk, characterized in that it comprises: - a step (105) of detecting at least one type of fuel in a geographical area to be mapped, - a step (110) of determining, by a calculation device, a parametric value representative of the presence of at least one fuel detected in at least one photograph of the zone, - a step (115) of determining, by a calculation device, a parametric value representative of wind movements in the zone, - a step (120) of determining, by a calculation device, a parametric value representative of a water content of the zone, - a step (125) of calculating, by a calculation device, a parametric value representative of a propagation of a fire in the zone, - a step (130) of estimating, by a calculation device, a parametric value representative of a probability of occurrence of a fire in the zone, - a step (133) of determining, by a calculation device, a value representative of the topography of the zone, - a step (135) of calculating, by a calculating device, at least one value representative of a risk of forest fire in the area as a function of: - the parametric value representative of the presence of at least one detected fuel, - the parametric value representative of wind movements, - the parametric value representative of water content, - the parametric value representative of topography, - the parametric value representative of the spread of a fire and - the parametric value representative of the probability of occurrence of a fire, - a step (140) of association, by a calculation device, of at least one value representative of a risk of forest fire with geographical coordinates and - a step (145) of providing, on an interface, a map representing at least one value representative of a forest fire risk produced according to the result of the association step.
2. The method (100) of claim 1, wherein the step (105) of detection comprises: - a step (106) of collecting plant samples in a geographical area to be mapped, - a step (107) of measuring values of predetermined physical quantities as a function of the samples collected and - a step (108) of determining parametric values of a model of typological representation of fuels as a function of the values of determined physical quantities.
3. Method (100) according to claim 2, wherein the fuel typological representation model is adapted to the representation of fuel types of the Mediterranean basin.
4. Method (100) according to one of claims 1 to 3, in which the step (110) of determining a parametric value representative of the presence of at least one detected fuel comprises a step (111) of processing, by a calculation device, at least one satellite photograph of the area.
5. The method (100) of claim 4, wherein the step (111) of processing at least one satellite photograph of the area implements a trained machine learning model.
6. Method (100) according to one of claims 1 to 5, in which the step (115) of determining wind movements is carried out as a function of at least one topography value determined during the determination step (133).
7. Method (100) according to one of claims 1 to 6, in which the step (120) of determining a parametric value representative of a water content comprises: - a step (121) of determining, by a calculation device, an index representative of the water content of the vegetation on a map of the area according to at least three classes, - a step (122) of locating, by a calculation device, each class at a minimum humidity value from the fuel cut-off network and - a step (123) of generating, by a calculation device, a map of the minimum water content in at least three classes.
8. Method (100) according to one of claims 1 to 7, in which the step (125) of calculating a parametric value representative of a propagation of a fire in the zone comprises: - a step (126) of determining, by a calculation device, a speed of propagation of the fire, - a step (127) of determining, by a calculation device, an intensity of the fire, - a step (128) of determining, by a calculation device, a value representative of a topography in a photograph of the area and - a step (129) of determining, by a calculation device, a value representative of a wind speed and direction in a photograph of the area, the step of calculating a parametric value representative of a fire propagation being carried out as a function of the determined propagation speed, the determined fire intensity, at least one determined topography value and at least one determined speed and direction value.
9. Method (100) according to one of claims 1 to 8, in which the step (130) of estimating a parametric value representative of a probability of occurrence of a fire comprises: - a step (131) of determining, by a calculation device, a probable occurrence value based on historical data for the area and - a step (132) of modeling, by a calculation device, a reference fire contour based on data representative of other zones.
10. System (200) for mapping a forest fire risk, characterized in that it comprises: - a means (205) for detecting at least one type of fuel in a geographical area to be mapped, - a means (210) for determining, by a calculation device, a parametric value representative of the presence of at least one fuel detected in at least one photograph of the area, - a means (215) for determining, by a calculation device, a parametric value representative of wind movements in the area, - a means (220) for determining, by a calculation device, a parametric value representative of a water content of the area, - a means (225) for calculating, by a calculation device, a parametric value representative of a propagation of a fire in the area, - a means (230) for estimating, by a calculation device, a parametric value representative of a probability of occurrence of a fire in the area, - a means (233) of determining, by a calculation device, a value representative of the topography of the zone, - a means (235) of calculating, by a calculating device, at least one value representative of a risk of forest fire in the area as a function of: - the parametric value representative of the presence of at least one detected fuel, - the parametric value representative of wind movements, - the parametric value representative of a water content, - the parametric value representative of the topography, - the parametric value representative of the spread of a fire and - the parametric value representing a probability of occurrence of a fire, - a means (240) of associating, by a calculation device, at least one value representative of a risk of forest fire with geographical coordinates and - a means (245) for providing, on an interface, a map representing at least one value representative of a forest fire risk produced as a function of the result of the association step.