阅读说明：本技术 农林草复合区分区火灾预测预警方法、装置与电子设备 (agricultural, forest and grass composite district fire prediction and early warning method and device and electronic equipment ) 是由 杨邦会 王树东 王春红 殷健 梁文秀 温莹莹 孙宁 胡乔利 刘利 于 2019-08-28 设计创作，主要内容包括：本发明实施例提供一种农林草复合区分区火灾预测预警方法、装置与电子设备,其中该方法包括：基于目标农林草复合区不同时间、空间和光谱分辨率的多或高光谱数据以及热红外数据,提取所述目标农林草复合区的热红外异常点；基于所述目标农林草复合区不同中高空间分辨率的多或高光谱数据,利用燃烧痕迹法,估算所述热红外异常点对应的火点发生时间和火点燃烧面积；基于所述火点发生时间和火点燃烧面积,结合分别统计的所述目标农林草复合区的历史农、林、草区域火灾发生总点数、火灾发生强度和火灾发生频率,对所述目标农林草复合区内的农、林、草区域分区进行火灾预测预警。本发明实施例能够有效提高数据提取的精度和预警效果。(The embodiment of the invention provides a fire prediction and early warning method, a device and electronic equipment for an agricultural, forestry and grasses composite subarea, wherein the method comprises the following steps: extracting thermal infrared abnormal points of the target agricultural forest and grass composite area based on multi-or hyperspectral data and thermal infrared data of different time, space and spectral resolutions of the target agricultural forest and grass composite area; based on multi-or hyperspectral data with different medium and high spatial resolutions in the target agricultural and forest grass composite area, estimating the fire point generation time and the fire point burning area corresponding to the thermal infrared abnormal point by using a burning trace method; and based on the fire point generation time and the fire burning area, carrying out fire prediction and early warning on the agricultural, forest and grass region subareas in the target agricultural, forest and grass composite region by combining the historical agricultural, forest and grass region fire occurrence total points, fire occurrence intensity and fire occurrence frequency of the target agricultural, forest and grass composite region which are respectively counted. The embodiment of the invention can effectively improve the precision of data extraction and the early warning effect.)

1. A fire prediction and early warning method for an agricultural, forestry and grass composite zoning is characterized by comprising the following steps:

Based on spectral data and thermal infrared data of different time, space and spectral resolution of a target agricultural forest and grass composite area, extracting risk space-time information aiming at agricultural, forest and grass areas respectively, extracting underlying surface information from a heat sensitive area of the target agricultural forest and grass composite area, and extracting thermal infrared abnormal points of the target agricultural forest and grass composite area;

Based on the spectral data of different medium and high spatial resolutions of the target agroforestry composite area, estimating the ignition point generation time and the ignition area of the fire corresponding to the thermal infrared abnormal point by detecting the combustion trace in the target agroforestry composite area and determining the time for detecting the combustion trace;

Based on the fire point generation time and the fire ignition burning area, combining the historical total fire occurrence points, fire occurrence intensity and fire occurrence frequency of the agricultural, forestry and grass areas in the target agricultural, forestry and grass composite area which are respectively counted, and carrying out fire prediction and early warning on the agricultural, forestry and grass areas in the target agricultural, forestry and grass composite area;

the spectral data is specifically multispectral data or hyperspectral data.

2. The fire prediction and early warning method for the agricultural, forestry and grass composite district according to claim 1, wherein the step of extracting the thermal infrared anomaly point of the target agricultural, forestry and grass composite district specifically comprises:

simulating and generating a phenological trend curve of the target agricultural, forestry and grass composite area based on thermal infrared data of the target agricultural, forestry and grass composite area at different time, space and spectral resolution, and removing fire background information and extracting risk space-time information respectively aiming at agricultural, forestry and grass areas by using a set rule based on the phenological trend curve;

Extracting the underlying surface information of the heat sensitive area based on the spectral data with medium and high resolution of the target agricultural and forest grass composite area, and carrying out vectorization processing on the underlying surface information;

and extracting the thermal infrared abnormal points based on the risk spatio-temporal information and the vectorized underlying surface information.

3. the fire prediction and early warning method for the agricultural, forestry and grass composite zoning according to claim 1 or 2, wherein the step of estimating the fire point generating time and the fire point burning area corresponding to the thermal infrared abnormal point specifically comprises the following steps:

detecting the combustion trace by using a combustion trace index BSI model based on the spectral data with different medium and high spatial resolutions, and determining the time point of the detected combustion trace;

And if the time of the occurrence of the thermal infrared abnormal point is before the time point of the combustion trace, and the interval duration between the time of the occurrence and the time point of the combustion trace is less than a set threshold value, determining that the time of the occurrence is the ignition point generation time of the thermal infrared abnormal point, and taking the combustion area of the thermal infrared abnormal point as the ignition combustion area of the fire.

4. the fire prediction and early warning method for the agricultural, forestry and grass composite district according to claim 1 or 2, wherein the step of performing fire prediction and early warning on the agricultural, forestry and grass district subareas in the target agricultural, forestry and grass composite district specifically comprises:

Determining the maximum number of spots of combustion traces of each monitoring unit, the maximum intensity of fire and the maximum frequency of fire in the target agro-forestry-grass composite area based on the fire point generation time and the fire ignition area;

Based on land utilization information, the maximum patch number of combustion traces of each monitoring unit, the maximum intensity of fire occurrence and the maximum frequency of fire occurrence, as well as the total fire occurrence number, the fire occurrence intensity and the fire occurrence frequency of the historical agriculture, forestry and grassland areas, respectively determining a straw burning early warning index, a forest land fire early warning index, a grassland fire early warning index, an extreme event fire early warning index and a comprehensive fire early warning index, and realizing the subarea joint fire prediction early warning of the agriculture, forestry and grassland areas in the target agriculture, forestry and grassland composite area.

5. The fire prediction and early warning method for the agricultural, forestry and grasses composite subarea is characterized in that the thermal infrared data is particularly high-time-resolution remote sensing data;

Correspondingly, the step of generating a phenological trend curve of the target agriculture, forestry and grasses composite area in a simulated mode specifically comprises the following steps of:

Calculating the NDVI of the time sequence of the target agricultural forest and grass composite area based on the remote sensing data with high time resolution, and simulating the NDVI of the time sequence to generate the phenological trend curve;

Dividing the phenological trend curve into a plurality of segments according to the trend of the phenological trend curve, wherein the plurality of segments comprise: starting to a rapid ascending section, stabilizing a plateau section of the curve and rapidly descending to a bottom section, wherein the slope of each section meets a set condition;

Determining a key time node of the phenological trend curve according to the end point of each segment comprises: a crop planting starting point, a boundary point which rapidly rises to a stable plateau area, a middle point of the stable plateau area, a boundary point which is from the plateau area to growth descending and a boundary point which rapidly descends to a bottom point;

the steps of removing fire background information and extracting risk spatio-temporal information respectively aiming at agricultural, forest and grass areas specifically comprise:

And respectively setting discrimination conditions based on the key time nodes aiming at the agricultural, forest and grass areas, and removing the fire background information and extracting the risk space-time information by using the discrimination conditions.

6. the fire prediction and early warning method for the agricultural, forestry and grass composite zoning according to claim 5, wherein the step of extracting the thermal infrared abnormal point specifically comprises the following steps:

superposing the risk space-time information and vectorized underlying surface information, and calculating the average temperature of the target agricultural, forest and grass composite area according to the superposition result;

and extracting pixels with abnormal temperature to form the thermal infrared abnormal point based on the average temperature and the actual temperature of each pixel point in the target agricultural forest and grass composite area.

7. the fire prediction and early warning method for the agriculture, forestry and grasses composite district according to claim 4, wherein before the step of performing fire prediction and early warning on the agriculture, forestry and grasses district in the target agriculture, forestry and grasses composite district, the method further comprises:

respectively aiming at a farmland area, an agricultural and forestry grass area and a forest and grass area, counting the total fire occurrence points according to the following formula:

wherein Pj represents the number of patches of combustion traces, each patch is a fire point, Δ pi represents the number of fire points newly added in the i-th year compared with the i-1 th year, when i is 1, Δ pi is 0, p1 represents the number of fire points in the 1 st year of statistics, n represents the total number of years of statistics, and sj represents the area of a statistic unit;

Respectively aiming at a farmland area, an agricultural and forestry grass area and a forest and grass area, calculating the fire occurrence intensity according to the following formula:

wherein Ppj represents the intensity of fire occurrence of statistical unit j, Δ spi represents the area of the newly added fire ignition trace patch in the i-th year compared with the i-1 th year, when i is 1, Δ spi is 0, sp1 represents the area of the statistical 1 st year fire trace patch, n represents the total number of years of statistics, and sj represents the area of statistical unit;

Respectively aiming at a farmland area, a farming, forestry and grassy area and a forest and grassy area, counting the fire occurrence frequency according to the following formula:

In the formula, Fpj represents the frequency of fire occurrence, Δ pi represents the number of fire points newly added in the ith year compared with the ith-1 st year, when i is 1, Δ pi is 0, p1 represents the number of fire points in the 1 st year counted, n represents the total number of years counted, and sj represents the area of the counting unit.

8. The utility model provides a fire prediction early warning device of combined subregion of agriculture, forestry and grasses which characterized in that includes:

the first processing module is used for extracting risk space-time information aiming at the agricultural, forest and grass areas respectively and extracting underlying surface information from a heat sensitive area of the target agricultural, forest and grass composite area based on spectral data and thermal infrared data of different time, space and spectral resolution of the target agricultural, forest and grass composite area, and extracting thermal infrared abnormal points of the target agricultural, forest and grass composite area;

the second processing module is used for estimating the ignition point generation time and the ignition burning area corresponding to the thermal infrared abnormal point by detecting the burning trace in the target agriculture, forestry and grasses composite area and determining the time for detecting the burning trace based on the spectral data of different medium and high spatial resolutions of the target agriculture, forestry and grasses composite area;

the prediction early warning module is used for carrying out fire prediction early warning on the agricultural, forest and grass region subareas in the target agricultural, forest and grass composite region by combining the historical agricultural, forest and grass region fire occurrence total points, fire occurrence intensity and fire occurrence frequency of the target agricultural, forest and grass composite region which are respectively counted;

The spectral data is specifically multispectral data or hyperspectral data.

9. An electronic device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method of any one of claims 1 to 7.

10. A non-transitory computer readable storage medium having stored thereon computer instructions, wherein the computer instructions, when executed by a computer, implement the steps of the agroforestry compound zoned fire prediction warning method of any one of claims 1 to 7.

Technical Field

the invention relates to the technical field of ecology and remote sensing, in particular to a fire prediction and early warning method, a fire prediction and early warning device and electronic equipment for an agricultural, forestry and grass composite zoning area.

Background

in addition to air pollution and destruction of the ecosystem, the straw burning in farmland, accidental or accidental fires in grassland and forests may cause loss of population and property. In recent years, fire disasters caused by people, climate and the like, particularly fire disasters in agriculture, forestry and grassy composite areas frequently occur, however, driving factors of fire disasters in farmlands and forestry and grassy areas are different, and differences of time periods and the like of fire disasters in the farmlands and the forestry and grassy areas are obvious, so that classification prediction and early warning of the fire disasters are very important.

And (4) carrying out fire prediction and early warning in the agriculture, forestry and grasses composite subareas, and obtaining information such as time, place, burning area, frequency and the like of fire occurrence from historical data. In order to accurately obtain the information, it is necessary to obtain the time point of combustion through thermal infrared or obtain the information of combustion trace through multispectral data, and then estimate the information of the time, place, intensity, etc. of the fire. Therefore, there are some methods of obtaining the occurrence time of a fire mainly by the identification of the thermal infrared abnormality information and some prediction and early warning methods based on the combustion trace in the existing methods.

However, due to the complexity of land and objects in the agricultural, forestry and grassy composite area, the accuracy of the fire point obtained by the thermal infrared-based method is not high, and the accuracy of the obtained result is very limited because the method mostly adopts remote sensing data with high time and low spatial resolution. Due to the complexity of background information of an agricultural, forestry and grass composite area, the applicability and the accuracy of a model in the existing combustion trace-based method have great uncertainty. Therefore, the accuracy of the ignition point obtained by only considering the thermal infrared sensor or the traditional combustion trace method has great uncertainty, and the accuracy of data extraction and the early warning effect are seriously influenced.

Disclosure of Invention

in order to overcome the above problems or at least partially solve the above problems, embodiments of the present invention provide a method, an apparatus, and an electronic device for predicting and warning a fire in an agroforestry composite zoning area, so as to effectively overcome the uncertainty in the fire prediction and warning in the agroforestry composite zoning area in the prior art, thereby effectively improving the accuracy of data extraction and the warning effect.

In a first aspect, an embodiment of the present invention provides a fire prediction and early warning method for an agricultural, forestry and grasses composite zoning area, including:

based on spectral data and thermal infrared data of different time, space and spectral resolution of a target agricultural forest and grass composite area, extracting risk space-time information aiming at agricultural, forest and grass areas respectively, extracting underlying surface information from a heat sensitive area of the target agricultural forest and grass composite area, and extracting thermal infrared abnormal points of the target agricultural forest and grass composite area;

based on the spectral data of different medium and high spatial resolutions of the target agroforestry composite area, estimating the ignition point generation time and the ignition area of the fire corresponding to the thermal infrared abnormal point by detecting the combustion trace in the target agroforestry composite area and determining the time for detecting the combustion trace;

Based on the fire point generation time and the fire ignition burning area, combining the historical total fire occurrence points, fire occurrence intensity and fire occurrence frequency of the agricultural, forestry and grass areas in the target agricultural, forestry and grass composite area which are respectively counted, and carrying out fire prediction and early warning on the agricultural, forestry and grass areas in the target agricultural, forestry and grass composite area;

The spectral data is specifically multispectral data or hyperspectral data.

In a second aspect, an embodiment of the present invention provides a fire prediction and early warning device for an agriculture, forestry and grasses composite zoning area, including:

The first processing module is used for extracting risk space-time information aiming at the agricultural, forest and grass areas respectively and extracting underlying surface information from a heat sensitive area of the target agricultural, forest and grass composite area based on spectral data and thermal infrared data of different time, space and spectral resolution of the target agricultural, forest and grass composite area, and extracting thermal infrared abnormal points of the target agricultural, forest and grass composite area;

the second processing module is used for estimating the ignition point generation time and the ignition burning area corresponding to the thermal infrared abnormal point by detecting the burning trace in the target agriculture, forestry and grasses composite area and determining the time for detecting the burning trace based on the spectral data of different medium and high spatial resolutions of the target agriculture, forestry and grasses composite area;

The prediction early warning module is used for carrying out fire prediction early warning on the agricultural, forest and grass region subareas in the target agricultural, forest and grass composite region by combining the historical agricultural, forest and grass region fire occurrence total points, fire occurrence intensity and fire occurrence frequency of the target agricultural, forest and grass composite region which are respectively counted;

the spectral data is specifically multispectral data or hyperspectral data.

In a third aspect, an embodiment of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor executes the computer program to implement the steps of the fire prediction and early warning method for agriculture, forestry, and grasses compound zoning.

In a fourth aspect, embodiments of the present invention provide a non-transitory computer-readable storage medium, on which computer instructions are stored, and when the computer instructions are executed by a computer, the method implements the steps of the agriculture, forestry and grasses composite zoning fire prediction and early warning method according to the first aspect.

According to the method, the device and the electronic equipment for predicting and early warning the fire in the agriculture, forestry and grass composite subareas, provided by the embodiment of the invention, the distribution and change characteristics of the area of combustion trace patches are obtained by cooperatively fusing thermal infrared time points and multispectral data through a plurality of sensors, so that the prediction and early warning of the fire in the agriculture, forestry and grass composite subareas can be realized, the uncertainty in the fire prediction and early warning in the agriculture, forestry and grass composite subareas in the prior art can be effectively overcome, and the accuracy of data extraction and the early warning effect are effectively 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, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.

FIG. 1 is a schematic flow chart of a fire prediction and early warning method for an agricultural, forestry and grasses composite zoning area provided by the embodiment of the invention;

FIG. 2 is a schematic structural diagram of a fire prediction and early warning device for an agricultural, forestry and grasses composite zoning area provided by the embodiment of the invention;

Fig. 3 is a schematic physical structure diagram of an electronic device according to an embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments obtained by persons of ordinary skill in the art based on the embodiments of the present invention without any creative efforts belong to the protection scope of the embodiments of the present invention.

The method aims at solving the problems that in the prior art, the method for obtaining fire point precision and traditional combustion traces by simply considering the thermal infrared sensor has great uncertainty and seriously influences the precision of data extraction and the early warning effect, and how to effectively combine thermal infrared and multispectral multisource data and optimize a corresponding model and method is very key for effectively extracting fire information of an agriculture, forestry and grassy composite area. According to the embodiment of the invention, the distribution and change characteristics of the combustion trace patch area are obtained by cooperatively fusing the thermal infrared time point and multispectral data through the multiple sensors, and the prediction and early warning of the fire in the agriculture, forestry and grass composite district are realized, so that the uncertainty in the prediction and early warning of the fire in the agriculture, forestry and grass composite district in the prior art can be effectively overcome, and the precision of data extraction and the early warning effect are effectively improved. Embodiments of the present invention will be described and illustrated with reference to various embodiments.

fig. 1 is a schematic flow chart of a fire prediction and early warning method for an agricultural, forestry and grasses composite zoning area provided by the embodiment of the invention, as shown in fig. 1, the method includes:

s101, based on spectral data and thermal infrared data of different time, space and spectral resolution of a target agricultural, forest and grass composite area, extracting risk space-time information respectively for the agricultural, forest and grass areas, extracting underlying surface information for a heat sensitive area of the target agricultural, forest and grass composite area, and extracting thermal infrared abnormal points of the target agricultural, forest and grass composite area, wherein the spectral data is specifically multispectral data or hyperspectral data.

it can be understood that, in the embodiment of the present invention, the abnormal information of the target agricultural and forest grass composite area, that is, the thermal infrared abnormal point, is extracted according to the acquired multispectral data or hyperspectral data and thermal infrared data of the target agricultural and forest grass composite area. Specifically, the embodiment of the invention acquires data of the target agricultural, forest and grass composite area in advance, wherein the data comprises multispectral data or hyperspectral data with different time, space and spectral resolutions and thermal infrared data. In particular, the thermal infrared data is remote sensing data with high time resolution, and the multi-or hyperspectral data with medium and high spatial resolution is different data sources with medium and high resolution, such as remote sensing data with medium and high resolution. The system also comprises other multi-source data, including DEM, land utilization data, meteorological data, microwave data, soil data and the like.

And then, extracting risk space-time information respectively aiming at the agricultural, forest and grass areas by generating a smooth phenological trend curve by utilizing the thermal infrared data acquired in advance. And simultaneously, extracting the underlying surface information of a heat sensitive area in a target agriculture, forestry and grasses composite area according to the obtained multi-or hyperspectral data, such as extracting the underlying surface information of asphalt roads, roofs, sand lands, cement roads and the like, water bodies, pools and the like for medium-high resolution remote sensing data.

And then, overlapping the respectively obtained risk space-time information and underlying surface information of the heat sensitive area, and identifying and extracting thermal infrared abnormal points in the target agricultural and forest grass composite area by monitoring the temperature of all pixels in the remote sensing image of the target agricultural and forest grass composite area.

s102, based on spectral data of different medium and high spatial resolutions of the target agroforestry composite area, by detecting a combustion trace in the target agroforestry composite area and determining the time of detecting the combustion trace, the fire point generation time and the fire ignition burning area corresponding to the thermal infrared abnormal point are estimated, wherein the spectral data are specifically multispectral data or hyperspectral data.

it is understood that the embodiment of the present invention combines the combustion trace method to determine the ignition point generation time of combustion and the ignition area of combustion on the basis of the above-described processing. Specifically, for multispectral data or hyperspectral data (remote sensing data) with different medium and high spatial resolutions, considering the difference that a plurality of sensors need to cooperatively extract combustion traces and sensor band settings, the embodiment of the invention utilizes a corresponding model to detect the combustion traces, and can determine the time point at which the combustion traces can be detected by utilizing the model, thereby determining the occurrence time of the ignition point corresponding to the thermal infrared abnormal point, and simultaneously obtaining the combustion area of the ignition point according to the combustion traces.

And S103, based on the fire point generation time and the fire burning area, combining the historical total fire occurrence points, fire occurrence intensity and fire occurrence frequency of the agricultural, forestry and grassy areas in the target agricultural, forestry and grassy composite area which are counted respectively, and performing fire prediction and early warning on the agricultural, forestry and grassy areas in the target agricultural, forestry and grassy composite area.

It can be understood that, finally, according to the obtained ignition point occurrence time of each ignition point and the ignition burning area of the fire caused by the ignition point, indexes such as the maximum patch number of burning traces of all monitoring units in the target agriculture, forestry and grasses composite area, the maximum intensity of fire occurrence, the maximum frequency of fire occurrence and the like can be determined. Meanwhile, fire occurrence indexes in any historical time period can be counted according to the fire occurrence conditions in the past year, and the number of fire occurrences, the fire occurrence intensity and the fire occurrence frequency in the historical time period in the past year are specifically counted according to a farmland area, an agricultural and forestry grass area and a forest and grass area. And then respectively determining fire early warning indexes corresponding to farmlands, woodlands and grasslands according to the obtained indexes such as the maximum patch number of the combustion traces of the monitoring units, the maximum intensity of fire occurrence, the maximum frequency of fire occurrence and the like, the number of fire occurrence points, the intensity of fire occurrence and the frequency of fire occurrence in the historical period of years, and accordingly realizing the regional fire prediction and early warning of the target agriculture, forestry and grassland composite area. If the early warning level is set according to the early warning index, and the regional fire prediction early warning of the target agricultural forest and grass composite area is realized according to different early warning levels.

according to the agricultural, forestry and grass composite district fire prediction and early warning method provided by the embodiment of the invention, the distribution and change characteristics of combustion trace patch areas are obtained by cooperatively fusing thermal infrared time points and multispectral data through a plurality of sensors, so that the agricultural, forestry and grass composite district fire prediction and early warning are realized, the uncertainty in the agricultural, forestry and grass composite district fire prediction and early warning in the prior art can be effectively overcome, and the accuracy and the early warning effect of data extraction are effectively improved.

Optionally, the step of extracting the thermal infrared abnormal point of the target agricultural and forest grass composite area specifically includes: simulating and generating a phenological trend curve of the target agricultural, forestry and grass composite area based on thermal infrared data of the target agricultural, forestry and grass composite area at different time, space and spectral resolution, and removing fire background information and extracting risk spatio-temporal information respectively aiming at agricultural, forestry and grass areas by using a set rule based on the phenological trend curve; extracting underlying surface information of a heat sensitive area based on high-resolution multispectral data or hyperspectral data of a target agricultural forest and grass composite area, and carrying out vectorization processing on the underlying surface information; and extracting thermal infrared abnormal points based on the risk space-time information and the vectorized underlying surface information.

Specifically, the NDVI is calculated according to the thermal infrared data of the target agricultural and forest grass composite area in different time, space and spectral resolutions, and a smooth trend curve, namely a phenological trend curve of the target agricultural and forest grass composite area, is generated in a simulation mode according to the calculation result. And then, extracting key time nodes according to the phenological trend curve, and removing fire background information and extracting risk spatiotemporal information aiming at the agricultural, forest and grass areas respectively according to the key time nodes. And simultaneously, extracting underlay surface information of a heat sensitive area in the target agricultural and forestry and grassy composite area, for example, extracting underlay surface information of an asphalt road, a roof, a sand ground, a cement road and the like, a water body, a pit and the like for medium-high resolution remote sensing data, vectorizing the underlay surface information, and establishing different planar vector layers. And finally, judging each pixel of the remote sensing data of the target agriculture, forestry and grasses composite area by combining risk space-time information and vectorized underlying surface information, and extracting abnormal points, namely thermal infrared abnormal points.

Optionally, according to the above embodiments, the thermal infrared data is specifically remote sensing data with high time resolution; correspondingly, the step of simulating and generating the phenological trend curve of the target agriculture, forestry and grasses composite area specifically comprises the following steps of: calculating the NDVI of the time sequence of the target agricultural forest and grass composite area based on the remote sensing data with high time resolution, and simulating the NDVI of the time sequence to generate a phenological trend curve; according to the trend of the object and climate trend curve, the object and climate trend curve is divided into a plurality of segments, and the plurality of segments comprise: starting to a rapid ascending section, stabilizing a plateau section of the curve and rapidly descending to a bottom section, wherein the slope of each section meets a set condition; determining the key time node of the phenological trend curve according to the end point of each segment comprises the following steps: the method comprises the following steps of starting crop planting, rapidly ascending to a stable plateau area boundary point, a stable plateau area midpoint, a plateau area to growth descending boundary point and a rapidly descending to bottom point boundary point.

that is to say, the embodiment of the invention generates the phenological trend curve of crop growth according to the remote sensing data with high time resolution, and identifies the key time node in the phenological trend curve. Therefore, NDVI of time sequences is calculated according to the remote sensing data with high time resolution of the target agricultural forest and grass composite area, then a smooth phenological trend curve is generated according to the NDVI values of the time sequences, and the simulated phenological trend curve is divided into 3 segments: namely an initial to fast rise F1(NDVI), a plateau F2(NDVI) and a fast fall to bottom F3(NDVI), and the slopes of these segments should satisfy the following conditions: f1 ' (NDVI) > a1, F2 ' (NDVI) ≈ 0, F3 ' (NDVI) < a2, where a1, a2 are both given thresholds. Finally, determining each key time node of the phenological trend curve according to the endpoints and the demarcation points of the segments, wherein the key time nodes comprise: the method comprises the following steps of a crop planting starting point t1, a boundary point t2 which rises rapidly to a stable plateau area, a middle point t3 of the stable plateau area, a boundary point t4 which falls from the plateau area to a growth descending point and a boundary point t5 which falls rapidly to a bottom point.

accordingly, the steps of removing the fire background information and extracting the risk spatiotemporal information respectively for the agricultural, forestry and grassy regions specifically include: and respectively setting discrimination conditions based on key time nodes aiming at agricultural, forest and grass areas, removing fire background information and extracting risk spatio-temporal information by using the discrimination conditions.

specifically, based on the above processing, removal of the background information of the fire in the agriculture, forestry and grasses and extraction of the risk spatiotemporal information are performed. Firstly, extracting fire risk space-time information of crops in one season and two seasons for a farmland.

firstly, the background information removal and risk spatio-temporal information extraction of main grain crops harvested in alternate spring and summer meet the following conditions:

in the formula, NDVIst1, NDVIst2, st2, st1 and a3 are an NDVI value of a crop planting starting point t1, an NDVI value which rapidly rises to a stable plateau area boundary point t2, a time node corresponding to t2, a time node corresponding to t1 and a third preset threshold, and NDVIst4, NDVIst5, st4, st5 and a4 are an NDVI value of a boundary point t4 from a plateau area to a growth decline, an NDVI value of a boundary point t5 which rapidly falls to a bottom point, a time node corresponding to t4, a time node corresponding to t5 and a fourth preset threshold, respectively.

That is, when it is detected that the vegetation growth characteristics satisfy the above conditions, it may be determined as a spring and summer alternate crop, and may be monitored as a key monitoring object.

Secondly, background information removal and risk spatio-temporal information extraction of main grain crops harvested in autumn meet the following conditions:

in the formula, ndvia 1, ndvia 2, at2, at1 and a5 are an NDVI value of a crop planting starting point t1, an NDVI value of a plateau area boundary t2 which rises rapidly to be stable, a time node corresponding to t2, a time node corresponding to t1 and a fifth preset threshold, and ndvia 4, ndvia 5, at4, at5 and a6 are an NDVI value of a plateau area to a boundary point t4 which grows and drops rapidly to a bottom point, an NDVI value of a boundary point t5 which drops rapidly to a bottom point, a time node corresponding to t4, a time node corresponding to t5 and a sixth preset threshold, respectively.

Secondly, for grasslands and woodlands, the distribution information of the grasslands and the woodlands is obtained from the land utilization information, and then fire background information removal and risk space-time information extraction are respectively carried out.

Firstly, grassland fire background information removal and risk spatio-temporal information extraction need to simultaneously meet the following three conditions:

In the formula, NDVIgt1, NDVIgt2, gt1, gt2 and a7 are an NDVI value of a grassland growth starting point t1, an NDVI value rapidly rising to a stable plateau area boundary point t2, a time node corresponding to t1, a time node corresponding to t2 and a seventh preset threshold respectively, NDVIgt4, NDVIgt5, gt4, gt5 and a8 are an NDVI value of a boundary point t4 from a plateau area to a growth decline, an NDVI value of a boundary point t5 rapidly falling to a bottom point, a time node corresponding to t4, a time node corresponding to t5 and an eighth preset threshold respectively, NDVIgt3-gt4 is an NDVI mean value of a corresponding time period from t3 to t4, and b1 is a preset threshold.

Secondly, forest fire background information removal and risk spatio-temporal information need to simultaneously meet the following three conditions:

In the formula, NDVIft1, NDVIft2, ft1, ft2 and a9 are the NDVI value of the forest land growth starting point t1, the NDVI value of the plateau area boundary point t2 which rises rapidly to be stable, the time node corresponding to t1, the time node corresponding to t2 and a ninth preset threshold respectively, NDVIft4, NDVIft5, ft4, ft5 and a10 are the NDVI value of the boundary point t4 from the plateau area to the growth descent, the NDVI value of the boundary point t5 which falls rapidly to the bottom point, the time node corresponding to t4, the time node corresponding to t5 and the tenth preset threshold respectively, NDVIft 3-4 is the NDVI mean value in the corresponding time period t3 to t4, and b2 is the preset threshold.

optionally, according to the above embodiments, the step of extracting the thermal infrared anomaly point specifically includes: superposing the risk space-time information and the vectorized underlying surface information, and calculating the average temperature of the target agricultural, forest and grass composite area according to the superposition result; and extracting pixels with abnormal temperature to form thermal infrared abnormal points based on the average temperature and the actual temperature of each pixel point in the target agricultural forest and grass composite area.

Specifically, first, the thermal sensitive underlying surface information and the thermal infrared information are superimposed. The thermal infrared information is surface temperature information obtained by thermal infrared remote sensing data, and represents the risk space-time information.

and secondly, calculating the average temperature of the target agricultural and forest grass composite area.

again, the temperature difference is calculated: in the formula, Tss is a pixel temperature difference value, Ts is a pixel actual temperature value, and is an area pixel temperature mean value.

when the temperature of any one pixel element meets Tss + delta omega i > b3, the pixel element is determined to be a temperature abnormal pixel element. Wherein Δ ω i is an additional value of the temperature-sensitive ground object, and is related to the type and occupied area of the temperature-sensitive ground object in the thermal infrared pixel, when i is 1, the asphalt road is asphalt road, 2 is sand ground, 3 is cement ground, 4 is water, 5 is general mixed artificial ground, and b3 is a preset threshold.

Optionally according to the above embodiments, the step of estimating the fire point occurrence time and the fire point burning area corresponding to the thermal infrared abnormal point specifically includes: detecting combustion traces by using a BSI maturity model based on spectral data with different medium and high spatial resolutions, and determining time points of the detected combustion traces; if the time of the thermal infrared abnormal point is before the time point of the combustion trace, and the interval duration between the time of the thermal infrared abnormal point and the time point of the combustion trace is less than a set threshold, determining that the time of the thermal infrared abnormal point is the time of the ignition point of the thermal infrared abnormal point, and taking the combustion area of the thermal infrared abnormal point as the area of the ignition point, wherein the spectral data is specifically multispectral data or hyperspectral data.

Specifically, for different remote sensing data with medium and high spatial resolutions, considering the difference of the settings of the combustion traces and the sensor wave bands which need to be extracted by the multi-sensor cooperation, the selected model is a BSI model, namely:

in the formula, BSI, Rnir, Rb and Rr are respectively a burning trace index, a near infrared band reflectivity, a blue light band reflectivity and a red light band reflectivity, c1, c2 and c3 are coefficients which can be respectively 1.9, 1.2 and 0.6, and a11 is an eleventh preset threshold.

then, for the extracted temperature anomaly point, the time when the temperature anomaly point appears is set to be rt1, after the time point, the time when the medium and high resolution remote sensing data is acquired is set to be rt2, and if the conditions are met: and (3) applying a combustion trace detected by a BSI model at the time point of rt2 to the remote sensing image with medium and high resolution, and determining that the time of the fire point is rt1 and simultaneously obtaining the combustion area of the fire point when the Δ rt is equal to rt2-rt1 < d 4. Where Δ rt is the duration of the interval of rt1 and rt2, and d4 is a preset threshold.

It is understood that in the case of an abnormal point found in the absence of thermal infrared, the fire point occurrence time can also be estimated from the occurrence time point of the combustion trace patch. Specifically, the time of occurrence of the fire point is estimated by the time point of occurrence of the combustion trace: let bt2 be the time point when the burning trace plaque is detected, let bt1 be the time node of repeated coverage of the remote sensing image before the time (numerically, satellite revisit cycle), and satisfy: Δ bt-bt 2-bt1 < d5, the time range for the fire to occur was determined to be between bt2 and bt1, while the area of the combustion trace was obtained. Where Δ bt is the duration of the interval of bt1 and bt2, and d5 is the preset threshold.

On the basis of the above, the estimation of the time point when the fire disaster occurs again at the adjacent point can be carried out. Assuming that the plaque area of the medium combustion traces is detected as Sbt1-t2, if the plaque area Sbt2-t3 is found at the time node bt2, and Δ lbt ═ bt3-bt2 < d6, and Δ s ═ Sbt1-t3-Sbt1-t2 > s3, it is determined that the fire is reoccurring at the same position. Δ lbt is the interval duration of bt2 and bt3, d6 is a preset threshold, Δ s is a patch area difference value, Sbt1-t3 is the area extracted by the high-resolution remote sensing image at the time point st3, Sbt1-t2 is the area extracted by the high-resolution remote sensing image at the time point st2, and s3 is the preset threshold.

in addition, on the basis of the above embodiments, before the step of performing fire prediction and early warning on the agricultural, forestry and grassy region subareas in the target agricultural, forestry and grassy composite region, the method of the embodiment of the present invention may further include:

Respectively aiming at a farmland area, an agricultural and forestry grass area and a forest and grass area, counting the total fire occurrence points according to the following formula:

Wherein Pj represents the number of patches of combustion traces, each patch is a fire point, Δ pi represents the number of fire points newly added in the i-th year compared with the i-1 th year, when i is 1, Δ pi is 0, p1 represents the number of fire points in the 1 st year of statistics, n represents the total number of years of statistics, and sj represents the area of a statistic unit;

Respectively aiming at a farmland area, an agricultural and forestry grass area and a forest and grass area, calculating the fire occurrence intensity according to the following formula:

Wherein Ppj represents the intensity of fire occurrence of statistical unit j, Δ spi represents the area of the newly added fire ignition trace patch in the i-th year compared with the i-1 th year, when i is 1, Δ spi is 0, sp1 represents the area of the statistical 1 st year fire trace patch, n represents the total number of years of statistics, and sj represents the area of statistical unit;

Respectively aiming at a farmland area, a farming, forestry and grassy area and a forest and grassy area, counting the fire occurrence frequency according to the following formula:

In the formula, Fpj represents the frequency of fire occurrence, Δ pi represents the number of fire points newly added in the ith year compared with the ith-1 st year, when i is 1, Δ pi is 0, p1 represents the number of fire points in the 1 st year counted, n represents the total number of years counted, and sj represents the area of the counting unit.

The method for carrying out fire prediction and early warning on the agricultural, forest and grass region subareas in the target agricultural, forest and grass composite region comprises the following steps: determining the maximum number of spots of combustion traces of each monitoring unit, the maximum intensity of fire and the maximum frequency of fire in a target agriculture, forestry and grasses composite area based on the time of each fire and the area of fire ignition; based on land utilization information, the maximum patch number of combustion traces of each monitoring unit, the maximum intensity of fire occurrence and the maximum frequency of fire occurrence, the total number of fire occurrence points, the fire occurrence intensity and the fire occurrence frequency of the historical agriculture, forestry and grass areas are respectively determined, a straw burning early warning index, a forest fire early warning index, a grassland fire early warning index, an extreme event fire early warning index and a comprehensive fire early warning index are respectively determined, and the subarea joint fire prediction early warning of the agriculture, forestry and grass areas in the target agriculture, forestry and grass composite area is realized.

specifically, the embodiment of the invention carries out fire joint early warning on the areas of agriculture, forestry and grasses in the target agricultural, forestry and grasses composite area. Firstly, overall monitoring is carried out according to the time of each fire point and the area of fire point, the maximum number of spots of combustion traces of each monitoring unit, the maximum intensity of fire occurrence and the maximum frequency of fire occurrence are determined, then fire early warning indexes of farmlands, forest lands and grasslands are respectively calculated according to the overall indexes and the counted historical indexes, and the regional fire prediction early warning of the whole target agriculture, forestry and grassland composite area is carried out on the basis. It can be understood that the zoning early warning can be realized according to different fallen leaf periods of crops, woodlands and grasslands. And expressing the characteristic growth time of different types of vegetation by using delta ct, delta ft, delta gt and the like, and calculating fire early warning indexes aiming at crops, forest lands and grasslands respectively.

Firstly, for agricultural-forestry-grassland fire early warning caused by farmland straw burning, the fire in the crop harvesting period is mainly considered:

In the formula, the EWI delta ct is a straw burning early warning index in a delta ct time period, Pj1, Ppj1 and Fpj1 are the number of spots of a burning trace of a monitoring unit j, the intensity of fire and the frequency of fire occurrence in the delta ct time period, the maximum number of spots of the burning trace of each monitoring unit Pmax, Ppmax and Fpmax, the maximum intensity of fire and the maximum frequency of fire occurrence, and the delta ct is a crop harvesting period.

If the farmland is identified to be connected with the burning trace patch of the forest and grass before the crop harvest time to the planting of the next year, the farmland is considered to be a farmland fire.

Secondly, the forest land caused agriculture-forestry-grassland fire hazard early warning:

in the formula, EWI Δ ft is a forest fire early warning index in a Δ ft time period, Pj2, Ppj2 and Fpj2 are the number of spots of combustion traces of a monitoring unit j, the intensity of fire occurrence and the fire occurrence frequency in the Δ ft time period, the maximum number of spots of combustion traces of each monitoring unit Pmax, Ppmax and Fpmax, the maximum intensity of fire occurrence and the maximum frequency of fire occurrence respectively, and Δ ft is the characteristic growth time of forest vegetation.

Thirdly, carrying out agricultural-forest-grassland fire early warning on grasslands:

In the formula, EWI Δ gt is a grassland fire early warning index in a Δ gt time period, Pj3, Ppj3 and Fpj3 are the number of burning traces of the monitoring unit j, the intensity of fire occurrence and the fire occurrence frequency in the Δ gt time period, the maximum number of burning traces of each monitoring unit Pmax, Ppmax and Fpmax, the maximum intensity of fire occurrence and the maximum frequency of fire occurrence, and Δ gt is the characteristic growth time of grassland vegetation.

On the basis, fire early warning of extreme climatic events and comprehensive fire early warning can be carried out.

wherein, for extreme weather event early warning:

In the formula, EWI Δ kt is an extreme event fire early warning index in a Δ kt time period, Pj4, Ppj4 and Fpj4 are the number of spots of combustion traces of a monitoring unit j, the intensity of fire occurrence and the fire occurrence frequency in the Δ kt time period, respectively, the maximum number of spots of combustion traces of each monitoring unit Pmax, Ppmax and Fpmax, the maximum intensity of fire occurrence and the maximum frequency of fire occurrence, and Δ kt is an extreme climate easy occurrence time.

For comprehensive early warning:

In the formula, EWI Δ ot is a comprehensive fire early warning index in a Δ ot time period, Pj5, Ppj5 and Fpj5 are the number of spots of combustion traces of a monitoring unit j, the intensity of fire occurrence and the fire occurrence frequency in the Δ ot time period, and the maximum number of spots of combustion traces of each monitoring unit Pmax, Ppmax and Fpmax, the maximum intensity of fire occurrence and the maximum frequency of fire occurrence respectively.

Based on the same concept, the embodiment of the invention provides a fire prediction and early warning device for an agricultural, forestry and grassy composite district according to the above embodiments, and the device is used for realizing the prediction and early warning of the fire of the agricultural, forestry and grassy composite district in the above embodiments. Therefore, the description and definition in the fire prediction and early warning method for agriculture, forestry and grasses composite zoning areas in the embodiments can be used for understanding the execution modules in the embodiments of the present invention, and specific reference may be made to the embodiments, which are not repeated herein.

according to an embodiment of the present invention, a structure of the fire prediction and early warning device for an agriculture, forestry and grasses composite district is shown in fig. 2, which is a schematic structural diagram of the fire prediction and early warning device for an agriculture, forestry and grasses composite district provided by the embodiment of the present invention, and the device can be used for realizing the prediction and early warning of fire in the agriculture, grasses composite district in the above method embodiments, and the device includes: a first processing module 201, a second processing module 202 and a predictive alert module 203. Wherein:

The first processing module 201 is configured to extract risk space-time information for the agricultural, forestry and grass areas respectively based on spectral data and thermal infrared data of different time, space and spectral resolution of the target agricultural, forestry and grass composite area, extract underlying surface information for a thermally sensitive area of the target agricultural, forestry and grass composite area, and extract thermal infrared outliers of the target agricultural, forestry and grass composite area; the second processing module 202 is configured to estimate fire point generation time and fire point burning area corresponding to the thermal infrared abnormal point by detecting a burning trace in the target agroforestry composite area and determining time when the burning trace is detected based on spectral data of different medium and high spatial resolutions of the target agroforestry composite area; the prediction early warning module 203 is configured to perform fire prediction and early warning on the agricultural, forestry and grass region partitions in the target agricultural, forestry and grass composite region based on the fire point occurrence time and the fire point burning area in combination with the historical agricultural, forestry and grass region fire occurrence total points, fire occurrence intensity and fire occurrence frequency of the target agricultural, forestry and grass composite region, which are counted respectively, wherein the spectral data is specifically multispectral data or hyperspectral data.

specifically, the first processing module 201 extracts risk spatiotemporal information for the agricultural, forestry and grassy regions respectively by generating a smooth phenological trend curve using thermal infrared data acquired in advance. And simultaneously, extracting the underlying surface information of a heat sensitive area in the target agriculture, forestry and grasses composite area according to the acquired multispectral data or hyperspectral data, such as for medium-high resolution remote sensing data, extracting the underlying surface information of asphalt roads, roofs, sand lands, cement roads and the like, water bodies, pools and the like. And then the first processing module 201 superposes the obtained risk space-time information and underlying surface information of the thermal sensitive area, and identifies and extracts thermal infrared abnormal points in the target agricultural and forest grass composite area by monitoring the temperature of all pixels in the remote sensing image of the target agricultural and forest grass composite area.

Then, for multispectral data or hyperspectral data (remote sensing data) with different medium and high spatial resolutions, considering the difference that a plurality of sensors need to cooperatively extract combustion traces and the setting of sensor wave bands, the second processing module 202 detects the combustion traces in the multispectral data or hyperspectral data by using corresponding models, determines the time point at which the combustion traces can be detected by using the models, then determines the occurrence time of the ignition point corresponding to the thermal infrared abnormal point according to the time point, and obtains the combustion area of the ignition point according to the combustion traces.

Finally, the prediction and early warning module 203 determines indexes such as the maximum patch number of combustion traces of all monitoring units in the target agriculture, forestry and grasses composite area, the maximum intensity of fire occurrence, the maximum frequency of fire occurrence and the like according to the obtained fire point occurrence time of each fire point and the fire burning area caused by the fire point. Meanwhile, the prediction and early warning module 203 counts fire occurrence indexes in any historical time period according to the fire occurrence conditions in the past year, and specifically counts the number of fire occurrences, the fire occurrence intensity and the fire occurrence frequency in the historical time period in the past year according to a farmland area, a farming and forestry and grassy area and a forestry and grassy area. Then, the prediction and early-warning module 203 determines fire early-warning indexes corresponding to farmlands, woodlands and grasslands respectively according to the obtained indexes such as the maximum patch number of the combustion traces of the monitoring units, the maximum intensity of fire occurrence, the maximum frequency of fire occurrence and the like, and the number of fire occurrence points, the intensity of fire occurrence and the frequency of fire occurrence in the historical period of years, and accordingly realizes the regional fire prediction and early-warning of the target agriculture, forestry and grassland composite area. If the early warning level is set according to the early warning index, and the regional fire prediction early warning of the target agricultural forest and grass composite area is realized according to different early warning levels.

according to the agricultural, forestry and grass composite district fire prediction and early warning device provided by the embodiment of the invention, the corresponding execution module is arranged, the distribution and change characteristics of the combustion trace patch area are obtained based on the cooperative fusion of the thermal infrared time point and the multispectral data of the multisensor, and the prediction and early warning of the agricultural, forestry and grass composite district fire are realized, so that the uncertainty of the prior art in the agricultural, forestry and grass composite district fire prediction and early warning can be effectively overcome, and the accuracy of data extraction and the early warning effect are effectively improved.

It is understood that, in the embodiment of the present invention, each relevant program module in the apparatus of each of the above embodiments may be implemented by a hardware processor (hardware processor). Moreover, the prediction and early warning device for the fire in the agroforestry and grass composite district according to the embodiment of the present invention can implement the prediction and early warning process for the fire in the agroforestry and grass composite district according to the above-mentioned method embodiments by using the above-mentioned program modules, and when the device is used for implementing the prediction and early warning for the fire in the agroforestry and grass composite district according to the above-mentioned method embodiments, the beneficial effects produced by the device according to the embodiment of the present invention are the same as those of the corresponding method embodiments, and the method embodiments may be referred to, and no further description is given here.

As another aspect of the embodiments of the present invention, the present embodiment provides an electronic device according to the above embodiments, where the electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor executes the computer program to implement the steps of the fire prediction and early warning method for agriculture, forestry, and grasses compound zoning according to the above embodiments.

further, the electronic device of the embodiment of the present invention may further include a communication interface and a bus. Referring to fig. 3, an entity structure diagram of an electronic device provided in an embodiment of the present invention includes: at least one memory 301, at least one processor 302, a communication interface 303, and a bus 304.

The memory 301, the processor 302 and the communication interface 303 complete mutual communication through the bus 304, and the communication interface 303 is used for information transmission between the electronic device and the original data acquisition device; the memory 301 stores a computer program operable on the processor 302, and the processor 302 executes the computer program to implement the steps of the fire prediction and early warning method for the agriculture, forestry and grasses composite zoning area according to the embodiments.

It can be understood that the electronic device at least includes a memory 301, a processor 302, a communication interface 303 and a bus 304, and the memory 301, the processor 302 and the communication interface 303 form a communication connection with each other through the bus 304 and can complete the communication with each other, for example, the processor 302 reads program instructions of the fire prediction and early warning method for the agriculture, forestry and grass composite zoning from the memory 301. In addition, the communication interface 303 may also implement communication connection between the electronic device and the original data acquisition device, and may complete mutual information transmission, such as reading of multispectral data or hyperspectral data, thermal infrared data, and the like, through the communication interface 303.

When the electronic device is running, the processor 302 calls the program instructions in the memory 301 to execute the methods provided by the above-mentioned method embodiments, including for example: on the basis of multi-or high-spectral data and thermal infrared data of different time, space and spectral resolutions of a target agricultural, forest and grass composite area, extracting risk time-space information respectively aiming at the agricultural, forest and grass areas, extracting underlying surface information from a heat sensitive area of the target agricultural, forest and grass composite area, and extracting thermal infrared abnormal points of the target agricultural, forest and grass composite area; based on multi-or hyperspectral data with different medium and high spatial resolutions in the target agroforestry composite area, estimating the fire point generation time and the fire ignition burning area corresponding to the thermal infrared abnormal point by detecting the burning trace in the target agroforestry composite area and determining the time for detecting the burning trace; and based on the fire point generation time and the fire burning area, carrying out fire prediction and early warning on the agricultural, forest and grass region subareas in the target agricultural, forest and grass composite region by combining the historical agricultural, forest and grass region fire occurrence total points, fire occurrence intensity and fire occurrence frequency of the target agricultural, forest and grass composite region which are respectively counted.

the program instructions in the memory 301 may be implemented in the form of software functional units and stored in a computer readable storage medium when sold or used as a stand-alone product. Alternatively, all or part of the steps for implementing the method embodiments may be implemented by hardware related to program instructions, where the program may be stored in a computer-readable storage medium, and when executed, the program performs the steps including the method embodiments; and the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.

Embodiments of the present invention further provide a non-transitory computer-readable storage medium according to the above embodiments, on which computer instructions are stored, and when the computer instructions are executed by a computer, the method for forecasting and warning fire in an agriculture, forestry and grasses composite zoning fire prediction and warning includes the following steps: on the basis of multi-or high-spectral data and thermal infrared data of different time, space and spectral resolutions of a target agricultural, forest and grass composite area, extracting risk time-space information respectively aiming at the agricultural, forest and grass areas, extracting underlying surface information from a heat sensitive area of the target agricultural, forest and grass composite area, and extracting thermal infrared abnormal points of the target agricultural, forest and grass composite area; based on multi-or hyperspectral data with different medium and high spatial resolutions in the target agroforestry composite area, estimating the fire point generation time and the fire ignition burning area corresponding to the thermal infrared abnormal point by detecting the burning trace in the target agroforestry composite area and determining the time for detecting the burning trace; and based on the fire point generation time and the fire burning area, carrying out fire prediction and early warning on the agricultural, forest and grass region subareas in the target agricultural, forest and grass composite region by combining the historical agricultural, forest and grass region fire occurrence total points, fire occurrence intensity and fire occurrence frequency of the target agricultural, forest and grass composite region which are respectively counted.

According to the electronic device and the non-transitory computer readable storage medium provided by the embodiment of the invention, by executing the steps of the prediction and early warning method for the fire disaster in the agriculture and forestry grass composite zoning area, the distribution and change characteristics of the combustion trace patch area are obtained by cooperatively fusing the thermal infrared time point and the multispectral data through the multisensor, so that the prediction and early warning for the fire disaster in the agriculture and forestry grass composite zoning area is realized, the uncertainty of the prior art in the prediction and early warning for the fire disaster in the agriculture and forestry grass composite zoning area can be effectively overcome, and the accuracy of data extraction and the early warning effect are effectively improved.

it is to be understood that the above-described embodiments of the apparatus, the electronic device and the storage medium are merely illustrative, and that elements described as separate components may or may not be physically separate, may be located in one place, or may be distributed on different network elements. Some or all of the modules can be selected according to actual needs to achieve the purpose of the scheme of the embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.

Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. Based on such understanding, the technical solutions mentioned above may be embodied in the form of a software product, which may be stored in a computer-readable storage medium, such as a usb disk, a removable hard disk, a ROM, a RAM, a magnetic or optical disk, etc., and includes several instructions for causing a computer device (such as a personal computer, a server, or a network device, etc.) to execute the methods described in the method embodiments or some parts of the method embodiments.

In addition, it should be understood by those skilled in the art that in the specification of the embodiments of the present invention, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.

in the description of the embodiments of the invention, numerous specific details are set forth. It is understood, however, that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description. Similarly, it should be appreciated that in the foregoing description of exemplary embodiments of the invention, various features of the embodiments of the invention are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of one or more of the various inventive aspects.

However, the disclosed method should not be interpreted as reflecting an intention that: that is, the claimed embodiments of the invention require more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of an embodiment of this invention.

Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the embodiments of the present invention, and not to limit the same; although embodiments of the present invention have been described in detail with reference to the foregoing embodiments, it should be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the spirit and scope of the technical solutions of the embodiments of the present invention.