Forestry intelligent surveying and mapping method and system based on regional feature feedback

Through intelligent surveying and mapping methods based on regional feature feedback, satellite remote sensing and drone data are used to generate priority monitoring grids and health index heat maps, and the category system is dynamically updated, which solves the problem of resource waste and response extension caused by full coverage scanning of drones, and achieves efficient forestry monitoring.

CN120526307AActive Publication Date: 2025-08-22SHANDONG ZHIHUI YUNTU GEOGRAPHIC INFORMATION ENG CO LTD

Patent Information

Application Number
CN202510607909.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-05-13
Publication Date
2025-08-22
Estimated Expiration
2045-05-13

Smart Images

  • Figure CN120526307A_ABST
    Figure CN120526307A_ABST
Patent Text Reader

Abstract

The invention discloses an intelligent forestry surveying and mapping method and system based on regional feature feedback, relates to the technical field of forest surveying and mapping, and provides the following scheme: extracting spectrum and texture features of a target forest image through satellite remote sensing data, realizing recognition and grading division of feature regions by using an intelligent segmentation algorithm, and obtaining a target forest image. A priority monitoring grid of the content evaluation indexes is generated; and scheduling the unmanned aerial vehicle to scan the priority monitoring grid for the first time, and integrating image data of the unmanned aerial vehicle and the satellite. According to the method, spectrum and texture features of a target forest region are extracted through satellite remote sensing data, a non-uniform priority monitoring grid containing a feature density index is generated, an unmanned aerial vehicle is scheduled to scan a high-priority grid for the first time, satellite multi-spectrum data, an unmanned aerial vehicle high-resolution image and LiDAR point cloud data are fused, and the target forest region is obtained. The tree species are identified, the health index thermodynamic diagram is generated, high-danger areas are rapidly identified, the scanning range of the unmanned aerial vehicle is remarkably compressed, and the monitoring efficiency is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of forest area surveying and mapping, and in particular to a forestry intelligent surveying and mapping method and system based on regional feature feedback. Background Art

[0002] Currently, forestry monitoring mainly adopts a collaborative operation mode of satellite remote sensing and drones. Satellites provide large-scale forest coverage data through multispectral images, combined with supervised classification, such as random forest classification of vegetation types. Drones use high-resolution images or LiDAR data to analyze tree species and detect areas with pests and diseases through the NDVI threshold method.

[0003] Existing drone scanning areas are based on a fixed division strategy of satellite imagery. This requires a mandatory full forest coverage scan to generate a health heat map, resulting in a large number of ineffective scans in healthy areas, causing significant resource redundancy and energy consumption. Furthermore, since a feature feedback mechanism is not established for the first scan, the identification of high-risk areas must wait until data from the entire forest area is collected. This significantly prolongs the response cycle from data collection to risk warning. Therefore, we propose a forestry intelligent mapping method and system based on regional feature feedback to address this problem. Summary of the Invention

[0004] In order to solve the above technical problems, a forestry intelligent mapping method and system based on regional feature feedback is provided. This technical solution solves the problem that the existing drone scanning area is fixedly divided based on satellite images proposed in the above background technology. It is necessary to complete the full forest area coverage scan before a health heat map can be generated, which causes the drone to perform a large number of invalid scans in the healthy area, resulting in significant resource redundancy and energy consumption. Moreover, since the feature feedback mechanism is not established in the first scan, the identification of high-risk areas must wait for the completion of data collection in the entire forest area, resulting in a significant extension of the response cycle from data collection to risk warning.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: The present invention provides a forestry intelligent surveying and mapping method based on regional feature feedback, the method comprising: Step 1: Extract the spectral and texture features of the target forest area image through satellite remote sensing data, use intelligent segmentation algorithms to identify and grade feature areas, and generate priority monitoring grids containing quantitative assessment indicators; Step 2: Dispatch drones to perform an initial scan of the priority monitoring grid, integrate drone and satellite image data, fuse spectral features and 3D structural parameters to build a tree species recognition model, and establish a classification system with tree species nouns as labels; Step 3: Based on the healthy tree species image feature threshold library, quantify the degree of deviation from the health status of each area, and generate a health index heat map of the target forest area through spatial statistical analysis; Step 4: Establish a dynamic mapping relationship between health index and risk level, dynamically update the classification system based on the drone scanning results, and plan the drone's secondary scanning of the target forest area according to the health risk level.

[0006] Furthermore, the spectral and texture features of the target forest area image are extracted through satellite remote sensing data, and the intelligent segmentation algorithm is used to realize the identification and classification of the feature area, and generate a priority monitoring grid containing quantitative evaluation indicators, specifically including: Obtain multispectral satellite remote sensing images of the target forest area, extract spectral reflectance after radiometric calibration and atmospheric correction preprocessing, and calculate texture feature parameters based on the gray-level co-occurrence matrix to generate preliminary feature regions; The normalized spectral reflectance and texture feature parameter values ​​are concatenated into a feature vector, and the preliminary feature regions are clustered based on the vector, and first-level cluster labels are assigned. The satellite remote sensing image and cluster labels are input into the semantic segmentation model. After optimization through the cross entropy loss function, the probability distribution map of each pixel belonging to different feature categories is output. Based on the probability distribution density, an adaptive method is used to divide the non-uniform grid, which is recorded as the monitoring grid. The ratio of the sum of the pixel category probabilities in each grid to the grid area is calculated as the feature density index, and the priority areas are divided according to the preset density threshold.

[0007] Furthermore, the dispatching drone performs an initial scan of the priority monitoring grid, integrates drone and satellite image data, fuses spectral features and three-dimensional structural parameters to construct a tree species recognition model, and establishes a classification system with tree species nouns as labels, specifically including: Hyperspectral imaging and LiDAR point cloud data of the area to be scanned are collected, and the Euclidean distance between each pixel and the standard tree species spectral library is calculated as the spectral fingerprint difference. At the same time, DEM and DSM are generated based on the point cloud, and leaf density gradient and canopy convexity are extracted as vertical structure parameters; The random forest model is used to fuse spectral fingerprint differences and vertical structure parameters to dynamically distinguish tree species, and the proportion of pixels of each tree species in the grid is calculated as its distribution area. Combine the drone spectral fingerprint with satellite remote sensing imagery to perform mixed pixel decomposition, calculate the proportion of each tree species in the satellite remote sensing image pixels, analyze the tree species composition of the preliminary feature area, and refine the first-level cluster labels into second-level cluster labels based on the tree species name; The tree species in the non-preliminary feature areas are clustered, and when the cumulative distribution area of ​​a certain tree species reaches the clustering threshold, a secondary cluster label is assigned.

[0008] Furthermore, the method quantifies the degree of deviation of the health status of each area based on the healthy tree species image feature threshold library, and generates a health index heat map of the target forest area through spatial statistical analysis, specifically including: For each tree species in the secondary clustering label, obtain the spectral reflectance threshold and canopy concavity threshold under its healthy state, and adjust each threshold dynamically according to the season; The spectral reflectance difference and canopy convexity difference between the target tree species and the health threshold are calculated pixel by pixel. In principle, abnormal difference data should be eliminated; The distribution area of ​​the tree species is used as the weight coefficient, and the weighted sum is used to generate the health coefficient of the tree species; The Kriging spatial interpolation method is used to compensate for data gaps, summarize the health coefficients of each tree species and perform weighted average, and output a health index heat map of the target forest area.

[0009] Furthermore, the establishment of a dynamic mapping relationship between health index and risk level, and the dynamic updating of the classification system based on the drone's scanning results, specifically includes: Set the health coefficient threshold interval, mark the secondary cluster labels whose total health coefficient exceeds the upper limit of the threshold interval as dangerous labels, those within the interval as potential dangerous labels, and those below the lower limit of the threshold interval as low dangerous labels; When the drone scans a new area to be scanned, the total health coefficient of the secondary cluster label is recalculated; Through the streaming computing engine, the total health coefficient is compared with the health coefficient threshold, and the secondary cluster labels and their label marks are updated.

[0010] Furthermore, the second scanning of the target forest area by the drone according to the health risk level specifically includes: Based on the health index heat map and health coefficient threshold, the target forest area is divided into dangerous areas, potential dangerous areas and low-danger areas; If the proportion of the hazardous area in the characteristic area does not reach the set proportion of the hazardous area coverage, the uncovered hazardous area will be scanned again by the drone, and a certain proportion of overlapping scanning bands will be set to perform sampling scans of non-hazardous areas. A graded sensor configuration will be used according to the regional hazard level. Oblique photography of the danger zone is carried out and modeling is used to generate a 3D diseased canopy model, which is then combined with an onboard artificial intelligence algorithm to identify pests and diseases in real time and adjust flight parameters.

[0011] Furthermore, a forestry intelligent surveying and mapping system based on regional feature feedback is proposed, which is used to implement any of the above-mentioned forestry intelligent surveying and mapping methods based on regional feature feedback, comprising: The multi-source remote sensing feature extraction and classification module is used to extract the spectral and texture features of the target forest area image through satellite remote sensing data, use intelligent segmentation algorithms to realize the identification and classification of feature areas, and generate priority monitoring grids containing quantitative evaluation indicators; The air-ground collaborative tree species identification module is used to dispatch drones to perform initial scans of priority monitoring grids, integrate drone and satellite image data, fuse spectral features and three-dimensional structural parameters to build a tree species identification model, and establish a classification system labeled by tree species nouns; The health dynamic assessment module is used to quantify the degree of deviation from the health status of each area based on the threshold library of healthy tree species image features, and generate a health index heat map of the target forest area through spatial statistical analysis; The adaptive response control module is used to establish a dynamic mapping relationship between health index and risk level, dynamically update the classification system based on the drone's scanning results, and plan the drone's secondary scanning of the target forest area according to the health risk level.

[0012] Compared with the prior art, the present invention has the following beneficial effects: Using satellite remote sensing data to extract the spectral and textural characteristics of target forest areas, a semantic segmentation algorithm is used to intelligently identify and grade characteristic regions, generating a non-uniform priority monitoring grid with a characteristic density index. Drones are dispatched to perform initial scans of high-priority grids. The system then integrates satellite multispectral data with high-resolution drone imagery and LiDAR point cloud data to identify tree species and generate health index heat maps. This allows for rapid identification of high-risk areas, significantly reducing the drone scanning range and improving monitoring efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 This is a flow chart of a forestry intelligent mapping method based on regional feature feedback proposed by the present invention; Figure 2 A flow chart of the method for tree species identification and classification in the present invention; Figure 3 This is a flow chart for obtaining the health index heat map in the present invention; Figure 4 This is a flow chart of the method for secondary scanning of a drone in the present invention; Figure 5 This is a structural block diagram of a forestry intelligent surveying and mapping system based on regional feature feedback proposed by the present invention. DETAILED DESCRIPTION

[0014] The following description is intended to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are merely examples, and those skilled in the art may conceive of other obvious variations.

[0015] Reference Figure 1-5 As shown, a forestry intelligent mapping method based on regional feature feedback includes: Step 1: Extract the spectral and texture features of the target forest area image through satellite remote sensing data, use intelligent segmentation algorithms to identify and grade feature areas, and generate priority monitoring grids containing quantitative assessment indicators; Step 2: Dispatch drones to perform an initial scan of the priority monitoring grid, integrate drone and satellite image data, fuse spectral features and 3D structural parameters to build a tree species recognition model, and establish a classification system with tree species nouns as labels; Step 3: Based on the healthy tree species image feature threshold library, quantify the degree of deviation from the health status of each area, and generate a health index heat map of the target forest area through spatial statistical analysis; Step 4: Establish a dynamic mapping relationship between health index and risk level, dynamically update the classification system based on the drone scanning results, and plan the drone's secondary scanning of the target forest area according to the health risk level.

[0016] Specifically, the method extracts the spectral and texture features of the target forest area image through satellite remote sensing data, uses an intelligent segmentation algorithm to identify and grade the feature areas, and generates a priority monitoring grid containing quantitative evaluation indicators, which specifically includes: Obtain multispectral satellite remote sensing images of the target forest area, extract spectral reflectance after radiometric calibration and atmospheric correction preprocessing, and calculate texture feature parameters based on the gray-level co-occurrence matrix to generate preliminary feature regions; The normalized spectral reflectance and texture feature parameter values ​​are concatenated into a feature vector, and the preliminary feature regions are clustered based on the vector, and first-level cluster labels are assigned. The satellite remote sensing image and cluster labels are input into the semantic segmentation model. After optimization through the cross entropy loss function, the probability distribution map of each pixel belonging to different feature categories is output. Based on the probability distribution density, an adaptive method is used to divide the non-uniform grid, which is recorded as the monitoring grid. The ratio of the sum of the pixel category probabilities in each grid to the grid area is calculated as the feature density index, and priority areas are divided according to the preset density threshold; When necessary, a topological check is performed on the divided priority areas to ensure the spatial continuity of adjacent grids; illustratively, the preset density threshold can be 0.85, which is obtained by querying the 2024 Technical Specifications for Forest Remote Sensing Monitoring.

[0017] Reference Figure 2 As shown, the dispatched drone performs the first scan of the priority monitoring grid, integrates the image data of the drone and the satellite, fuses the spectral characteristics and three-dimensional structure parameters to build a tree species recognition model, and establishes a classification system with tree species nouns as labels, specifically including: Hyperspectral imaging and LiDAR point cloud data of the area to be scanned are collected, and the Euclidean distance between each pixel and the standard tree species spectral library is calculated as the spectral fingerprint difference. At the same time, DEM and DSM are generated based on the point cloud, and leaf density gradient and canopy convexity are extracted as vertical structure parameters; The random forest model is used to fuse spectral fingerprint differences and vertical structure parameters to dynamically distinguish tree species, and the proportion of pixels of each tree species in the grid is calculated as its distribution area. Combine the drone spectral fingerprint with satellite remote sensing imagery to perform mixed pixel decomposition, calculate the proportion of each tree species in the satellite remote sensing image pixels, analyze the tree species composition of the preliminary feature area, and refine the first-level cluster labels into second-level cluster labels based on the tree species name; Cluster the tree species in the non-preliminary feature areas. When the cumulative distribution area of ​​a certain tree species reaches the clustering threshold, assign a secondary cluster label. Exemplarily, the clustering threshold is dynamically calculated based on the total grid area of ​​the target forest area, and the clustering threshold = total grid area × 0.15%.

[0018] Reference Figure 3 As shown, the method quantifies the degree of deviation of the health status of each area based on the healthy tree species image feature threshold library, and generates a health index heat map of the target forest area through spatial statistical analysis, specifically including: For each tree species in the secondary clustering label, the spectral reflectance threshold and canopy concavity threshold under its healthy state are obtained, and the thresholds are dynamically adjusted according to the season. The seasonal dynamic adjustment rule is that the threshold is increased by 15% in summer and decreased by 10% in winter. The spectral reflectance difference and canopy convexity difference between the target tree species and the health threshold are calculated pixel by pixel. In principle, eliminate abnormal difference data; 3 In principle, data exceeding ±2.5 times the standard deviation of the mean were specifically eliminated; The distribution area of ​​the tree species is used as the weight coefficient, and the weighted sum is used to generate the health coefficient of the tree species; The Kriging spatial interpolation method is used to compensate for data gaps, summarize the health coefficients of each tree species and perform weighted average, and output a health index heat map of the target forest area.

[0019] Specifically, the establishment of a dynamic mapping relationship between health index and risk level, and the dynamic updating of the classification system based on the drone's scanning results, specifically include: Set the health coefficient threshold interval, mark the secondary cluster labels whose total health coefficient exceeds the upper limit of the threshold interval as dangerous labels, those within the interval as potential dangerous labels, and those below the lower limit of the threshold interval as low dangerous labels; For example, the health coefficient threshold interval can be set to [0.35, 0.65], where 0.35 corresponds to the disease outbreak threshold defined in the technical regulations for forest health assessment, and 0.65 is the 25th percentile of the healthy tree species sample library; When the drone scans a new area to be scanned, the total health coefficient of the secondary cluster label is recalculated; Through the streaming computing engine, the total health coefficient is compared with the health coefficient threshold, and the secondary cluster labels and their label marks are updated.

[0020] Reference Figure 4 As shown, the second scanning of the target forest area by the drone according to the health risk level specifically includes: Based on the health index heat map and health coefficient threshold, the target forest area is divided into dangerous areas, potential dangerous areas and low-danger areas; If the proportion of the hazardous area in the characteristic area does not reach the set proportion of the hazardous area coverage, the uncovered hazardous area will be scanned again by the drone, and a certain proportion of overlapping scanning bands will be set to perform sampling scans of non-hazardous areas. A graded sensor configuration will be used according to the regional hazard level. For example, 50m low-altitude multispectral LiDAR scanning is used in dangerous areas, 100m high-altitude visible light + thermal infrared monitoring is used in potential dangerous areas, and sample visible light scanning is implemented in low-risk areas to comprehensively improve the monitoring efficiency of drones; For example, the coverage ratio of the dangerous area can be set to 80%, and the overlapping scanning band width can be 12±2%. The 80% coverage ensures the monitoring integrity of the key area, and the 12% overlapping band can meet the image stitching accuracy requirements; Oblique photography of the danger zone is carried out and modeling is used to generate a 3D diseased canopy model, which is then combined with an onboard artificial intelligence algorithm to identify pests and diseases in real time and adjust flight parameters.

[0021] Reference Figure 5 As shown, this solution proposes a forestry intelligent surveying and mapping system based on regional feature feedback, which is used to implement the above-mentioned forestry intelligent surveying and mapping method based on regional feature feedback, including: The multi-source remote sensing feature extraction and classification module is used to extract the spectral and texture features of the target forest area image through satellite remote sensing data, use intelligent segmentation algorithms to realize the identification and classification of feature areas, and generate priority monitoring grids containing quantitative evaluation indicators; The air-ground collaborative tree species identification module is used to dispatch drones to perform initial scans of priority monitoring grids, integrate drone and satellite image data, fuse spectral features and three-dimensional structural parameters to build a tree species identification model, and establish a classification system labeled by tree species nouns; The health dynamic assessment module is used to quantify the degree of deviation from the health status of each area based on the threshold library of healthy tree species image features, and generate a health index heat map of the target forest area through spatial statistical analysis; The adaptive response control module is used to establish a dynamic mapping relationship between health index and risk level, dynamically update the classification system based on the drone's scanning results, and plan the drone's secondary scanning of the target forest area according to the health risk level.

[0022] The air-ground collaborative tree species identification module specifically includes: Multi-source data fusion and feature extraction unit, used to achieve collaborative processing and feature quantification of hyperspectral and LiDAR data; Dynamic tree species classification decision unit, used to integrate multi-dimensional features to achieve accurate tree species classification and area statistics; The label system optimization unit is used to complete the dynamic construction and update of the label system.

[0023] The health dynamic assessment module specifically includes: A dynamic management unit for health characteristic thresholds, used to achieve seasonal adaptive adjustment and verification of health thresholds for multiple tree species; Health anomaly detection and coefficient calculation subunit, used to accurately quantify the health of individual tree species; The spatial modeling and heat map generation unit is used to build and visualize the global health spatial distribution model.

[0024] The adaptive response control module specifically includes: Dynamic risk grading and label management unit, used to achieve dynamic quantification of health risks and real-time label updates; Intelligent scanning scheduling unit for adaptive drone mission planning based on risk level; Real-time decision-making and 3D modeling units are used to achieve accurate disease identification and dynamic optimization of flight parameters.

[0025] The above shows and describes the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions merely illustrate the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. A forestry intelligent mapping method based on regional feature feedback, characterized in that: The method comprises: Step 1: Extract the spectral and texture features of the target forest area image through satellite remote sensing data, use intelligent segmentation algorithms to identify and grade feature areas, and generate priority monitoring grids containing quantitative assessment indicators; Step 2: Dispatch drones to perform an initial scan of the priority monitoring grid, integrate drone and satellite image data, fuse spectral features and 3D structural parameters to build a tree species recognition model, and establish a classification system with tree species nouns as labels; Step 3: Based on the healthy tree species image feature threshold library, quantify the degree of deviation from the health status of each area, and generate a health index heat map of the target forest area through spatial statistical analysis; Step 4: Establish a dynamic mapping relationship between health index and risk level, dynamically update the classification system based on the drone scanning results, and plan the drone's secondary scanning of the target forest area according to the health risk level.

2. The method according to claim 1, characterized in that The method extracts the spectral and texture features of the target forest area image through satellite remote sensing data, uses an intelligent segmentation algorithm to identify and grade feature areas, and generates a priority monitoring grid containing quantitative evaluation indicators, specifically including: Obtain multispectral satellite remote sensing images of the target forest area, extract spectral reflectance after radiometric calibration and atmospheric correction preprocessing, and calculate texture feature parameters based on the gray-level co-occurrence matrix to generate preliminary feature regions; The normalized spectral reflectance and texture feature parameter values ​​are concatenated into a feature vector, and the preliminary feature regions are clustered based on the vector, and first-level cluster labels are assigned. The satellite remote sensing image and cluster labels are input into the semantic segmentation model. After optimization through the cross entropy loss function, the probability distribution map of each pixel belonging to different feature categories is output. Based on the probability distribution density, an adaptive method is used to divide the non-uniform grid, which is recorded as the monitoring grid. The ratio of the sum of the pixel category probabilities in each grid to the grid area is calculated as the feature density index, and the priority areas are divided according to the preset density threshold.

3. The method according to claim 1, characterized in that The dispatched drone performs an initial scan of the priority monitoring grid, integrates drone and satellite image data, fuses spectral features and three-dimensional structural parameters to construct a tree species recognition model, and establishes a classification system with tree species nouns as labels, specifically including: Collect hyperspectral imaging and LiDAR point cloud data of the area to be scanned, calculate the Euclidean distance between each pixel and the standard tree species spectral library as the spectral fingerprint difference, and generate DEM and DSM based on the point cloud, extracting the leaf layer density gradient and canopy convexity as vertical structure parameters; The random forest model is used to fuse spectral fingerprint differences and vertical structure parameters to dynamically distinguish tree species, and the proportion of pixels of each tree species in the grid is calculated as its distribution area. Combine the drone spectral fingerprint with satellite remote sensing imagery to perform mixed pixel decomposition, calculate the proportion of each tree species in the satellite remote sensing image pixels, analyze the tree species composition of the preliminary feature area, and refine the first-level cluster labels into second-level cluster labels based on the tree species name; The tree species in the non-preliminary feature areas are clustered, and when the cumulative distribution area of ​​a certain tree species reaches the clustering threshold, a secondary cluster label is assigned.

4. The method according to claim 3, characterized in that The method is based on the healthy tree species image feature threshold library, quantifies the degree of deviation of the health status of each area, and generates a health index heat map of the target forest area through spatial statistical analysis, specifically including: For each tree species in the secondary clustering label, obtain the spectral reflectance threshold and canopy concavity threshold under its healthy state, and adjust each threshold dynamically according to the season; The difference between the spectral reflectance of the target tree species and the health threshold, as well as the canopy concavity and convexity, was calculated pixel by pixel using 3 In principle, abnormal difference data should be eliminated; The distribution area of ​​the tree species is used as the weight coefficient, and the weighted sum is used to generate the health coefficient of the tree species; The Kriging spatial interpolation method is used to compensate for data gaps, summarize the health coefficients of each tree species and perform weighted average, and output a health index heat map of the target forest area.

5. The method according to claim 4, characterized in that The establishment of a dynamic mapping relationship between health index and risk level and the dynamic updating of the classification system based on the drone's scanning results specifically include: Set the health coefficient threshold interval, mark the secondary cluster labels whose total health coefficient exceeds the upper limit of the threshold interval as dangerous labels, those within the interval as potential dangerous labels, and those below the lower limit of the threshold interval as low dangerous labels; When the drone scans a new area to be scanned, the total health coefficient of the secondary cluster label is recalculated; Through the streaming computing engine, the total health coefficient is compared with the health coefficient threshold, and the secondary cluster labels and their label marks are updated.

6. The method according to claim 5, characterized in that The second scanning of the target forest area by the drone according to the health risk level specifically includes: Based on the health index heat map and health coefficient threshold, the target forest area is divided into dangerous areas, potential dangerous areas and low-danger areas; If the proportion of the hazardous area in the characteristic area does not reach the set proportion of the hazardous area coverage, the uncovered hazardous area will be scanned again by the drone, and a certain proportion of overlapping scanning bands will be set to perform sampling scans of non-hazardous areas. A graded sensor configuration will be used according to the regional hazard level. Oblique photography of the danger zone is carried out and modeling is used to generate a 3D diseased canopy model, which is then combined with an onboard artificial intelligence algorithm to identify pests and diseases in real time and adjust flight parameters.

7. A forestry intelligent surveying and mapping system based on regional feature feedback, characterized in that: A method for implementing a forestry intelligent surveying and mapping method based on regional feature feedback as described in any one of claims 1 to 6, comprising: The multi-source remote sensing feature extraction and classification module is used to extract the spectral and texture features of the target forest area image through satellite remote sensing data, use intelligent segmentation algorithms to realize the identification and classification of feature areas, and generate priority monitoring grids containing quantitative evaluation indicators; The air-ground collaborative tree species identification module is used to dispatch drones to perform initial scans of priority monitoring grids, integrate drone and satellite image data, fuse spectral features and three-dimensional structural parameters to build a tree species identification model, and establish a classification system labeled by tree species nouns; The health dynamic assessment module is used to quantify the degree of deviation from the health status of each area based on the threshold library of healthy tree species image features, and generate a health index heat map of the target forest area through spatial statistical analysis; The adaptive response control module is used to establish a dynamic mapping relationship between health index and risk level, dynamically update the classification system based on the drone's scanning results, and plan the drone's secondary scanning of the target forest area according to the health risk level.

8. The forestry intelligent surveying and mapping system based on regional feature feedback according to claim 7 is characterized in that: The air-ground collaborative tree species identification module specifically includes: Multi-source data fusion and feature extraction unit, used to achieve collaborative processing and feature quantification of hyperspectral and LiDAR data; Dynamic tree species classification decision unit, used to integrate multi-dimensional features to achieve accurate tree species classification and area statistics; The label system optimization unit is used to complete the dynamic construction and update of the label system.

9. The forestry intelligent surveying and mapping system based on regional feature feedback according to claim 7, characterized in that: The health dynamic assessment module specifically includes: A dynamic management unit for health characteristic thresholds, used to achieve seasonal adaptive adjustment and verification of health thresholds for multiple tree species; Health anomaly detection and coefficient calculation subunit, used to accurately quantify the health of individual tree species; The spatial modeling and heat map generation unit is used to build and visualize the global health spatial distribution model.

10. The forestry intelligent surveying and mapping system based on regional feature feedback according to claim 7, characterized in that: The adaptive response control module specifically includes: Dynamic risk grading and label management unit, used to achieve dynamic quantification of health risks and real-time label updates; Intelligent scanning scheduling unit for adaptive drone mission planning based on risk level; Real-time decision-making and 3D modeling units are used to achieve accurate disease identification and dynamic optimization of flight parameters.

Citation Information

Patent Citations

  • Individual tree biomass estimation method based on hyperspectrum and air-ground collaborative LiDAR

    CN118570677A

  • Priority-based, facial recognition-assisted attendance determination and validation system

    US10789796B1

  • Landslide hazard monitoring and early warning method and system based on real 3D

    US12130401B1

  • Fire detection systems and methods

    US5832187A

  • Disease examination system and method based on eye image

    WO2022252107A1

Cited By

  • Unmanned aerial vehicle detection and control method and system for regional defense

    CN121028853A

  • Permanent basic farmland quality intelligent monitoring system and method

    CN121169088A