Forest and grass informatization-oriented ecological space multi-source data integration method and system

Abstract

The invention relates to the technical field of data processing, and discloses a forest grass informatization-oriented ecological space multi-source data integration method and system, and the method comprises the steps: endowing forest grass business data and Internet of Things sensing data with geographic coordinates and forest grass compartment codes based on a forest region geographic base map, and constructing a time-space fusion modeling data set in combination with a time label; extracting space coordinates of the unmanned aerial vehicle aerial image and the remote sensing image based on the space-time fusion modeling data set to obtain a first coordinate point set, and splicing the unmanned aerial vehicle aerial image and the remote sensing image to obtain a forest region static geographic map; based on the space-time fusion modeling data set, a control point is determined from the static geographic map of the forest region, a second coordinate point set is obtained from the real-time video of the unmanned aerial vehicle, and then the real-time video of the unmanned aerial vehicle and the static geographic map of the forest region are fused to obtain a dynamic geographic map layer of the forest region; and obtaining forest region multi-source data corresponding to the forest grass ecological space based on the forest region dynamic geographic map layer, the space-time fusion modeling data set, the forest grass business data and the internet-of-things perception data.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and relates to, but is not limited to, a method and system for integrating multi-source ecological spatial data for forestry and grassland informatization. Background Technology

[0002] With the increasing demand for refined forestry and grassland ecological protection and management, the data sources involved in forestry and grassland operations are becoming increasingly diverse, covering various heterogeneous data such as business approvals, IoT sensing, remote sensing imagery, and basic support. Currently, in the construction of forestry and grassland informatization, geographic information systems (GIS) can be used to acquire geospatial and vegetation distribution images, quantify and aggregate topographic, climate, and soil data, extract vegetation features using deep learning, and match multi-source data to generate biodiversity trend data. However, the accuracy of geospatial and vegetation distribution images acquired through GIS is relatively low. Summary of the Invention

[0003] In view of this, embodiments of the present invention provide a method and system for integrating multi-source data of ecological space for forestry and grassland informatization, which enables the integrated data to accurately reflect the actual condition of forestry and grassland ecological space.

[0004] The specific technical solutions of this invention are as follows: The first aspect of this application provides a method and system for integrating multi-source ecological spatial data for forestry and grassland informatization, including: Acquire forestry and grassland business data, IoT sensing data, remote sensing images, and forest area geographic base maps of forest and grassland ecological space; the forest area geographic base map is a vector map including multiple small plot areas, each of which is a minimum management unit, associated with a unique plot code and geographic coordinates; Based on the forest area geographic base map, geographic coordinates and forest and grassland sub-compartment codes are assigned to forestry and grassland business data and IoT sensing data, and combined with time labels, a spatiotemporal fusion modeling dataset is constructed. Based on the spatiotemporal fusion modeling dataset, spatial coordinates are extracted from UAV aerial images and remote sensing images in IoT sensing data to obtain the first coordinate point set. Based on the first coordinate point set, the UAV aerial images and remote sensing images are stitched together to obtain a static geographic map of the forest area. Based on the spatiotemporal fusion modeling dataset, control points are determined from the geographic boundary information corresponding to the target forest and grassland sub-compartment codes on the static geographic map of the forest area. Keyframes are extracted from real-time UAV videos in the IoT sensing data. Image feature extraction algorithms are used to extract image feature points and their corresponding coordinate information from the keyframes. The image feature points and their coordinate information corresponding to the real-time UAV videos are determined as the second coordinate point set. Robust estimation algorithms are used to register the control points and the second coordinate point set to obtain the second transformation parameters. Based on the second transformation parameters, the images of each frame in the real-time UAV videos are mapped in real time and superimposed onto the second coordinate space corresponding to the static geographic base map of the forest area to obtain the dynamic geographic layer of the forest area. Based on the dynamic geographic layer of the forest area, the spatiotemporal fusion modeling dataset, forestry and grassland business data, and IoT sensing data, multi-source data of the forest area corresponding to the forestry and grassland ecological space are obtained.

[0005] In some embodiments, geographic coordinates include first geographic coordinates and second geographic coordinates; forest and grassland sub-compartment codes include first forest and grassland sub-compartment codes and second forest and grassland sub-compartment codes; Based on the forest area geographic base map, geographic coordinates and forest and grassland sub-compartment codes are assigned to forestry and grassland operational data and IoT sensing data. Combined with time labels, a spatiotemporal fusion modeling dataset is constructed, including: Spatial matching is performed between the location information in the forestry and grassland business data and the forest area geographic base map to assign the forestry and grassland business data the corresponding first geographic coordinates and the first forestry and grassland sub-compartment code. Business time information is obtained from the forestry and grassland business data and standardized to obtain the first timestamp. The spatiotemporal labels corresponding to the IoT sensing data are standardized to obtain the second geographic coordinates and the second timestamp. Based on the forest area geographic base map, the second geographic coordinates are associated with the corresponding second forest and grassland sub-compartment code. Using the first and second timestamps as time labels, and combining them with the first geographic coordinates, the second geographic coordinates, the first forest and grassland sub-compartment code, and the second forest and grassland sub-compartment code, a spatiotemporal fusion modeling dataset is constructed.

[0006] In some embodiments, based on a spatiotemporal fusion modeling dataset, spatial coordinates are extracted from UAV aerial images and remote sensing images in IoT sensing data to obtain a first set of coordinate points, including: Based on the geographical boundary information corresponding to the first and second forest and grassland sub-compartment codes, image feature points are extracted from UAV aerial images and remote sensing images, respectively. The image feature points extracted from the UAV aerial image are matched with the image feature points extracted from the remote sensing image to obtain matching point pairs, and a first set of coordinate points is generated based on the matching point pairs.

[0007] In some embodiments, based on a first set of coordinate points, aerial images taken by the UAV and remote sensing images are stitched together to obtain a static geographic map of the forest area, including: A robust estimation algorithm is used to filter matching point pairs in the first coordinate point set to obtain target matching point pairs. Based on the target matching point pairs, the first transformation parameters for geometrically correcting the UAV aerial image to the first coordinate space corresponding to the remote sensing image are determined. Based on the first transformation parameters, the UAV aerial images are geometrically corrected, and the geometrically corrected UAV aerial images are pixel-level stitched with remote sensing images to obtain a static geographic base map of the forest area.

[0008] In some embodiments, based on control points and a second set of coordinate points, real-time UAV video and a static geographic map of the forest area are fused to obtain a dynamic geographic layer of the forest area, including: The second set of coordinate points and control points are registered using a robust estimation algorithm to obtain the second transformation parameters; Based on the second transformation parameters, the images of each frame in the real-time video of the UAV are mapped in real time and superimposed onto the second coordinate space corresponding to the static geographic base map of the forest area to obtain the dynamic geographic layer of the forest area.

[0009] In some embodiments, before acquiring forestry and grassland operational data, IoT sensing data, remote sensing imagery, and forest area geographic base maps of forest and grassland ecological spaces, the method further includes: Acquire initial forestry and grassland business data and initial IoT sensing data; Based on preset forestry and grassland cleaning rules, the initial forestry and grassland business data and the initial IoT sensing data are cleaned separately to obtain forestry and grassland business data and IoT sensing data.

[0010] In some embodiments, the method further includes: Sensitive information in forestry and grassland business data and IoT sensing data is desensitized based on provincial-municipal-county three-level permission rules to generate permission label dataset; Based on the permission tag dataset, a data access interface matching the user's permissions is generated; the permission tag dataset is used to constrain and control the access scope and content of data services. The system receives query commands triggered by users through a data access interface; these commands include user permission tags and query information. In response to a query command, the system verifies the permission scope corresponding to the user's permission tag from the permission tag dataset. Based on the query information, it retrieves and displays the data within the permission scope from the multi-source data of the forest area.

[0011] In some embodiments, the method further includes: Based on the preset hierarchical storage rules, and the respective data formats of forestry and grassland business data, IoT sensing data, remote sensing images, and forest area geographic base maps, the forestry and grassland business data, IoT sensing data, remote sensing images, and forest area geographic base maps are stored.

[0012] A second aspect of this application provides an ecological spatial multi-source data integration system for forestry and grassland informatization, comprising: The acquisition module is used to acquire forestry and grassland business data, IoT sensing data, remote sensing images, and forest area geographic base maps of forest and grassland ecological spaces. The forest area geographic base map is a vector map that includes multiple small plot areas. Each small plot area is a minimum management unit, associated with a unique plot code and geographic coordinates. The module is used to assign geographic coordinates and forestry and grassland sub-compartment codes to forestry and grassland business data and IoT sensing data based on the forest area geographic base map, and to build a spatiotemporal fusion modeling dataset by combining time labels. The processing module is used to extract spatial coordinates from UAV aerial images and remote sensing images in IoT sensing data based on the spatiotemporal fusion modeling dataset, to obtain a first coordinate point set. Based on the first coordinate point set, the UAV aerial images and remote sensing images are stitched together to obtain a static geographic map of the forest area. Based on the spatiotemporal fusion modeling dataset, control points are determined from the geographic boundary information corresponding to the target forest and grassland sub-compartment codes on the static geographic map of the forest area. Keyframes are extracted from the real-time video of the UAV in the IoT sensing data. Image feature extraction algorithms are used to extract image feature points and their corresponding coordinate information from the keyframes. The image feature points and their coordinate information corresponding to the real-time video of the UAV are determined as a second coordinate point set. A robust estimation algorithm is used to register the control points and the second coordinate point set to obtain a second transformation parameter. Based on the second transformation parameter, each frame of the real-time video of the UAV is mapped in real time and superimposed onto the second coordinate space corresponding to the static geographic base map of the forest area to obtain a dynamic geographic layer of the forest area. The integration module is used to obtain multi-source data of forest areas corresponding to forest and grassland ecological spaces based on dynamic geographic layers of forest areas, spatiotemporal fusion modeling datasets, forestry and grassland business data, and IoT sensing data.

[0013] The beneficial effects of the technical solutions provided by the embodiments of the present invention include at least the following: In this embodiment of the invention, firstly, forestry and grassland operational data, IoT sensing data, remote sensing images, and a forest area geographic base map are acquired. Then, based on the forest area geographic base map, geographic coordinates and forestry and grassland sub-compartment codes are assigned to the forestry and grassland operational data and IoT sensing data, and combined with time tags, a spatiotemporal fusion modeling dataset is constructed. Next, based on the spatiotemporal fusion modeling dataset, spatial coordinates are extracted from UAV aerial images and remote sensing images in the IoT sensing data to obtain a first coordinate point set. Based on the first coordinate point set, the UAV aerial images and remote sensing images are stitched together to obtain a static geographic map of the forest area. Based on the spatiotemporal fusion modeling dataset, control points are determined from the static geographic map of the forest area, and control points are extracted from the IoT sensing data... A second set of coordinate points is acquired from real-time video of the UAV. Based on the control points and the second set of coordinate points, the real-time video of the UAV and the static geographic map of the forest area are fused to obtain a dynamic geographic layer of the forest area. Finally, based on the dynamic geographic layer of the forest area, the spatiotemporal fusion modeling dataset, forestry and grassland operational data, and IoT sensing data, multi-source data of the forest area corresponding to the forest and grassland ecological space are obtained. In this way, a dynamic geographic layer of the forest area is constructed through four types of heterogeneous data (forestry and grassland operational data, IoT sensing data, remote sensing imagery, and forest area geographic base map), thereby realizing the integration of multi-source data of the forest and grassland ecological space. This can improve the comprehensiveness of the integrated data (multi-source data of the forest area) and enable the multi-source data of the forest area to accurately reflect the actual situation of the forest and grassland ecological space. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort, wherein: Figure 1 A flowchart illustrating a method for integrating multi-source ecological spatial data for forestry and grassland informatization, provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of an ecological space multi-source data integration system for forestry and grassland informatization provided in an embodiment of the present invention. Detailed Implementation

[0015] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. The following embodiments are used to illustrate the present invention, but are not intended to limit the scope of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] In the following description, references are made to “some embodiments,” which describe a subset of all possible embodiments. However, it is understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0017] It should be noted that the terms "first, second, and third" used in the embodiments of the present invention are only used to distinguish similar objects and do not represent a specific ordering of objects. It is understood that "first, second, and third" can be interchanged in a specific order or sequence where permitted, so that the embodiments of the present invention described herein can be implemented in an order other than that illustrated or described herein.

[0018] It will be understood by those skilled in the art that, unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which these embodiments of the invention pertain. It should also be understood that terms such as those defined in general dictionaries should be understood to have the same meaning as in the context of the prior art and should not be interpreted in an idealized or overly formal sense unless specifically defined as herein.

[0019] Figure 1 This is a flowchart illustrating a method for integrating multi-source ecological spatial data for forestry and grassland informatization, provided in an embodiment of the present invention. The method can be executed via a control device, which includes at least one of a personal computer, laptop computer, smartphone, tablet computer, and portable wearable device; this embodiment does not limit the specific type of control device.

[0020] like Figure 1 As shown, the ecological space multi-source data integration method for forestry and grassland informatization provided in this embodiment of the invention may include steps S101-S105.

[0021] S101. Acquire forestry and grassland business data, IoT sensing data, remote sensing images, and forest area geographic base maps of forest and grassland ecological spaces.

[0022] In some embodiments, before acquiring forestry and grassland business data, IoT sensing data, remote sensing images, and forest area geographic base maps of the forest and grassland ecological space, the multi-source data integration method for ecological space oriented towards forestry and grassland informatization provided in this application embodiment further includes: acquiring initial forestry and grassland business data and initial IoT sensing data; and cleaning the initial forestry and grassland business data and initial IoT sensing data respectively based on preset forestry and grassland cleaning rules to obtain forestry and grassland business data and IoT sensing data.

[0023] In some embodiments, initial forestry and grassland business data may include forest vegetation coverage, forest pest and disease occurrence information, forest fire risk level, etc.; initial IoT sensing data may include drone aerial images, drone real-time aerial patrol videos (also known as drone real-time videos), infrared camera species images, forest fire monitoring videos, etc.

[0024] For example, the control device can obtain initial forestry and grassland business data from the business processing system of forestry administrative departments at all levels (such as provinces, cities, and counties), and obtain initial IoT sensing data through multiple IoT sensing devices (such as weather stations, drones, infrared cameras, etc.) set up in the forest area.

[0025] In some embodiments, forestry and grassland cleaning rules are used to process outliers, missing values, duplicate values, etc., in initial forestry and grassland business data and initial IoT sensing data. These forestry and grassland cleaning rules are pre-configured within the control device and can be configured according to actual needs. This application embodiment does not limit the specific content of the forestry and grassland cleaning rules.

[0026] For example, for initial forestry and grassland operational data, the control device can, based on preset forestry and grassland cleaning rules, fill missing values ​​in the initial forestry and grassland operational data with mean or median, remove outliers (values ​​exceeding preset thresholds), and deduplicate duplicate data records. By cleaning the initial data based on preset forestry and grassland cleaning rules, higher-quality forestry and grassland operational data and IoT sensing data that better meet operational needs can be obtained, providing a reliable foundation for subsequent data integration and analysis. For initial IoT sensing data, the control device can, based on preset forestry and grassland cleaning rules, fill missing frames in the initial IoT sensing data with interpolation, remove outlier frames (frames with severe distortion or signal loss) and repair them by estimating data from preceding and following frames, and deduplicate repeatedly collected data in the initial IoT sensing data.

[0027] In some embodiments, the remote sensing imagery is satellite imagery, which can be acquired through satellite remote sensing technology. The forest area geographic base map is a vector map comprising multiple small plots, each of which is a minimum management unit, associated with a unique plot code and geographic coordinates, and can be obtained through a Geographic Information System (GIS) platform.

[0028] S102. Based on the forest area geographic base map, assign geographic coordinates and forest and grassland sub-compartment codes to forestry and grassland business data and IoT sensing data, and combine them with time labels to construct a spatiotemporal fusion modeling dataset.

[0029] In some embodiments, the geographic coordinates are latitude and longitude coordinates; the forest and grassland sub-compartment code is the identification code corresponding to each sub-compartment area according to the division of the forest area.

[0030] In some embodiments, geographic coordinates include first geographic coordinates and second geographic coordinates; forest and grassland sub-compartment codes include first forest and grassland sub-compartment codes and second forest and grassland sub-compartment codes; based on the forest area geographic base map, geographic coordinates and forest and grassland sub-compartment codes are assigned to forest and grassland operational data and IoT sensing data, and combined with time labels, a spatiotemporal fusion modeling dataset is constructed by: spatially matching the location information in the forest and grassland operational data with the forest area geographic base map to assign the corresponding first geographic coordinates and first forest and grassland sub-compartment codes to the forest and grassland operational data, and obtaining operational time information from the forest and grassland operational data and standardizing it to obtain a first timestamp; standardizing the spatiotemporal labels corresponding to the IoT sensing data to obtain second geographic coordinates and a second timestamp, and associating the second geographic coordinates with the corresponding second forest and grassland sub-compartment codes based on the forest area geographic base map; using the first timestamp and the second timestamp as time labels, and combining the first geographic coordinates, the second geographic coordinates, the first forest and grassland sub-compartment codes, and the second forest and grassland sub-compartment codes to construct a spatiotemporal fusion modeling dataset.

[0031] For example, for forestry and grassland operational data, the control device can spatially match the location information (such as place names) in the data with the forest area geographic base map. Based on the spatial matching results, it determines the corresponding sub-compartment area from the base map and assigns the geographic coordinates and sub-compartment code associated with the sub-compartment area as the first geographic coordinates and the first forestry and grassland sub-compartment code to the operational data. Simultaneously, it extracts operational time information, such as approval date and effective date, from the data and processes it according to a preset time format, such as Coordinated Universal Time (UTC), to obtain a first timestamp. For IoT sensing data, the control device standardizes the spatiotemporal tags corresponding to the data. These tags include original coordinates and an original timestamp. The original coordinates can be format-converted and unified, such as converting them to WGS84 decimal coordinates to obtain second geographic coordinates. The original timestamp is also standardized according to a preset time format to obtain a second timestamp. After obtaining the second geographic coordinates, the corresponding sub-compartment area can be determined on the forest area geographic base map. The second geographic coordinates are then associated with the corresponding second forest and grassland sub-compartment code to ensure that the IoT sensing data can accurately correspond to the geospatial location on the forest area geographic base map. Subsequently, the first and second timestamps are used as time labels, combined with the first and second geographic coordinates, the first and second forest and grassland sub-compartment codes, to construct a spatiotemporal fusion modeling dataset. This dataset includes time labels, the first and second geographic coordinates, the first and second forest and grassland sub-compartment codes, and the second forest and grassland sub-compartment code.

[0032] S103. Based on the spatiotemporal fusion modeling dataset, spatial coordinates are extracted from UAV aerial images and remote sensing images in the IoT sensing data to obtain the first coordinate point set. Based on the first coordinate point set, the UAV aerial images and remote sensing images are stitched together to obtain a static geographic map of the forest area.

[0033] In some embodiments, based on a spatiotemporal fusion modeling dataset, spatial coordinate extraction is performed on UAV aerial images and remote sensing images in IoT sensing data to obtain a first coordinate point set. This includes: extracting image feature points from the UAV aerial images and remote sensing images respectively based on the geographic boundary information corresponding to the first and second forest and grassland sub-compartment codes; matching the image feature points extracted from the UAV aerial images with the image feature points extracted from the remote sensing images to obtain matching point pairs; and generating the first coordinate point set based on the matching point pairs.

[0034] For example, the control device can map the geographic boundary information corresponding to the first forest and grassland sub-compartment code onto the UAV aerial imagery to define the image feature extraction area in the UAV aerial imagery and extract image feature points within the image feature extraction area. Similarly, the control device can project the geographic boundary information corresponding to the second forest and grassland sub-compartment code onto the remote sensing imagery and extract image feature points. Here, image feature points represent at least one of the following in the UAV aerial imagery or remote sensing imagery: corner points, patchy area centers, and high-texture areas. In some application scenarios, image feature point detection algorithms, such as Scale-Invariant Feature Transform (SIFT) and Speeded-Up Robust Features (SURF), can be used for image feature point extraction. The control device can match image feature points extracted from UAV aerial images with image feature points extracted from remote sensing images. Specifically, it calculates the similarity (e.g., Euclidean distance) between the image feature points extracted from the UAV aerial images and those extracted from the remote sensing images, and identifies two image feature points with a similarity greater than or equal to a preset similarity threshold as matching point pairs. Each matching point pair contains one image feature point extracted from the UAV aerial image and one image feature point extracted from the remote sensing image. After obtaining the matching point pairs, the control device can organize and format them to generate a first set of coordinate points.

[0035] In some embodiments, stitching together UAV aerial images and remote sensing images based on a first set of coordinate points to obtain a static geographic map of the forest area includes: using a robust estimation algorithm to filter matching point pairs in the first set of coordinate points to obtain target matching point pairs, and based on the target matching point pairs, determining a first transformation parameter for geometrically correcting the UAV aerial images to the first coordinate space corresponding to the remote sensing images; performing geometric correction on the UAV aerial images based on the first transformation parameter, and stitching the geometrically corrected UAV aerial images and remote sensing images pixel-level to obtain a static geographic base map of the forest area.

[0036] For example, the image feature points extracted from UAV aerial images are the pixel coordinates of the UAV aerial images, while the image feature points extracted from remote sensing images are the geographic coordinates of the remote sensing images. Since there may be mismatched point pairs in the matching point pairs in the first coordinate point set, to improve the accuracy of subsequent integrated data, a robust estimation algorithm, such as the Random Sample Consensus (RANSAC) algorithm, can be used to filter the matching point pairs in the first coordinate point set to obtain target matching points. Based on the target matching point pairs, a mathematical optimization method (such as the least squares method) is used to calculate the first transformation parameter. The first transformation parameter is used to geometrically correct the UAV aerial images to the first coordinate space corresponding to the remote sensing images. After obtaining the first transformation parameter, the UAV aerial images can be geometrically corrected using the first transformation parameter, that is, the pixel coordinates on the UAV aerial images are transformed according to the first transformation parameter so that they are in the same coordinate space as the geographic coordinates on the remote sensing images. The geometrically corrected UAV aerial images are spatially aligned with the remote sensing images. At this point, the geometrically corrected UAV images can be fused with the reference remote sensing images to generate the final static base map.

[0037] In some application scenarios, when geometrically corrected UAV images are fused with reference remote sensing images, pixels with small differences in pixel values ​​in the overlapping areas can be selected for fusion to reduce stitching artifacts.

[0038] S104. Based on the spatiotemporal fusion modeling dataset, control points are determined from the geographic boundary information corresponding to the target forest and grassland sub-compartments on the static geographic map of the forest area. Keyframes are extracted from the real-time video of UAVs in the IoT sensing data. Image feature extraction algorithms are used to extract image feature points and their corresponding coordinate information from the keyframes. The image feature points and their coordinate information corresponding to the real-time video of the UAVs are determined as the second coordinate point set. Robust estimation algorithms are used to register the control points and the second coordinate point set to obtain the second transformation parameters. Based on the second transformation parameters, the images of each frame in the real-time video of the UAVs are mapped in real time and superimposed on the second coordinate space corresponding to the static geographic base map of the forest area to obtain the dynamic geographic layer of the forest area.

[0039] In some embodiments, the control device determines at least one relevant target forest and grassland sub-compartment code from the spatiotemporal fusion modeling dataset based on the location information corresponding to the UAV real-time video, and maps the geographic boundary information corresponding to the target forest and grassland sub-compartment code onto a static geographic map of the forest area to determine the corresponding control point (such as the corner point of the sub-compartment area) on the static geographic map of the forest area. Then, the control device extracts keyframes from the UAV real-time video and uses image feature extraction algorithms (such as SIFT and SURF algorithms) to extract image feature points and their corresponding coordinate information from the keyframes, determining the image feature points and their coordinate information corresponding to the UAV real-time video as a second set of coordinate points.

[0040] In some embodiments, determining control points from the geographic boundary information corresponding to the target forest and grassland sub-compartment code on a static geographic map of a forest area based on a spatiotemporal fusion modeling dataset includes: determining the target forest and grassland sub-compartment code from the spatiotemporal fusion modeling dataset based on the location information corresponding to the real-time video of a UAV; obtaining the geographic boundary information corresponding to the target forest and grassland sub-compartment code, and determining the sub-compartment boundary corner points as a first set of control points based on the geographic boundary information; obtaining high-precision surveying and mapping control points around the location corresponding to the real-time video of a UAV from the control point database as a second set of control points based on the location information corresponding to the real-time video of a UAV; and fusing the first set of control points and the second set of control points to obtain the control points.

[0041] In some embodiments, the control point database is a database pre-configured within the control device. The high-precision mapping control points in the control point database can be obtained through Real-Time Kinematic Global Positioning System (RTK-GPS) or total station surveying. The planar accuracy of each high-precision mapping control point can reach the centimeter to sub-meter level, which is higher than the accuracy of the corner points of the sub-block boundary (5-20 meters).

[0042] Understandably, the control device merges the first set of control points and the second set of control points to obtain control points, which can improve the accuracy and coverage of the control points.

[0043] In some embodiments, during the fusion process, the control device can perform weighted processing on the first control point set and the second control point set according to preset weighting rules. For example, the spatial deviation between the corner points of the small-scale boundary and the high-precision mapping control points can be optimized by interpolation algorithms or spatial smoothing algorithms, making the control point sets more geometrically consistent.

[0044] In some embodiments, fusing real-time UAV video and static geographic map of forest area based on control points and a second set of coordinate points to obtain dynamic geographic layer of forest area includes: registering the second set of coordinate points and control points using a robust estimation algorithm to obtain second transformation parameters; and mapping and superimposing each frame of image in real-time from the real-time UAV video onto the second coordinate space corresponding to the static geographic base map of forest area based on the second transformation parameters to obtain dynamic geographic layer of forest area.

[0045] For example, while the second set of coordinate points extracted from the real-time video of the drone can reflect the spatial trajectory of the drone during filming, it may contain errors. To improve the accuracy of the subsequently integrated data, a robust estimation algorithm, such as the random sample consensus algorithm, can be used to register the second set of coordinate points with control points, thereby selecting multiple interior points and their corresponding control points from the second set of coordinate points. Then, based on the multiple interior points and their corresponding control points, a mathematical optimization method (such as the least squares method) is used to calculate the second transformation parameters. These second transformation parameters define the spatial mapping relationship from the real-time drone video to the static geographic map of the forest area. After obtaining the second transformation parameters, the position parameters (such as coordinates, scale, etc.) of each frame of the UAV real-time video in the forest static geographic map are determined based on the second transformation parameters. Then, geometric correction and resampling are performed on each frame of the UAV real-time video according to the position parameters to obtain the corrected image. The corrected image is used as a semi-transparent dynamic overlay layer, which is rendered and superimposed on the forest static geographic base map in real time. This allows each frame of the UAV real-time video to be mapped and superimposed on the second coordinate space corresponding to the forest static geographic base map in real time, thereby obtaining the forest dynamic geographic layer.

[0046] S105. Based on the dynamic geographic layer of the forest area, the spatiotemporal fusion modeling dataset, forestry and grassland business data, and IoT sensing data, multi-source data of the forest area corresponding to the forestry and grassland ecological space are obtained.

[0047] For example, the dynamic geographic layer of the forest area, the spatiotemporal fusion modeling dataset, the forestry and grassland business data and the IoT sensing data can be integrated to obtain the multi-source data of the forest area corresponding to the forestry and grassland ecological space and store it in the memory of the control device.

[0048] It is understood that the embodiments of this application construct a dynamic geographic layer of forest area through four types of heterogeneous data (forestry and grassland business data, Internet of Things sensing data, remote sensing images and forest area geographic base map), thereby realizing the integration of multi-source data of forest and grassland ecological space, which can improve the comprehensiveness of integrated data (multi-source data of forest area) and enable the multi-source data of forest area to accurately reflect the actual situation of forest and grassland ecological space.

[0049] In some embodiments, the ecological space multi-source data integration method for forestry and grassland informatization provided in this application further includes: storing forestry and grassland business data, IoT sensing data, remote sensing images and forest area geographic base maps based on preset hierarchical storage rules, the respective data formats of forestry and grassland business data, IoT sensing data, remote sensing images and forest area geographic base maps.

[0050] In some embodiments, hierarchical storage rules are used to define how data of different data forms are stored in a categorized manner.

[0051] For example, forestry and grassland operational data is structured data, IoT sensing data is semi-structured and unstructured streaming data, remote sensing imagery is unstructured data, and forest area geographic base maps are multi-format data combining vector spatial data and attribute data. In some application scenarios, the storage hierarchy of the control device can include an operational data layer, a sensing data layer, and a remote sensing data layer. Forestry and grassland operational data can be stored in the operational data layer, IoT sensing data can be stored in the sensing data layer, remote sensing imagery can be stored in the remote sensing data layer, geometric data from the forest area geographic base map can be stored in the sensing data layer or the remote sensing data layer, and attribute data can be stored in the operational data layer. Among them, the operational data layer can adopt a columnar database; the sensing data layer can adopt an object storage + metadata table architecture, where video, image, and other data in IoT sensing data can be stored in object storage, and other data (such as acquisition time, path, etc.) can be stored in the metadata table; the remote sensing data layer can adopt a distributed file system, and a slicing algorithm can be used to slice remote sensing imagery into tile files and store them in the remote sensing data layer.

[0052] It is understood that the embodiments of this application use a hierarchical storage method to store forestry and grassland business data, IoT sensing data, remote sensing images and forest area geographic base maps, which can improve the efficiency of data management and the speed of data reading.

[0053] In some embodiments, the ecological space multi-source data integration method for forestry and grassland informatization provided in this application further includes: desensitizing sensitive information in forestry and grassland business data and IoT sensing data based on provincial-municipal-county three-level permission rules to generate a permission tag dataset; generating a data access interface matching user permissions based on the permission tag dataset; receiving query instructions triggered by users through the data access interface; and, in response to the query instructions, verifying the permission scope corresponding to the user permission tags from the permission tag dataset, and obtaining and displaying data within the permission scope from the forest area multi-source data based on the query information. The permission tag dataset is used to constrain and control the access scope and content of data services; the query instructions include user permission tags and query information.

[0054] In some embodiments, the provincial-municipal-county three-tier permission rules are used to define the query permission scope for forestry and grassland multi-source data corresponding to the provincial, municipal, and county levels, respectively. In practical applications, the provincial-municipal-county three-tier permission rules can be formulated according to actual needs. For example, provincial users can access forestry and grassland multi-source data throughout the province, including all sensitive information; municipal users can only access forestry and grassland multi-source data within their jurisdiction (district), and certain sensitive fields (such as precise coordinates and personal ID numbers) need to be anonymized; county users can only access forestry and grassland multi-source data within their county (district), and the anonymization scope is broader, such as only displaying the township level and hiding the beneficiary's name, retaining only the surname, etc.

[0055] For example, the control device performs anonymization processing on sensitive information (such as forest tenure holders, ID numbers, contact information, and latitude and longitude coordinates of rare and endangered plants and animals) in forestry and grassland business data and IoT sensing data based on provincial-municipal-county three-level permission rules. This anonymization can be achieved by masking, generalizing, or replacing sensitive information. Simultaneously, the control device can generate at least one permission tag for each sensitive field or each anonymized record. After the sensitive information is anonymized, all permission tags will constitute a permission tag dataset. The control device can then generate corresponding data access interfaces for users with different permission levels (such as provincial, municipal, and county-level users) based on the permission tag dataset. When a user triggers a query command through a client or other terminal device, the query command will carry the user's permission tag and query information, and will be sent to the control device through the data access interface. Upon receiving a query command, the control device will respond by verifying the permission scope corresponding to the user's permission tag to confirm whether the user has the right to access the requested data and the corresponding permission scope. After successful verification, the control device will filter the data within the permission scope from the multi-source data of the forest area based on the query information and display the data in an appropriate format (such as tables, charts, maps, etc.) for the user to view and analyze.

[0056] In some embodiments, a query interface may be displayed on the client or other terminal device. Users can enter their account and password to log in on the query interface, and after logging in, enter query information, such as "view the current video at XX location", and then click the query control to trigger the query command.

[0057] It is understood that the embodiments of this application can ensure that users with different permission levels can obtain the corresponding data according to their permission scope, thereby improving data security.

[0058] Based on the same inventive concept, this application also provides an ecological spatial multi-source data integration system for forestry and grassland informatization. The solution provided by this system is similar to the solution described in the above-mentioned method; therefore, please refer to the limitations of the ecological spatial multi-source data integration method for forestry and grassland informatization described above, and will not be repeated here. Specifically, Figure 2 This is a schematic diagram of the structure of an ecological spatial multi-source data integration system for forestry and grassland informatization, as described in an embodiment of this application. Figure 2 As shown, the ecological spatial multi-source data integration system for forestry and grassland informatization includes: The acquisition module 210 is used to acquire forestry and grassland business data, IoT sensing data, remote sensing images and forest area geographic base map of forest and grassland ecological space; the forest area geographic base map is a vector map including multiple small plot areas, each small plot area is a minimum management unit, associated with a unique plot code and geographic coordinates; Module 220 is used to assign geographic coordinates and forestry and grassland sub-compartment codes to forestry and grassland business data and IoT sensing data based on the forest area geographic base map, and to build a spatiotemporal fusion modeling dataset by combining time labels. The processing module 230 is used to extract spatial coordinates from UAV aerial images and remote sensing images in IoT sensing data based on the spatiotemporal fusion modeling dataset to obtain a first coordinate point set. Based on the first coordinate point set, the UAV aerial images and remote sensing images are stitched together to obtain a static geographic map of the forest area. Based on the spatiotemporal fusion modeling dataset, control points are determined from the geographic boundary information corresponding to the target forest and grassland sub-compartment codes on the static geographic map of the forest area. Keyframes are extracted from the real-time video of the UAV in IoT sensing data. Image feature extraction algorithms are used to extract image feature points and their corresponding coordinate information from the keyframes. The image feature points and their coordinate information corresponding to the real-time video of the UAV are determined as a second coordinate point set. A robust estimation algorithm is used to register the control points and the second coordinate point set to obtain a second transformation parameter. Based on the second transformation parameter, each frame of the real-time video of the UAV is mapped in real time and superimposed onto the second coordinate space corresponding to the static geographic base map of the forest area to obtain a dynamic geographic layer of the forest area. The integration module 240 is used to obtain multi-source data of forest areas corresponding to forest and grassland ecological spaces based on dynamic geographic layers of forest areas, spatiotemporal fusion modeling datasets, forestry and grassland business data and IoT sensing data.

[0059] In some embodiments, geographic coordinates include first geographic coordinates and second geographic coordinates; forest and grassland sub-compartment codes include first forest and grassland sub-compartment codes and second forest and grassland sub-compartment codes; The construction module 220 is also used to spatially match the location information in the forestry and grassland business data with the forest area geographic base map to assign the forestry and grassland business data the corresponding first geographic coordinates and the first forestry and grassland sub-compartment code, and to obtain the business time information from the forestry and grassland business data and perform standardization processing to obtain the first timestamp; to standardize the spatiotemporal labels corresponding to the IoT sensing data to obtain the second geographic coordinates and the second timestamp, and to associate the second geographic coordinates with the corresponding second forestry and grassland sub-compartment code based on the forest area geographic base map; and to construct a spatiotemporal fusion modeling dataset by using the first timestamp and the second timestamp as time labels, combined with the first geographic coordinates, the second geographic coordinates, the first forestry and grassland sub-compartment code and the second forestry and grassland sub-compartment code.

[0060] In some embodiments, the processing module 230 is further configured to extract image feature points from UAV aerial images and remote sensing images respectively based on the geographical boundary information corresponding to the first forest and grassland sub-compartment code and the second forest and grassland sub-compartment code; match the image feature points extracted from the UAV aerial images with the image feature points extracted from the remote sensing images to obtain matching point pairs, and generate a first set of coordinate points based on the matching point pairs.

[0061] In some embodiments, the processing module 230 is further configured to use a robust estimation algorithm to filter matching point pairs in the first coordinate point set to obtain target matching point pairs, and based on the target matching point pairs, determine the first transformation parameters for geometrically correcting the UAV aerial image to the first coordinate space corresponding to the remote sensing image; perform geometric correction on the UAV aerial image based on the first transformation parameters, and stitch the geometrically corrected UAV aerial image with the remote sensing image at the pixel level to obtain a static geographic base map of the forest area.

[0062] In some embodiments, the processing module 230 is further configured to: determine the target forest and grassland sub-compartment code from the spatiotemporal fusion modeling dataset based on the location information corresponding to the UAV real-time video; obtain the geographic boundary information corresponding to the target forest and grassland sub-compartment code, and determine the sub-compartment boundary corner point as the first control point set based on the geographic boundary information; obtain the high-precision surveying and mapping control points around the location corresponding to the UAV real-time video from the control point database as the second control point set based on the location information corresponding to the UAV real-time video; and fuse the first control point set and the second control point set to obtain the control point.

[0063] In some embodiments, the acquisition module 210 is further configured to acquire initial forestry and grassland business data and initial Internet of Things sensing data; The processing module 230 is also used to clean the initial forestry and grassland business data and the initial IoT sensing data based on preset forestry and grassland cleaning rules, so as to obtain forestry and grassland business data and IoT sensing data.

[0064] In some embodiments, the processing module 230 is further configured to perform desensitization processing on sensitive information in forestry and grassland business data and IoT sensing data based on provincial-municipal-county three-level permission rules, and generate a permission tag dataset; generate a data access interface matching user permissions based on the permission tag dataset; wherein, the permission tag dataset is used to constrain and control the access scope and content of data services; receive query instructions triggered by users through the data access interface; wherein, the query instructions include user permission tags and query information; in response to the query instructions, verify the permission scope corresponding to the user permission tags from the permission tag dataset, and obtain and display data within the permission scope from multi-source data of forest areas based on the query information.

[0065] In some embodiments, the processing module 230 is further configured to store forestry and grassland business data, IoT sensing data, remote sensing images and forest area geographic base maps based on preset hierarchical storage rules and the respective data formats of forestry and grassland business data, IoT sensing data, remote sensing images and forest area geographic base maps.

[0066] It should be understood that the phrase "one embodiment" or "an embodiment" throughout the specification means that a specific feature, structure, or characteristic related to the embodiment is included in at least one embodiment of the invention. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. Furthermore, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. It should be understood that in the various embodiments of the invention, the sequence numbers of the above-described processes do not imply a sequential order of execution; the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the invention. The sequence numbers of the above-described embodiments of the invention are merely descriptive and do not represent the superiority or inferiority of the embodiments.

[0067] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0068] In the several embodiments provided by this invention, it should be understood that the disclosed methods can be implemented in other ways. The methods disclosed in the several method embodiments provided by this invention can be arbitrarily combined without conflict to obtain new method embodiments. The features disclosed in the several method embodiments provided by this invention can be arbitrarily combined without conflict to obtain new method embodiments.

[0069] The above description is merely an embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for integrating multi-source ecological spatial data for forestry and grassland informatization, characterized in that, include: Acquire forestry and grassland business data, IoT sensing data, remote sensing images, and forest area geographic base maps of forest and grassland ecological spaces; The base map of the forest area is a vector map that includes multiple small compartments. Each small compartment is a minimum management unit, associated with a unique compartment code and geographic coordinates. Based on the forest area geographic base map, geographic coordinates and forest and grassland sub-compartment codes are assigned to forestry and grassland business data and IoT sensing data, and combined with time labels, a spatiotemporal fusion modeling dataset is constructed. Based on the spatiotemporal fusion modeling dataset, spatial coordinates are extracted from UAV aerial images and remote sensing images in IoT sensing data to obtain the first coordinate point set. Based on the first coordinate point set, the UAV aerial images and remote sensing images are stitched together to obtain a static geographic map of the forest area. Based on the spatiotemporal fusion modeling dataset, control points are determined from the geographic boundary information corresponding to the target forest and grassland sub-compartment codes on the static geographic map of the forest area. Keyframes are extracted from real-time UAV videos in the IoT sensing data. Image feature extraction algorithms are used to extract image feature points and their corresponding coordinate information from the keyframes. The image feature points and their coordinate information corresponding to the real-time UAV videos are determined as the second coordinate point set. Robust estimation algorithms are used to register the control points and the second coordinate point set to obtain the second transformation parameters. Based on the second transformation parameters, the images of each frame in the real-time UAV videos are mapped in real time and superimposed onto the second coordinate space corresponding to the static geographic base map of the forest area to obtain the dynamic geographic layer of the forest area. Based on the dynamic geographic layer of the forest area, the spatiotemporal fusion modeling dataset, forestry and grassland business data, and IoT sensing data, multi-source data of the forest area corresponding to the forestry and grassland ecological space are obtained.

2. The method according to claim 1, characterized in that, Geographic coordinates include the first geographic coordinate and the second geographic coordinate; forest and grassland sub-compartment codes include the first forest and grassland sub-compartment code and the second forest and grassland sub-compartment code; Based on the forest area geographic base map, geographic coordinates and forest and grassland sub-compartment codes are assigned to forestry and grassland operational data and IoT sensing data. Combined with time labels, a spatiotemporal fusion modeling dataset is constructed, including: Spatial matching is performed between the location information in the forestry and grassland business data and the forest area geographic base map to assign the forestry and grassland business data the corresponding first geographic coordinates and the first forestry and grassland sub-compartment code. Business time information is obtained from the forestry and grassland business data and standardized to obtain the first timestamp. The spatiotemporal labels corresponding to the IoT sensing data are standardized to obtain the second geographic coordinates and the second timestamp. Based on the forest area geographic base map, the second geographic coordinates are associated with the corresponding second forest and grassland sub-compartment code. Using the first and second timestamps as time labels, and combining them with the first geographic coordinates, the second geographic coordinates, the first forest and grassland sub-compartment code, and the second forest and grassland sub-compartment code, a spatiotemporal fusion modeling dataset is constructed.

3. The method according to claim 2, characterized in that, Based on the spatiotemporal fusion modeling dataset, spatial coordinates are extracted from UAV aerial images and remote sensing images in IoT sensing data to obtain the first set of coordinate points, including: Based on the geographical boundary information corresponding to the first and second forest and grassland sub-compartment codes, image feature points are extracted from UAV aerial images and remote sensing images, respectively. The image feature points extracted from the UAV aerial image are matched with the image feature points extracted from the remote sensing image to obtain matching point pairs, and a first set of coordinate points is generated based on the matching point pairs.

4. The method according to claim 3, characterized in that, Based on the first set of coordinate points, the drone aerial images and remote sensing images are stitched together to obtain a static geographic map of the forest area, including: A robust estimation algorithm is used to filter matching point pairs in the first coordinate point set to obtain target matching point pairs. Based on the target matching point pairs, the first transformation parameters for geometrically correcting the UAV aerial image to the first coordinate space corresponding to the remote sensing image are determined. Based on the first transformation parameters, the UAV aerial images are geometrically corrected, and the geometrically corrected UAV aerial images are pixel-level stitched with remote sensing images to obtain a static geographic base map of the forest area.

5. The method according to claim 1 or 2, characterized in that, Based on a spatiotemporal fusion modeling dataset, control points are determined from the static geographic map of the forest area based on the geographic boundary information corresponding to the target forest and grassland sub-compartment codes, including: Based on the location information corresponding to the real-time video of the UAV, the target forest and grassland sub-compartment code is determined from the spatiotemporal fusion modeling dataset; Obtain the geographic boundary information corresponding to the target forest and grassland sub-compartment code, and determine the sub-compartment boundary corner points as the first control point set based on the geographic boundary information; Based on the location information corresponding to the real-time video of the UAV, high-precision mapping control points around the location corresponding to the real-time video of the UAV are obtained from the control point database as the second control point set. By merging the first set of control points and the second set of control points, control points are obtained.

6. The method according to claim 1, characterized in that, Before acquiring forestry and grassland operational data, IoT sensing data, remote sensing imagery, and forest area geographic base maps of the forest and grassland ecological space, the method further includes: Acquire initial forestry and grassland business data and initial IoT sensing data; Based on preset forestry and grassland cleaning rules, the initial forestry and grassland business data and the initial IoT sensing data are cleaned separately to obtain forestry and grassland business data and IoT sensing data.

7. The method according to claim 1, characterized in that, The method further includes: Sensitive information in forestry and grassland business data and IoT sensing data is desensitized based on provincial-municipal-county three-level permission rules to generate permission label dataset; Based on the permission tag dataset, a data access interface matching the user's permissions is generated; the permission tag dataset is used to constrain and control the access scope and content of data services. The system receives query commands triggered by users through a data access interface; these commands include user permission tags and query information. In response to a query command, the system verifies the permission scope corresponding to the user's permission tag from the permission tag dataset. Based on the query information, it retrieves and displays the data within the permission scope from the multi-source data of the forest area.

8. The method according to claim 1, characterized in that, The method further includes: Based on the preset hierarchical storage rules, and the respective data formats of forestry and grassland business data, IoT sensing data, remote sensing images, and forest area geographic base maps, the forestry and grassland business data, IoT sensing data, remote sensing images, and forest area geographic base maps are stored.

9. A multi-source ecological spatial data integration system for forestry and grassland informatization, characterized in that, include: The acquisition module is used to acquire forestry and grassland business data, IoT sensing data, remote sensing images, and forest area geographic base maps of forest and grassland ecological spaces. The base map of the forest area is a vector map that includes multiple small compartments. Each small compartment is a minimum management unit, associated with a unique compartment code and geographic coordinates. The module is used to assign geographic coordinates and forestry and grassland sub-compartment codes to forestry and grassland business data and IoT sensing data based on the forest area geographic base map, and to build a spatiotemporal fusion modeling dataset by combining time labels. The processing module is used to extract spatial coordinates from UAV aerial images and remote sensing images in IoT sensing data based on the spatiotemporal fusion modeling dataset, obtain a first set of coordinate points, and stitch the UAV aerial images and remote sensing images based on the first set of coordinate points to obtain a static geographic map of the forest area. Based on the spatiotemporal fusion modeling dataset, control points are determined from the geographic boundary information corresponding to the target forest and grassland sub-compartment codes on the static geographic map of the forest area. Keyframes are extracted from real-time UAV videos in the IoT sensing data. Image feature extraction algorithms are used to extract image feature points and their corresponding coordinate information from the keyframes. The image feature points and their coordinate information corresponding to the real-time UAV videos are determined as the second coordinate point set. Robust estimation algorithms are used to register the control points and the second coordinate point set to obtain the second transformation parameters. Based on the second transformation parameters, the images of each frame in the real-time UAV videos are mapped in real time and superimposed onto the second coordinate space corresponding to the static geographic base map of the forest area to obtain the dynamic geographic layer of the forest area. The integration module is used to obtain multi-source data of forest areas corresponding to forest and grassland ecological spaces based on dynamic geographic layers of forest areas, spatiotemporal fusion modeling datasets, forestry and grassland business data, and IoT sensing data.

Images