Forest intelligent management method and system based on satellite remote sensing and ground sensor network

Abstract

The application discloses a forest intelligent management method and system based on satellite remote sensing and ground sensing network, relates to the technical field of forest intelligent management, and comprises the following steps: obtaining thermal anomaly patch information in a satellite thermal infrared image, and extracting a radiation temperature gradient change amplitude of a region corresponding to the thermal anomaly patch; and marking a region with a radiation temperature gradient change amplitude exceeding a preset threshold as a potential hidden heat source region; based on the potential hidden heat source region, scheduling adjacent ground gas sensing nodes to perform high-frequency gas concentration collection, and quantitatively processing according to a gas concentration change amplitude and spatial distribution dispersion. The application realizes early identification of a potential hidden heat source region in a gas non-response stage by fusing satellite thermal infrared and ground gas information, and improves smoldering detection capability by using a gas weak signal; and meanwhile, thermal inertia characteristics are extracted by combining thermal diffusion analysis and thermal disturbance mechanism, so that hierarchical response control is driven to complete fire suppression before an open fire appears.

Description

Technical Field

[0001] This application relates to the field of intelligent forest management technology, specifically to intelligent forest management methods and systems based on satellite remote sensing and ground sensor networks. Background Technology

[0002] Satellite remote sensing refers to the use of Earth observation satellites such as Landsat and Sentinel-2, equipped with optical, multispectral, thermal infrared, or radar sensors, to periodically image and invert parameters of surface targets on a large scale, thereby obtaining macroscopic information such as forest vegetation coverage, leaf area index, water content, and fire anomalies.

[0003] Ground-based sensor networks refer to distributed Internet of Things (IoT) sensing systems deployed within forest areas, consisting of nodes for temperature and humidity, soil moisture, gas concentration, wind speed and direction, etc., which achieve continuous data collection and refined characterization of local environmental conditions through wireless communication.

[0004] When conducting intelligent forest management based on both methods, a large-scale dynamic monitoring base map can be formed through satellite remote sensing, and abnormal areas can be quickly located. Guided by the location results, high-frequency data of the corresponding area of ​​the ground sensor network can be scheduled for fine-grained verification and supplementation, thereby constructing a hierarchical perception mechanism of "macroscopic scanning and local verification". On this basis, time-series data is integrated to perform change detection, disaster early warning and growth assessment, and finally drive decision-making processes such as forest fire prevention and control, pest and disease identification and resource allocation, so as to achieve a closed loop of intelligent forest management that takes into account both wide coverage and precise details.

[0005] In the process of forest area monitoring based on the collaboration of satellite remote sensing and ground sensor networks, when satellite thermal infrared data captures local thermal anomaly patches, but ground gas sensor nodes fail to simultaneously detect combustion characteristic gases such as carbon monoxide and carbon dioxide, the forest fire assessment mechanism, during the multi-source data consistency verification process, will identify the high-temperature patch as an abnormal fluctuation caused by environmental disturbance and ignore it, thus failing to trigger further refined detection or manual verification of the area. If this assessment path persists, the understory layer of dead branches and leaves may have already entered a low-intensity smoldering state. This type of smoldering process releases heat slowly and gas escape is hindered, making it difficult for ground gas sensor nodes to capture typical combustion product signals in a timely manner, thus creating a perception decoupling phenomenon of "high temperature is evident but gas does not respond." As the smoldering area gradually expands to the surrounding areas under continuous combustible material distribution, heat accumulates continuously in the local space, causing the temperature field below the surface to rise continuously and further altering the pyrolysis state of the surrounding combustibles, causing more areas to gradually approach the critical combustion state. Once the accumulated heat exceeds the threshold, the previously concealed smoldering area will rapidly transform into open flames, forming a large-scale, sudden fire. Due to the lack of effective early warning and intervention, such fires already possess high intensity and spread capability when they are exposed, causing not only large-scale damage to forest vegetation but also significantly increasing the difficulty of firefighting and posing ecological security risks.

[0006] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0007] The purpose of this application is to provide a method and system for intelligent forest management based on satellite remote sensing and ground sensor networks, so as to solve the problems mentioned in the background art.

[0008] To achieve the above objectives, this application provides the following technical solution: a smart forest management method based on satellite remote sensing and ground sensor networks, comprising the following steps: Acquire information on thermal anomaly patches in satellite thermal infrared images, extract the amplitude of radiation temperature gradient change in the corresponding region of thermal anomaly patches, and mark the region where the amplitude of radiation temperature gradient change exceeds a preset threshold as a potential hidden heat source area. Based on potential hidden heat source areas, nearby ground gas sensing nodes are scheduled to perform high-frequency gas concentration acquisition. The gas concentration change amplitude and spatial distribution dispersion are quantified to obtain sparse characteristic values ​​of gas response. Based on the sparse characteristic values ​​of the gas response, the local heat diffusion trajectory of the potential hidden heat source area is calculated, the heat outward expansion direction offset angle and expansion rate parameters are extracted, and the heat accumulation intensity index is generated. Based on the heat accumulation intensity index, point-by-point thermal disturbance excitation is implemented in potential concealed heat source areas. By introducing small-amplitude environmental disturbances, the temperature rebound amplitude is obtained, the thermal inertia enhancement characteristics are extracted, and the concealed combustion criterion is updated. Based on the updated criteria for concealed combustion, graded response control is implemented for potential concealed heat source areas. Differentiated intervention strategies are triggered according to the magnitude of thermal inertia enhancement characteristics to suppress concealed fires before they appear as open flames.

[0009] Preferably, areas where the amplitude of the radiation temperature gradient change exceeds a preset threshold are designated as potential hidden heat source areas. The specific steps are as follows: Acquire satellite thermal infrared image data within the forest coverage area, establish a unified spatial grid according to the image pixel correspondence, arrange the radiation temperature values ​​of the same spatial location at different times, filter spatial locations with increasing temperature at three consecutive times and expand to eight adjacent areas, delineate temperature rising areas and identify thermal anomaly patches. The area covered by the thermal anomaly patch was divided into sub-regions and the radiation temperature values ​​were recorded. The temperature difference in each direction and the change of continuous time interval were calculated in adjacent sub-regions. The spatial difference changes and the temporal changes were summarized and superimposed to obtain the radiation temperature gradient change amplitude of the corresponding sub-region. By comparing the amplitude of the radiation temperature gradient change in each sub-region with a preset threshold, candidate heat source units that exceed the preset threshold for three consecutive time moments are marked. The spatial adjacency relationship is combined for merging and the spatial range is continuously updated, thereby identifying the region where the amplitude of the radiation temperature gradient change exceeds the preset threshold as a potential hidden heat source region.

[0010] Preferably, the sparse characteristic values ​​of the gas response are obtained through the following steps: The boundary of the potential hidden heat source area is selected as the reference starting point. The spatial position of the ground gas sensing node is located one by one and the straight distance between it and the nearest point of the potential hidden heat source area boundary is calculated. Ground gas sensing nodes with a distance less than a fixed distance range are included in the scheduling range and a unified time starting point is assigned to perform gas concentration collection. Based on the gas concentration values ​​obtained by ground gas sensor nodes within the scheduling range, the gas concentration changes are recorded in chronological order and the gas concentration change amplitude is calculated. The difference in gas concentration change amplitude between adjacent ground gas sensor nodes is arranged in spatial order to obtain the gas concentration change amplitude sequence and spatial distribution data sequence. The effective change segments are selected based on the gas concentration change amplitude sequence, and the proportion of ground gas sensing nodes and the number of continuous segments are counted. The numerical level is output as the gas response sparsity feature value according to the matching relationship between the node proportion value and the number of continuous segments in the corresponding table.

[0011] Preferably, in the process of screening effective change segments from the gas concentration change amplitude sequence, the time period corresponding to the change amplitude that is greater than zero in a continuous time interval is marked as an effective change segment, and the number of effective change segments corresponding to each ground gas sensing node is counted. Combined with the spatial distribution data sequence, the distribution of effective change segments of adjacent nodes is checked in spatial order to divide continuous segments.

[0012] Preferably, the specific steps for generating a heat accumulation intensity index are as follows: The geometric center location is determined based on the potential hidden heat source region corresponding to the sparse eigenvalues ​​of the gas response, and the observation direction is divided. The radiation temperature values ​​of continuous sub-regions are read along each observation direction to construct a temperature change record that includes spatial order and temporal variation. By combining temperature change records to filter observation directions and construct heat diffusion paths, connecting the temperature change locations of continuous sub-regions, extracting the path deflection of each observation direction and the change in the number of sub-regions per unit time, we can obtain the heat outward expansion direction offset angle and expansion rate parameters. The primary and secondary diffusion directions are divided by the offset angle and expansion rate parameters of the heat diffusion direction. The diffusion state is statistically analyzed and matched with the corresponding relationship table. The corresponding numerical level is output to obtain the heat accumulation intensity index.

[0013] Preferably, the primary and secondary diffusion directions are divided using the offset angle of the heat diffusion direction and the expansion rate parameter, including: taking the angle between each observation direction path and a pre-selected reference direction as the offset angle; marking the primary diffusion direction when the offset angle is less than or equal to 45 degrees and the expansion rate parameter is increasing; and marking the secondary diffusion direction when the offset angle is greater than 45 degrees or the expansion rate parameter is decreasing or remains unchanged.

[0014] Preferably, the thermal inertia enhancement features are extracted and the concealed combustion criterion is updated. The specific steps are as follows: Using each sub-region within a potential hidden heat source area as a processing unit, the radiation temperature values ​​are recorded and short-term airflow changes are introduced to obtain a temperature recording sequence. Read the temperature record sequence and select the radiation temperature value at the end of the disturbance as the rebound start reference value. Calculate the temperature difference and extract the temperature rebound amplitude. Arrange the temperature rebound amplitude and record it in correspondence with the heat accumulation intensity index. The temperature rebound amplitude is divided and thermal inertia enhancement units are screened. Continuous regions are merged and spatial coverage is statistically analyzed to update the concealed combustion criteria.

[0015] Preferably, the process of reading the temperature record sequence and selecting the radiation temperature value at the end of the disturbance as the rebound start reference value further includes: comparing the radiation temperature values ​​at the end of the disturbance and at continuous time points after the end of the disturbance, selecting the maximum temperature difference as the temperature rebound amplitude, and recording the temperature rebound amplitude in correspondence with the heat accumulation intensity index.

[0016] Preferably, a differentiated intervention strategy is triggered based on the magnitude of the enhanced thermal inertia feature. The specific steps are as follows: Obtain the concealed combustion criterion level and thermal inertia enhancement characteristic value, classify the state into high-level state, intermediate-level state and low-level state, and incorporate them into the first response sequence, second response sequence and third response sequence respectively, and sort the response execution units according to the distance of the center location of the potential concealed heat source area. Based on the sorting results of the response execution units, the values ​​of thermal inertia enhancement characteristics are statistically analyzed and the average value is calculated. High-intensity intervention intervals and low-intensity intervention intervals are divided. Continuous airflow disturbance operation or single airflow disturbance operation is performed for the first response sequence, interval airflow disturbance operation is performed for the second response sequence, and the state record is retained for the third response sequence. Based on the temperature change trend over continuous time, the first response sequence, the second response sequence, and the third response sequence are adjusted, and differentiated intervention strategies are used to achieve differentiated response control of potential hidden heat sources.

[0017] The intelligent management system for forest areas based on satellite remote sensing and ground sensor networks includes a thermal infrared anomaly identification module, a gas sparse feature extraction module, a thermal diffusion trajectory analysis module, a thermal inertial feature extraction module, and a hierarchical response control module. The thermal infrared anomaly identification module acquires information on thermal anomaly patches in satellite thermal infrared images and extracts the amplitude of radiation temperature gradient changes in the corresponding areas of thermal anomaly patches. Areas with radiation temperature gradient changes exceeding a preset threshold are identified as potential hidden heat source areas. The gas sparse feature extraction module, based on the potential hidden heat source area, schedules nearby ground gas sensing nodes to perform high-frequency gas concentration acquisition, and performs quantification based on the gas concentration change amplitude and spatial distribution dispersion to obtain gas response sparse feature values. The heat diffusion trajectory analysis module calculates the local heat diffusion trajectory of potential hidden heat source areas based on the sparse characteristic values ​​of the gas response, extracts the offset angle and expansion rate parameters of the heat expansion direction, and generates a heat accumulation intensity index. The thermal inertia feature extraction module implements point-by-point thermal disturbance excitation in potential concealed heat source areas based on the heat accumulation intensity index. By introducing small-amplitude environmental disturbances, it obtains the temperature rebound amplitude, extracts thermal inertia enhancement features, and updates the concealed combustion criteria. The graded response control module performs graded response control for potential concealed heat source areas according to the updated concealed combustion criteria, and triggers differentiated intervention strategies based on the magnitude of the thermal inertia enhancement characteristics.

[0018] The technical effects and advantages provided by this application in the above technical solution are as follows: This application constructs a collaborative analysis mechanism based on satellite thermal infrared imagery and ground-based gas sensing information. After identifying thermal anomaly patches, it further introduces the amplitude of radiation temperature gradient changes for constraint judgment, enabling potential concealed heat source areas to be located in advance before gas signals respond, thus avoiding the judgment lag problem caused by relying solely on gas concentration changes. By jointly characterizing the amplitude of gas concentration changes and spatial distribution dispersion, even weak gas response signals can be effectively utilized, thereby improving the ability to identify concealed smoldering states and enabling fires to be detected at an early stage and enter subsequent processing procedures.

[0019] This application, based on the identification of potential concealed heat source areas, introduces heat diffusion trajectory analysis and a point-by-point thermal disturbance excitation mechanism. By acquiring temperature rebound amplitude, it extracts thermal inertia enhancement characteristics and updates the concealed combustion criteria accordingly, thereby driving a graded response control process. This allows different areas to implement corresponding intervention strategies based on their thermal state differences. By prioritizing continuous intervention in high-risk areas and combining it with dynamic response adjustments, the heat release process is continuously regulated, thereby achieving proactive suppression before the fire develops into an open flame stage, reducing the risk of subsequent expansion, and improving the overall controllability of forest area monitoring and management. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings.

[0021] Figure 1 This is a flowchart illustrating the overall method of this application.

[0022] Figure 2 This is a schematic diagram of the system modules of this application. Detailed Implementation

[0023] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0024] like Figure 1As shown, this application provides a method for intelligent forest management based on satellite remote sensing and ground sensor networks, including the following steps: Step 1: Obtain information on thermal anomaly patches in satellite thermal infrared images, extract the amplitude of radiation temperature gradient change in the corresponding area of ​​thermal anomaly patches, and mark the area where the amplitude of radiation temperature gradient change exceeds a preset threshold as a potential hidden heat source area. Areas where the amplitude of the radiation temperature gradient change exceeds a preset threshold are identified as potential hidden heat source areas. The specific steps are as follows: Within the forest coverage area, satellite thermal infrared image data of the corresponding area are acquired at fixed time intervals. Each acquired thermal infrared image contains radiation temperature values ​​corresponding to multiple spatial locations. After acquiring thermal infrared images at multiple time points, a unified spatial grid is established based on the pixel correspondence of each spatial location in the image as the basic unit. The radiation temperature values ​​of the same spatial location at different time points are arranged in chronological order to generate a point-by-point temperature change record.

[0025] Subsequently, the radiation temperature value at the current time is compared point by point with the radiation temperature values ​​at the previous two time points. When a certain spatial location shows an increasing temperature over three consecutive time points, the spatial location is marked as the point of continuous temperature increase. Then, with the point of continuous temperature increase as the center, adjacent spatial locations within a fixed range are selected around it. The fixed range is selected as an eight-adjacent region based on the center point. The radiation temperature values ​​of all spatial locations within this range are compared synchronously. When more than half of the spatial locations within this range simultaneously meet the condition of increasing temperature over three consecutive time points, the entire range is defined as the temperature rising region.

[0026] Based on this, the temperature rise area is compared with its surrounding adjacent areas. When the radiation temperature value of all spatial locations within the temperature rise area is higher than the radiation temperature value of the corresponding location in the surrounding area, the continuously distributed temperature rise area is identified as a thermal anomaly patch.

[0027] "Thermal anomaly" refers to the sustained temperature rise and relatively high temperature state of a continuous spatial range in a satellite thermal infrared image relative to the time dimension and the surrounding adjacent spatial range. Specifically, taking each spatial location in the image as the basic unit, the radiation temperature values ​​of the same spatial location at different times are arranged in chronological order. When a spatial location shows an increasing radiation temperature for three consecutive times, the spatial location is marked as the point of continuous temperature rise, and eight adjacent regions are determined with the point of continuous temperature rise as the center.

[0028] When more than half of the spatial locations within an eight-adjacent region simultaneously meet the condition of increasing radiation temperature for three consecutive time periods, the eight-adjacent region is designated as a temperature-rising region. Furthermore, the radiation temperature of the temperature-rising region is compared with that of its surrounding adjacent regions. When the radiation temperature values ​​of all spatial locations within the temperature-rising region are higher than the corresponding radiation temperature values ​​of the surrounding regions, the temperature-rising region is deemed to meet the criteria for thermal anomaly determination, and continuously distributed temperature-rising regions that meet the criteria for thermal anomaly determination are identified as thermal anomaly patches.

[0029] "Thermal anomalies" are not determined based on subjective judgment or undefined absolute high temperatures. Instead, they are determined by the increasing relationship of radiation temperature over three consecutive time points as the time dimension criterion, the fact that more than half of the spatial locations within eight adjacent areas satisfy the increasing relationship of radiation temperature as the local spatial continuity criterion, and the fact that the radiation temperature of all spatial locations within the temperature-rising area is higher than the radiation temperature of the corresponding locations in the surrounding area as the spatial relative temperature criterion. Areas that simultaneously meet the above time dimension, local spatial continuity, and spatial relative temperature criterion are considered to be "thermal anomalies". Areas that do not meet the above conditions are not identified as thermal anomaly patches.

[0030] After identifying the thermal anomaly patch, the area covered by the thermal anomaly patch is divided into sections of a fixed size. Each pixel is numbered as an independent sub-region, and the radiation temperature value of each sub-region at each time point is recorded as the basic data for subsequent processing.

[0031] After completing the sub-region division and temperature recording within the thermal anomaly patch, each sub-region is used as a processing unit. At the same time, adjacent sub-regions in the east-west, north-south, and two diagonal directions of that sub-region are selected. The radiation temperature values ​​of each pair of adjacent sub-regions are compared one by one, and the temperature difference in each direction is recorded and saved according to the direction. Subsequently, between two adjacent time points, the temperature difference changes of the same pair of sub-regions are compared again. The temperature difference of the later time point is subtracted from the temperature difference of the earlier time point, and the change process is recorded. The changes in three consecutive time intervals are accumulated and recorded in chronological order.

[0032] After recording the temperature difference changes in all directions and at all time intervals, the temperature difference changes in each direction of the same sub-region are summarized item by item, and the changes in the time dimension are superimposed one by one to obtain a numerical result that can simultaneously reflect the spatial difference changes and the time change process. This numerical result is used as the amplitude of the radiation temperature gradient change of the sub-region.

[0033] For sub-regions located at the edge of thermal anomaly patches, when performing the above processing, the temperature difference between the sub-region and the adjacent spatial location outside the thermal anomaly patch is also introduced, so that the amplitude of the radiation temperature gradient change of the sub-region simultaneously includes the temperature change information of both the inside and the outside, thereby ensuring that the amplitude of the radiation temperature gradient change of each sub-region fully reflects the process of heat spreading from the inside to the outside.

[0034] After obtaining the amplitude of radiation temperature gradient changes in all sub-regions within the thermal anomaly patch, a reference range is established in advance based on long-term historical forest area observation data. The amplitude of radiation temperature gradient changes at each spatial location in the historical data under normal environmental conditions is statistically recorded, and continuous variation intervals are sorted out in chronological order. The upper limit of this continuous variation interval is used as the preset threshold.

[0035] In the actual calibration process, for each sub-region, the amplitude of the obtained radiation temperature gradient change is compared with the preset threshold one by one. When the amplitude of the radiation temperature gradient change of a certain sub-region exceeds the preset threshold for three consecutive time moments, the sub-region is marked as a candidate heat source unit. After all sub-regions are marked as candidates, based on spatial adjacency, candidate heat source units with spatial boundary contact relationships are merged. Spatial adjacency is defined as two sub-regions having a common boundary in the horizontal or vertical direction. When multiple candidate heat source units meet the adjacency relationship, they are merged into the same continuous region, and the continuous region is marked as a potential hidden heat source area.

[0036] After calibration, the changes of the potential hidden heat source area in subsequent time periods are continuously updated. When a sub-region no longer meets the preset threshold condition within the next three time periods, it is removed from the potential hidden heat source area. When a new sub-region meets the preset threshold condition within three consecutive time periods, it is included in the potential hidden heat source area. This ensures that the spatial range of the potential hidden heat source area is always consistent with the changes in the radiation temperature gradient and provides a continuously updated spatial basis for subsequent processing.

[0037] Step 2: Based on the potential hidden heat source area, the nearby ground gas sensing nodes are scheduled to perform high-frequency gas concentration acquisition. The gas concentration change amplitude and spatial distribution dispersion are quantified to obtain the gas response sparse characteristic value. The specific steps for obtaining sparse eigenvalues ​​of the gas response are as follows: After determining the spatial range of the potential hidden heat source area, the boundary of the potential hidden heat source area is used as a reference starting point to locate the ground gas sensing nodes distributed around the area one by one, record the spatial position of each ground gas sensing node, and calculate the straight-line distance between it and the nearest point on the boundary of the potential hidden heat source area.

[0038] Ground gas sensing nodes within a fixed distance range are included in the scheduling range. The fixed distance range is set to cover multiple consecutive spatial locations outside the boundary of the potential concealed heat source area, and it is ensured that the ground gas sensing nodes within this range are spatially distributed continuously around the potential concealed heat source area.

[0039] After the scheduling range is determined, a unified time starting point is assigned to all ground gas sensing nodes included in the scheduling range, and gas concentration collection operations are performed at fixed time intervals. The fixed time interval is an integer division of the original collection interval, so that multiple gas concentration values ​​can be obtained within each original collection cycle.

[0040] During the data collection process, the gas concentration values ​​acquired by each ground gas sensing node at each time point are recorded one by one in chronological order. At the same time, the gas concentration values ​​of all ground gas sensing nodes at the same time point are arranged accordingly, thereby constructing a high-frequency gas concentration data set that covers multiple spatial locations around potential hidden heat source areas and has a continuous temporal order.

[0041] After obtaining the high-frequency gas concentration data set, each ground gas sensing node is used as an independent processing unit. The gas concentration values ​​at multiple consecutive time points are read sequentially according to the time sequence. The gas concentration value at the current time point is compared with the gas concentration value at the previous time point, and the difference between the two is recorded. This difference is then compared with the difference at the time point before that, so as to obtain the changes within two consecutive time intervals.

[0042] Based on this, the two differences formed between three consecutive time points are summed to obtain the gas concentration change amplitude within that time period. This change amplitude is then recorded segment by segment in chronological order to form a gas concentration change amplitude sequence for each ground gas sensing node. After obtaining the change amplitude sequences of all ground gas sensing nodes, the ground gas sensing nodes are sorted according to their spatial location. Using the center of the potential hidden heat source area as a reference point, the ground gas sensing nodes are arranged in order of increasing distance.

[0043] Subsequently, the amplitude of gas concentration changes between two adjacent ground gas sensing nodes is compared one by one, the difference between the amplitude of changes between adjacent nodes is recorded, and the data is arranged in spatial order to form a data sequence that reflects the spatial distribution of the amplitude of gas concentration changes. This data sequence is used to describe the continuous change state of gas concentration changes within a spatial range.

[0044] After obtaining the gas concentration change amplitude sequence and its spatial distribution data sequence, focusing on the generation process of sparse characteristic values ​​of gas response, the gas concentration change amplitude sequence of each ground gas sensing node is first screened. The time period corresponding to the change amplitude that is greater than zero for three consecutive time points is marked as the effective change segment, and the number of effective change segments corresponding to each ground gas sensing node is counted.

[0045] After completing the statistical analysis of effective change segments for all ground gas sensing nodes, the number of ground gas sensing nodes with effective change segments is compared with the total number of ground gas sensing nodes to obtain the proportion of nodes involved in gas concentration changes. Subsequently, based on the spatial distribution data sequence, the spatial distribution locations of ground gas sensing nodes with effective change segments are marked one by one, and adjacent nodes are checked in spatial order to see if they simultaneously possess effective change segments. When adjacent nodes simultaneously possess effective change segments, they are grouped into the same continuous segment; when adjacent nodes do not simultaneously possess effective change segments, they are divided into different segments, and the number of all continuous segments is counted.

[0046] After completing the node ratio statistics and continuous segment statistics, the node ratio and the number of continuous segments are recorded accordingly. When the node ratio value is small and the number of continuous segments is greater than one, the state is marked as a discrete gas distribution state. When the node ratio value is large and the number of continuous segments is equal to one, the state is marked as a concentrated gas distribution state.

[0047] After completing the node ratio and continuous segment statistics, the node ratio and the number of continuous segments are recorded accordingly. The node ratio is defined as the ratio between the number of ground gas sensing nodes with effective change segments and the total number of ground gas sensing nodes included in the scheduling range. In the specific judgment process, the node ratio is divided into two intervals. When the node ratio is less than or equal to one-half, it is considered that the node ratio value is small. When the node ratio is greater than one-half, it is considered that the node ratio value is large. Based on this, a joint judgment is made in combination with the number of continuous segments. When the node ratio is less than or equal to one-half and the number of continuous segments is greater than one, this state is marked as a discrete gas distribution state. At this time, it means that the number of ground gas sensing nodes participating in gas concentration changes is small and they are spatially distributed in multiple discontinuous areas.

[0048] When the node ratio is greater than one-half and the number of continuous segments is equal to one, the state is marked as a concentrated gas distribution state. This indicates that the number of ground gas sensing nodes participating in gas concentration changes accounts for a large proportion and forms a single continuous distribution area in space.

[0049] For cases where the node ratio is less than or equal to one-half and the number of consecutive segments is equal to one, or where the node ratio is greater than one-half and the number of consecutive segments is greater than one, the case is classified into a transitional state and further subdivided according to a pre-set classification rule to ensure that the gas distribution state classification process has complete boundary definitions, thereby avoiding cases that cannot be classified.

[0050] Finally, based on the combination relationship between the node ratio values ​​and the number of consecutive segments, the corresponding numerical level is output by matching each item according to the pre-set correspondence table. This numerical level is then used as the gas response sparsity feature value for further processing of potential hidden heat source areas.

[0051] Step 3: Based on the sparse characteristic values ​​of the gas response, calculate the local heat diffusion trajectory of the potential hidden heat source area, extract the heat outward expansion direction offset angle and expansion rate parameters, and generate a heat accumulation intensity index. The specific steps for generating a heat accumulation intensity index are as follows: After obtaining the sparse characteristic values ​​of the gas response, the spatial range of the potential hidden heat source area is used as the basic processing area, and the geometric center of the potential hidden heat source area is used as the reference starting point. Multiple observation directions are divided within the potential hidden heat source area according to fixed directions. The observation directions include four groups of directions: east-west, north-south, and two diagonal directions. Each group of directions extends outward from the reference starting point to the boundary of the potential hidden heat source area.

[0052] In each direction, following the same spatial grid division method as in step one, consecutive sub-regions distributed along that direction are selected sequentially, and the radiation temperature values ​​corresponding to each sub-region at multiple consecutive time points are read. For each consecutive sub-region in each direction, the temperature changes between adjacent sub-regions are recorded point by point in order from the reference starting point outwards. The temperature of the sub-region closer to the reference starting point is compared with the temperature of its adjacent sub-regions outside it, and this process continues to expand outwards step by step to form a temperature change sequence in that direction.

[0053] Subsequently, the temperature change sequence in the same direction is compared moment by moment over multiple consecutive time intervals, and the changes in the temperature change sequence in the time dimension are recorded, thereby obtaining a temperature change record in each direction that simultaneously contains both spatial and temporal change sequences.

[0054] After recording temperature changes in all directions, the gas response sparsity characteristic value is used as the criterion. When the gas response sparsity characteristic value corresponds to a discrete state of gas distribution, the direction with the larger change amplitude in the temperature change sequence is selected as the dominant direction. When the gas response sparsity characteristic value corresponds to a concentrated state of gas distribution, the temperature change sequences in all directions are included in the processing range, thereby providing a basis for direction selection for subsequent calculation of local thermal diffusion trajectory.

[0055] After completing the temperature change recording and direction selection for each observation direction, the temperature change positions of each sub-region are connected sequentially in each selected direction, starting from the reference starting point, according to spatial order, to construct a heat diffusion path along that direction. The temperature changes of each sub-region along the path are recorded segment by segment at continuous time.

[0056] For three consecutive sub-regions in the same direction, the sub-region closest to the reference starting point is used as the starting point. The middle sub-region and the outer sub-region are connected in sequence to form a spatial path, and the temperature change position of the path at different times is recorded. Then, by comparing the spatial position changes of adjacent path segments in the same direction at different times, the expansion trend of heat in this direction is determined, and the number of sub-regions through which heat expands outward in each time interval is recorded, thus obtaining the expansion path sequence in this direction.

[0057] After obtaining the expansion path sequences in each direction, the expansion paths in different directions are compared. Taking the reference starting point as the baseline, the deflection of each path relative to the reference direction is recorded. The angle between each path and the east-west direction is used as the reference angle, and the offset of other paths relative to this reference angle is recorded one by one, thus obtaining the offset angle of the heat expansion direction. At the same time, in each direction, based on the change in the number of sub-regions traversed by the heat expansion within a continuous time period, the number of sub-regions whose path length increases per unit time is recorded. This number is used as the expansion rate parameter in that direction, so that the expansion rate parameter can reflect the speed change of the heat expansion outward along that direction.

[0058] After obtaining the parameters of the heat expansion direction offset angle and expansion rate, the specific operation involves the generation process of the heat accumulation intensity index: First, the spread rate parameters of all observation directions within the potential hidden heat source area are compiled and arranged according to their numerical values, and the range of difference between the maximum and minimum values ​​is recorded. On this basis, the offset angle of the heat spread direction is included in the analysis range. For directions with small offset angles and continuously increasing spread rate parameters, they are marked as primary diffusion directions. For directions with large offset angles or discontinuous changes in spread rate parameters, they are marked as secondary diffusion directions.

[0059] When analyzing the offset angle of the heat expansion direction, a pre-selected reference direction is used as the reference direction, and the angle between each directional path and this reference direction is taken as the offset angle; in the specific determination process: The offset angle is divided into two intervals: when the offset angle is less than or equal to 45 degrees, it is considered a small offset angle; when the offset angle is greater than 45 degrees, it is considered a large offset angle. Based on this, combined with the changes in the spreading rate parameter, when the spreading rate parameter in a certain direction increases over three consecutive time points, it is considered a continuous increase in the spreading rate parameter; when the spreading rate parameter decreases or remains unchanged between adjacent time points, it is considered a discontinuous change in the spreading rate parameter. This provides a clear basis for determining the primary and secondary spreading directions.

[0060] Subsequently, the number of main diffusion directions was counted, and the changes in their corresponding spread rate parameters were recorded. When the number of main diffusion directions was a single direction and the spread rate parameter continued to increase over a continuous period of time, the state was recorded as a concentrated diffusion state. When the number of main diffusion directions was multiple directions and the spread rate parameter distributions of each direction differed, the state was recorded as a dispersed diffusion state.

[0061] After classifying the diffusion states, the numerical range of the expansion rate parameter is combined with the diffusion state and recorded. The combinations are matched item by item according to a pre-set correspondence table, and different combinations are mapped to different levels of numerical results. These numerical results are used as an index of heat accumulation intensity. A higher level is corresponding to a larger expansion rate parameter and a concentrated diffusion state, while a lower level is corresponding to a smaller expansion rate parameter and a dispersed diffusion state. Other combinations are classified as intermediate levels. This allows the heat accumulation intensity index to simultaneously reflect the heat diffusion rate and the degree of concentration of the diffusion direction, providing a basis for subsequent treatment of potential hidden heat source areas.

[0062] Step 4: Based on the heat accumulation intensity index, implement point-by-point thermal disturbance excitation in the potential concealed heat source area, obtain the temperature rebound amplitude by introducing small-amplitude environmental disturbances, extract the thermal inertia enhancement characteristics, and update the concealed combustion criterion. The specific steps for extracting thermal inertia enhancement features and updating the concealed combustion criterion are as follows: Given that the potential concealed heat source area and the corresponding heat accumulation intensity index have been determined, each sub-region within the potential concealed heat source area is used as a processing unit. Sub-regions are selected one by one according to the spatial grid order, and the radiation temperature value of each sub-region at the current time is recorded as the initial temperature value. On this basis, a fixed form of environmental disturbance is introduced into the spatial range corresponding to each sub-region using a unified time control method. The environmental disturbance is limited to applying a short-term airflow change with a fixed time interval to the spatial range corresponding to the sub-region. The airflow change is generated at the beginning of the disturbance and stops after the fixed time interval ends. The direction of airflow is kept consistent during the disturbance.

[0063] After introducing the airflow change, the radiation temperature values ​​of the sub-region are recorded hourly at the time before the disturbance begins, at the time of the disturbance beginning, at the time of the disturbance ending, and at three consecutive time points after the disturbance ends, forming a temperature record sequence containing at least five consecutive time points, thus providing a complete data foundation for the subsequent extraction of temperature rebound amplitude.

[0064] After the temperature record sequence of each sub-region is formed, the corresponding temperature record sequence is read in chronological order for each sub-region. The radiation temperature value at the end of the disturbance is used as the rebound start reference value, and the radiation temperature values ​​at three consecutive time points after the end of the disturbance are read in sequence.

[0065] During this process, the radiation temperature value at each time point after the disturbance ends is compared with the rebound initiation reference value one by one, the temperature difference at each time point is calculated, and the maximum temperature difference is selected as the temperature rebound amplitude of the sub-region. After the temperature rebound amplitude of all sub-regions is extracted, the temperature rebound amplitude of each sub-region is arranged according to its spatial location and recorded one by one with the corresponding heat accumulation intensity index.

[0066] Furthermore, the temperature rebound amplitude of all sub-regions is sorted according to its numerical value, and the number of sub-regions with a temperature rebound amplitude greater than zero is counted. This number is then compared with the total number of sub-regions to obtain the temperature rebound response ratio, thus providing a unified basis for the extraction of thermal inertia enhancement features.

[0067] After determining the temperature rebound amplitude, temperature rebound response ratio, and corresponding heat accumulation intensity index, the specific operation involves extracting the enhanced thermal inertia features and updating the concealed combustion criteria. First, all sub-regions are divided into two categories based on their temperature rebound amplitudes. A sub-region with a temperature rebound amplitude greater than the average of all sub-regions is marked as a high-rebound sub-region, while a sub-region with a temperature rebound amplitude less than or equal to the average is marked as a low-rebound sub-region. High-rebound sub-regions are then jointly screened with their corresponding heat accumulation intensity indices. A sub-region is marked as a thermal inertia enhancement unit when it simultaneously satisfies both a temperature rebound amplitude greater than the average and a corresponding heat accumulation intensity index within the top half of the ranking results for all sub-region heat accumulation intensity indices.

[0068] Subsequently, the adjacency relationships between all thermal inertia enhancement units are examined within the spatial range. Thermal inertia enhancement units with common boundaries in the horizontal or vertical directions are grouped into continuous regions, and the number of continuous regions and their spatial coverage are counted.

[0069] When the number of continuous regions is equal to one and the continuous region covers the center of the potential concealed heat source area, the concealed combustion criterion is updated to a high-level state. When the number of continuous regions is greater than one, the concealed combustion criterion is updated to an intermediate-level state. When there are no continuous regions, the concealed combustion criterion is updated to a low-level state, thus completing the concealed combustion criterion update process based on the enhanced thermal inertia feature.

[0070] Step 5: According to the updated concealed combustion criteria, implement graded response control for potential concealed heat source areas, and trigger differentiated intervention strategies based on the magnitude of thermal inertia enhancement characteristics to suppress concealed fires before the appearance of open flames. The differentiated intervention strategy is triggered based on the magnitude of the thermal inertia enhancement feature. The specific steps are as follows: With the updated concealed combustion criteria determined, the spatial range of the potential concealed heat source area is taken as the control object. According to the sub-region division method in the aforementioned steps, each sub-region within the potential concealed heat source area is used as a response execution unit, and the concealed combustion criterion level and thermal inertia enhancement characteristic value corresponding to each sub-region are read one by one. On this basis, the concealed combustion criteria are divided into three level intervals: high-level state, intermediate-level state, and low-level state, and each sub-region within the potential concealed heat source area is classified and organized according to the level order.

[0071] Sub-regions in a high-level state are included in the first response sequence, sub-regions in an intermediate-level state are included in the second response sequence, and sub-regions in a low-level state are included in the third response sequence. Subsequently, using the center of the potential hidden heat source area as a reference point, the sub-regions in each response sequence are sorted in order of spatial distance from near to far, so that the sub-regions closer to the center of the potential hidden heat source area are given priority in response control, thus forming a hierarchical response control basis with spatial priority.

[0072] After the response sequence division and spatial sorting of each sub-region are completed, the intervention intensity interval is further divided within the corresponding response sequence based on the thermal inertia enhancement feature value corresponding to each sub-region. The thermal inertia enhancement feature values ​​of all sub-regions are statistically analyzed and the average value is calculated. When the thermal inertia enhancement feature value of a certain sub-region is greater than the average value, it is classified into the high-intensity intervention interval. When the thermal inertia enhancement feature value is less than or equal to the average value, it is classified into the low-intensity intervention interval.

[0073] For different combinations of response sequences and intervention intensity ranges, corresponding differentiated intervention strategies are determined one by one: For the sub-region of the first response sequence that belongs to the high-intensity intervention range, airflow disturbance operations are continuously performed at fixed time intervals within its corresponding spatial range. The fixed time interval is the time length between two adjacent temperature recording time intervals, and airflow disturbance operations are performed in three consecutive time intervals; For the sub-region of the first response sequence that belongs to the low-intensity intervention range, airflow disturbance operations are performed only once within a single time interval within its corresponding spatial range.

[0074] For the sub-region of the second response sequence, airflow disturbance is performed in an interval manner within its corresponding spatial range. That is, airflow disturbance is performed only in the first time interval of three consecutive time intervals, and no disturbance is performed in the other two time intervals.

[0075] For the sub-regions of the third response sequence, no airflow disturbance operation is performed; only the status record is retained, thus forming a differentiated intervention strategy system with different intervention frequencies and durations corresponding to different levels.

[0076] During the implementation of differentiated intervention strategies in each sub-region, the temperature changes in each sub-region within the potential hidden heat source area are continuously recorded, and three consecutive time points are used as judgment windows. Within each judgment window, the temperature change trend of each sub-region is analyzed one by one: When the temperature value of a sub-region in the first response sequence decreases successively over three consecutive time intervals, the sub-region is moved to the second response sequence. When the temperature value of a sub-region in the second response sequence remains unchanged over three consecutive time intervals (i.e., the temperature value at the later time interval is less than or equal to the temperature value at the previous time interval), the sub-region is moved to the third response sequence. When a sub-region in the third response sequence again meets the criteria for the thermal inertia enhancement unit over three consecutive subsequent time intervals, it is reintegrated into the second response sequence.

[0077] By using the unified judgment rule based on three consecutive time points, each sub-region within the potential hidden heat source area always corresponds to a response sequence that matches its temperature change state.

[0078] Throughout the graded response control process, high-frequency intervention is prioritized for the first response sequence sub-region, and the response level is gradually reduced after the temperature change trend changes. This allows the heat inside the potential concealed heat source area to be continuously released and gradually weakened, thereby suppressing the concealed fire before the open flame combustion occurs. This achieves a closed-loop graded response control process based on the updated concealed combustion criterion and the enhanced thermal inertia characteristics.

[0079] like Figure 2As shown, this application provides a forest area intelligent management system based on satellite remote sensing and ground sensor network, including a thermal infrared anomaly identification module, a gas sparse feature extraction module, a thermal diffusion trajectory analysis module, a thermal inertial feature extraction module, and a hierarchical response control module; The thermal infrared anomaly identification module acquires information on thermal anomaly patches in satellite thermal infrared images and extracts the amplitude of radiation temperature gradient changes in the corresponding areas of thermal anomaly patches. Areas with radiation temperature gradient changes exceeding a preset threshold are identified as potential hidden heat source areas. The gas sparse feature extraction module, based on the potential hidden heat source area, schedules nearby ground gas sensing nodes to perform high-frequency gas concentration acquisition, and performs quantification based on the gas concentration change amplitude and spatial distribution dispersion to obtain gas response sparse feature values. The heat diffusion trajectory analysis module calculates the local heat diffusion trajectory of potential hidden heat source areas based on the sparse characteristic values ​​of the gas response, extracts the offset angle and expansion rate parameters of the heat expansion direction, and generates a heat accumulation intensity index. The thermal inertia feature extraction module implements point-by-point thermal disturbance excitation in potential concealed heat source areas based on the heat accumulation intensity index. By introducing small-amplitude environmental disturbances, it obtains the temperature rebound amplitude, extracts thermal inertia enhancement features, and updates the concealed combustion criteria. The graded response control module performs graded response control for potential concealed heat source areas according to the updated concealed combustion criteria, and triggers differentiated intervention strategies based on the magnitude of the thermal inertia enhancement characteristics.

[0080] The intelligent forest management method based on satellite remote sensing and ground sensor network provided in this embodiment of the invention is implemented through the above-mentioned intelligent forest management system based on satellite remote sensing and ground sensor network. For details of the specific methods and processes of the intelligent forest management system based on satellite remote sensing and ground sensor network, please refer to the embodiments of the intelligent forest management method based on satellite remote sensing and ground sensor network, which will not be repeated here.

[0081] The foregoing has only described certain exemplary embodiments of this application by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of this application. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of this application.

Claims

1. A method for intelligent forest management based on satellite remote sensing and ground sensor networks, characterized in that, Includes the following steps: Acquire information on thermal anomaly patches in satellite thermal infrared images, extract the amplitude of radiation temperature gradient change in the corresponding region of thermal anomaly patches, and mark the region where the amplitude of radiation temperature gradient change exceeds a preset threshold as a potential hidden heat source area. Based on potential hidden heat source areas, nearby ground gas sensing nodes are scheduled to perform high-frequency gas concentration acquisition. The gas concentration change amplitude and spatial distribution dispersion are quantified to obtain sparse characteristic values ​​of gas response. Based on the sparse characteristic values ​​of the gas response, the local heat diffusion trajectory of the potential hidden heat source area is calculated, the heat outward expansion direction offset angle and expansion rate parameters are extracted, and the heat accumulation intensity index is generated. Based on the heat accumulation intensity index, point-by-point thermal disturbance excitation is implemented in potential concealed heat source areas. By introducing small-amplitude environmental disturbances, the temperature rebound amplitude is obtained, the thermal inertia enhancement characteristics are extracted, and the concealed combustion criterion is updated. Based on the updated concealed combustion criteria, graded response control is implemented for potential concealed heat source areas, and differentiated intervention strategies are triggered according to the magnitude of the thermal inertia enhancement characteristics.

2. The intelligent forest management method based on satellite remote sensing and ground sensor network according to claim 1, characterized in that, Areas where the amplitude of the radiation temperature gradient change exceeds a preset threshold are identified as potential hidden heat source areas. The specific steps are as follows: Acquire satellite thermal infrared image data within the forest coverage area, establish a unified spatial grid according to the image pixel correspondence, arrange the radiation temperature values ​​of the same spatial location at different times, filter spatial locations with increasing temperature at three consecutive times and expand to eight adjacent areas, delineate temperature rising areas and identify thermal anomaly patches. The area covered by the thermal anomaly patch was divided into sub-regions and the radiation temperature values ​​were recorded. The temperature difference in each direction and the change of continuous time interval were calculated in adjacent sub-regions. The spatial difference changes and the temporal changes were summarized and superimposed to obtain the radiation temperature gradient change amplitude of the corresponding sub-region. The amplitude of the radiation temperature gradient change in each sub-region is compared with a preset threshold. Candidate heat source units that exceed the preset threshold for three consecutive time points are marked. The spatial adjacency relationship is combined for merging and the spatial range is continuously updated. Regions whose radiation temperature gradient change amplitude exceeds the preset threshold are marked as potential hidden heat source areas.

3. The intelligent forest management method based on satellite remote sensing and ground sensor network according to claim 2, characterized in that, The specific steps for obtaining sparse eigenvalues ​​of the gas response are as follows: The boundary of the potential hidden heat source area is selected as the reference starting point. The spatial position of the ground gas sensing node is located one by one and the straight distance between it and the nearest point of the potential hidden heat source area boundary is calculated. Ground gas sensing nodes with a distance less than a fixed distance range are included in the scheduling range and a unified time starting point is assigned to perform gas concentration collection. Based on the gas concentration values ​​obtained by ground gas sensor nodes within the scheduling range, the gas concentration changes are recorded in chronological order and the gas concentration change amplitude is calculated. The difference in gas concentration change amplitude between adjacent ground gas sensor nodes is arranged in spatial order to obtain the gas concentration change amplitude sequence and spatial distribution data sequence. The effective change segments are selected based on the gas concentration change amplitude sequence, and the proportion of ground gas sensing nodes and the number of continuous segments are counted. The numerical level is output as the gas response sparsity feature value according to the matching relationship between the node proportion value and the number of continuous segments in the corresponding table.

4. The intelligent forest management method based on satellite remote sensing and ground sensor network according to claim 3, characterized in that, In the process of screening effective change segments from the gas concentration change amplitude sequence, the time period corresponding to the change amplitude that is greater than zero in a continuous time interval is marked as an effective change segment. The number of effective change segments corresponding to each ground gas sensing node is counted. Combined with the spatial distribution data sequence, the distribution of effective change segments of adjacent nodes is checked in spatial order to divide continuous segments.

5. The intelligent forest management method based on satellite remote sensing and ground sensor network according to claim 3, characterized in that, The specific steps for generating a heat accumulation intensity index are as follows: The geometric center location is determined based on the potential hidden heat source region corresponding to the sparse eigenvalues ​​of the gas response, and the observation direction is divided. The radiation temperature values ​​of continuous sub-regions are read along each observation direction to construct a temperature change record that includes spatial order and temporal variation. By combining temperature change records to filter observation directions and construct heat diffusion paths, connecting the temperature change locations of continuous sub-regions, extracting the path deflection of each observation direction and the change in the number of sub-regions per unit time, we can obtain the heat outward expansion direction offset angle and expansion rate parameters. The primary and secondary diffusion directions are divided by the offset angle and expansion rate parameters of the heat diffusion direction. The diffusion state is statistically analyzed and matched with the corresponding relationship table. The corresponding numerical level is output to obtain the heat accumulation intensity index.

6. The intelligent forest management method based on satellite remote sensing and ground sensor network according to claim 5, characterized in that, The primary and secondary diffusion directions are divided using the offset angle of the heat diffusion direction and the expansion rate parameter. This includes: taking the angle between each observation path and a pre-selected reference direction as the offset angle; marking the primary diffusion direction when the offset angle is less than or equal to 45 degrees and the expansion rate parameter is increasing; and marking the secondary diffusion direction when the offset angle is greater than 45 degrees or the expansion rate parameter is decreasing or remains unchanged.

7. The intelligent forest management method based on satellite remote sensing and ground sensor network according to claim 5, characterized in that, The specific steps for extracting thermal inertia enhancement features and updating the concealed combustion criterion are as follows: Using each sub-region within a potential hidden heat source area as a processing unit, the radiation temperature values ​​are recorded and short-term airflow changes are introduced to obtain a temperature recording sequence. Read the temperature record sequence and select the radiation temperature value at the end of the disturbance as the rebound start reference value. Calculate the temperature difference and extract the temperature rebound amplitude. Arrange the temperature rebound amplitude and record it in correspondence with the heat accumulation intensity index. The temperature rebound amplitude is divided and thermal inertia enhancement units are screened. Continuous regions are merged and spatial coverage is statistically analyzed to update the concealed combustion criteria.

8. The intelligent forest management method based on satellite remote sensing and ground sensor network according to claim 7, characterized in that, The process of reading the temperature record sequence and selecting the radiation temperature value at the end of the disturbance as the rebound start reference value also includes: comparing the radiation temperature values ​​at the end of the disturbance and at continuous time points after the end of the disturbance, selecting the maximum temperature difference as the temperature rebound amplitude, and recording the temperature rebound amplitude in correspondence with the heat accumulation intensity index.

9. The intelligent forest management method based on satellite remote sensing and ground sensor network according to claim 7, characterized in that, The differentiated intervention strategy is triggered based on the magnitude of the thermal inertia enhancement feature. The specific steps are as follows: Obtain the concealed combustion criterion level and thermal inertia enhancement characteristic value, classify the state into high-level state, intermediate-level state and low-level state, and incorporate them into the first response sequence, second response sequence and third response sequence respectively, and sort the response execution units according to the distance of the center location of the potential concealed heat source area. Based on the sorting results of the response execution units, the values ​​of thermal inertia enhancement characteristics are statistically analyzed and the average value is calculated. High-intensity intervention intervals and low-intensity intervention intervals are divided. Continuous airflow disturbance operation or single airflow disturbance operation is performed for the first response sequence, interval airflow disturbance operation is performed for the second response sequence, and the state record is retained for the third response sequence. Based on the temperature change trend over continuous time, the first response sequence, the second response sequence, and the third response sequence are adjusted, and differentiated intervention strategies are used to achieve differentiated response control of potential hidden heat sources.

10. A forest area intelligent management system based on satellite remote sensing and ground sensor networks, used to implement the forest area intelligent management method based on satellite remote sensing and ground sensor networks as described in any one of claims 1-9, characterized in that, It includes a thermal infrared anomaly identification module, a gas rarefaction feature extraction module, a thermal diffusion trajectory analysis module, a thermal inertial feature extraction module, and a graded response control module; The thermal infrared anomaly identification module acquires information on thermal anomaly patches in satellite thermal infrared images and extracts the amplitude of radiation temperature gradient changes in the corresponding areas of thermal anomaly patches. Areas with radiation temperature gradient changes exceeding a preset threshold are identified as potential hidden heat source areas. The gas sparse feature extraction module, based on potential hidden heat source areas, schedules nearby ground gas sensing nodes to perform high-frequency gas concentration acquisition, and performs quantification based on the gas concentration change amplitude and spatial distribution dispersion to obtain gas response sparse feature values. The heat diffusion trajectory analysis module calculates the local heat diffusion trajectory of potential hidden heat source areas based on the sparse characteristic values ​​of the gas response, extracts the offset angle and expansion rate parameters of the heat expansion direction, and generates a heat accumulation intensity index. The thermal inertia feature extraction module implements point-by-point thermal disturbance excitation in potential concealed heat source areas based on the heat accumulation intensity index. By introducing small-amplitude environmental disturbances, it obtains the temperature rebound amplitude, extracts thermal inertia enhancement features, and updates the concealed combustion criteria. The graded response control module performs graded response control for potential concealed heat source areas according to the updated concealed combustion criteria, and triggers differentiated intervention strategies based on the magnitude of the thermal inertia enhancement characteristics.

Images