Data fusion method of forestry data management system

Abstract

The invention relates to the field of data fusion processing, and particularly discloses a data fusion method of a forestry data management system, which is used for processing real-time acquired data of the forestry data management system. Dividing the forest resource data of the target area into different independent areas, and establishing a special forest resource growth quantitative model for the forest resource data of each independent area; detecting whether a forest resource data conflict exists or not; when the forest resource data of a certain independent region conflicts, the conflict data is corrected through the forest resource growth quantification model of the independent region; when the forest resource data of a plurality of independent regions conflicts, the conflict data is corrected through the combination of the forest resource growth quantification models of the plurality of independent regions; and correcting the conflict data and then completing data fusion.

Description

Technical Field

[0001] The present application relates to the field of data fusion processing, and particularly to a data fusion method for a forestry data management system. Background Art

[0002] Forestry data has significant spatio-temporal characteristics. In terms of time, the growth and change of forest resources is a dynamic process, and in terms of space, there are differences in forest ecosystems at different geographical locations. During the spatio-temporal data fusion process, it is difficult to accurately match and fuse data at different time and space scales. There are many conflicting data. For example, when combining satellite images from different periods with ground data to analyze the dynamic changes of forest resources, it is difficult to accurately correspond the data at the same spatial position at different time points for unification. Summary of the Invention

[0003] In view of the deficiencies of the prior art, the present application provides a data fusion method for a forestry data management system, effectively solving the existing problems.

[0004] To achieve the above object, the present application is realized through the following technical solutions: A data fusion method for a forestry data management system of the present application includes the following steps: Perform unified data format, data quality purification, spatial coordinate normalization, and time series standardization processing on the real-time collected data of the forestry data management system; Divide the forest resource data of the target area into different independent areas, and establish a special forest resource growth quantification model for the forest resource data of each independent area; The forestry data management system detects whether there is a conflict in forest resource data before data fusion; the conflict in forest resource data refers to inconsistent growth indicators of forest resources in the same area collected at different time points; When there is a conflict in the forest resource data of a certain independent area, correct the conflicting data through the forest resource growth quantification model of this independent area; When there is a conflict in the forest resource data of a certain number of independent areas, jointly correct the conflicting data through the forest resource growth quantification models of these independent areas; After correcting the conflicting data, complete the data fusion.

[0005] On the other hand, the present application also discloses an electronic device, including: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the data fusion method of the forestry data management system described above.

[0006] The present application provides a data fusion method for a forestry data management system, which has at least the following significant beneficial effects: Conflict detection is performed before data fusion to discover the problem that the growth indicators of forest resources in the same area collected at different time points are inconsistent, avoiding inaccurate analysis results caused by data conflicts. For data conflicts in a single independent area and multiple independent areas, corresponding correction methods are respectively adopted, and a forest resource growth quantification model is used to correct the data, improving the accuracy and reliability of the data and ensuring the overall consistency and rationality of multi-area data. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Figure 1 It is a flowchart of the data fusion method of the forestry data management system of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0008] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0009] The present application discloses a data fusion method for a forestry data management system, as Figure 1 , the method includes the steps of: S1. performing unified data format, data quality purification, spatial coordinate normalization, and time series standardization processing on the real-time collected data of the forestry data management system; S2. dividing the forest resource data of the target area into different independent areas; S3. establishing a dedicated forest resource growth quantification model for the forest resource data of each independent area; S4. the forestry data management system detecting whether there is a conflict in the forest resource data before data fusion, where the forest resource data conflict refers to the inconsistent growth indicators of the forest resources in the same area collected at different time points; S5. when there is a conflict in the forest resource data of a certain independent area, correcting the conflict data through the forest resource growth quantification model of the independent area; S6. when there is a conflict in the forest resource data of a certain multiple independent areas, jointly correcting the conflict data through the forest resource growth quantification models of the multiple independent areas; S7. completing data fusion after correcting the conflict data.

[0010] Specifically, for S1, the real-time collected data of the forestry data management system is processed in terms of data format unification, data quality purification, spatial coordinate normalization, and time series standardization. The data format unification of the real-time collected data of the forestry data management system specifically includes spatial data specification and attribute data management. In the spatial data specification, for vector data, whether it is forest resource monitoring points or forest area plot boundary data, it is stored in the GeoJSON format. This format can accurately describe various geometric types such as points, lines, and surfaces, and has good readability and interoperability. For raster data, satellite remote sensing images, etc. are unified into the 16-bit TIFF format, and at the same time, the WGS84 geographic coordinate system and UTM projection are embedded.

[0011] Exemplarily, the projection zone is automatically divided according to the longitude and latitude of different regions. For example, the area between 30°N and 40°N uses the UTM50N zone to ensure accurate spatial positioning and display of spatial data under different sources and uses. The structured data in the attribute data management, such as the key information of tree diameter at breast height, tree height, and growth increment of forest trees, is stored in the PostgreSQL database. Using its structured data management ability, it is convenient to query, statistically analyze, and analyze the data. For unstructured data, such as the metadata of remote sensing images, it is managed through the HDFS distributed file system. During the attribute data management process, the "Forestry Data Element Standard" (LY / T 1662-2006) is strictly followed to ensure the unity of the units and coding rules of data fields.

[0012] The data quality purification of the real-time collected data of the forestry data management system is specifically carried out from several aspects such as data integrity check, consistency processing, and outlier and noise data processing to ensure that the data quality meets the requirements of subsequent analysis and fusion.

[0013] Specifically, for the data integrity check, traverse the structured attribute data to check whether there are missing values in the key fields. For missing data, if it is ground monitoring data, the manual data entry method is preferred, and it is filled by on-site verification or by calling data in historical similar regions; if the amount of missing data is large, estimation is carried out for data with time series characteristics; for unstructured data, check whether the metadata information is complete, such as the key metadata of the acquisition time, resolution, and bands of remote sensing images. If there are missing values, obtain supplements from the data source or relevant metadata repositories. For the spatial data integrity verification, for vector data, check whether the geometric elements of forest resource monitoring points, forest area plot boundary data, etc. are complete, such as whether there are broken lines, fragmented surfaces, etc. Conduct topological verification on the vector data and repair the discovered topological errors. For raster data, check whether there are data loss or damage areas in the image. By calculating the checksum of the image and comparing it with the original storage record, if inconsistency is found, obtain the corresponding area data from the backup data source again.

[0014] Specifically, for data consistency processing, it includes the adjustment of attribute data consistency. During the management of attribute data, the attribute fields of different data sources are unified and standardized. For cases where there are multiple expressions for the same attribute, such as "tree species" and "tree species name", they are unified and standardized as "tree species", and standard coding is carried out. For numerical attribute data, it is checked whether the units of different data sources are unified. For example, data on "forest growth" with different units are uniformly converted to "cubic meters per hectare". When there are multiple numerical records for the same attribute (such as the tree breast diameters recorded by different monitoring devices in the same plot), weighted averaging is performed according to the data credibility weights. The credibility weights are determined based on the accuracy and collection frequency of the data collection devices. For example, the weight of data from high-precision sensors is 0.8, and the weight of manual measurement data is 0.5, to generate a unique and accurate attribute value.

[0015] Specifically, it also includes the integration of spatial data consistency. For vector data, through spatial topological relationships (such as intersection, inclusion), the forest resource boundaries and monitoring points from different sources are accurately matched. Overlap analysis is carried out on the forest farm zoning boundaries and the forest boundaries interpreted by satellite remote sensing, and the deviated boundary lines are corrected so that the spatial error between the two does not exceed one grid unit.

[0016] For raster data, based on the UTM projection zone, seamless mosaic technology is used for splicing. During the mosaic process, the brightness and color differences of adjacent images are adjusted through histogram matching to ensure that the tone of the fused image is consistent. At the same time, resampling technology is used to unify images with different resolutions to a 10-meter resolution to ensure the consistency of spatial details.

[0017] Specifically, in the processing of outliers and noise data, the median absolute deviation method is used to replace outliers to reduce their impact on the data analysis results. For high-frequency data collected by ground sensors (such as minute-level data), methods such as moving average filtering and median filtering are used to remove random noise in the data. Taking the data of the "soil moisture" sensor as an example, by setting an appropriate moving window size (such as 15 minutes), the average value or median value of the data within the window is calculated as the filtered data value. For salt-and-pepper noise and other noises in remote sensing image data, algorithms such as median filtering and mean filtering are used to remove them, while retaining the details of the image and improving the quality of the image data.

[0018] Specifically, perform spatial coordinate normalization on the real-time collected data of the forestry data management system, unify all forestry data to the national geodetic coordinate system, use the Gauss-Krüger projection (3° zone) for the plane coordinates, and the elevation datum is the 1985 National Elevation Datum. For historical data, use the georegistration tool for correction. Through precise control point matching and geometric transformation, ensure that the spatial error between the ground monitoring points and the remote sensing pixels is controlled within 10 meters, that is, not exceeding the size of 1 pixel with a resolution of 10 meters, so as to achieve precise spatial alignment of data from different sources. Divide the study area into regular grids of 100m×100m, and assign a unique ID to each grid. Through spatial overlay analysis, associate data with different resolutions, such as remote sensing images with a resolution of 30m and ground monitoring point data with a resolution of 5m, to the corresponding grids. In this way, when performing data processing and analysis, "grid-level" spatial alignment can be achieved based on the grids, which is convenient for quickly retrieving and processing multi-source data in the same spatial area.

[0019] Perform time series standardization processing on the real-time collected data of the forestry data management system. Specifically, uniformly adopt UTC time for the timestamps of all data, accurate to the second. For satellite remote sensing data (time resolution of 5 - 16 days), ground sensor data (minute level), and forest resource inventory data (annual), uniformly convert them to the daily scale (YYYY-MM-DD) to ensure the consistency of data with different time frequencies in the time dimension, which is convenient for time series analysis and data fusion. For high-frequency ground sensor data, perform aggregation downsampling by calculating the mean value to convert minute-level data to daily-scale data. For example, generate daily average values from data collected every 15 minutes. For low-frequency forest resource inventory data, use the cubic spline interpolation method to expand the annual data into a continuous time series, ensuring that the time axis error does not exceed 24 hours to meet the subsequent requirements for forest resource growth model analysis and dynamic monitoring based on time series.

[0020] Specifically, for S2, divide the forest resource data of the target area into different independent regions; in the forestry data management system, dividing the forest resource data of the target area into different independent regions is the basis for achieving precise data management and conflict handling. Specifically, integrate spatial data such as remote sensing images, terrain data (DEM), soil type maps, and vegetation distribution maps, and use the overlay analysis function of the Geographic Information System (GIS) to identify regions with similar characteristics. For example, overlay the vegetation type map with the soil type map to find the distribution areas of specific vegetation and soil combinations. Combine vector data such as administrative division boundaries and forest farm boundaries to adjust the initially divided regions to make the division results more in line with actual management needs.

[0021] Specifically, the division method includes the assistance of clustering algorithms, that is, based on the spatial distribution characteristics of forest resources (such as tree density, average breast height diameter, tree height) and ecological environment factors (altitude, slope, aspect), the target area is automatically divided into several clustering clusters. The clustering results are manually verified and optimized to ensure the homogeneity of forest resources within each independent area and clear and distinguishable boundaries.

[0022] The division method also includes grid division optimization, that is, based on the constructed 100m×100m regular grid, adjacent grids with similar attributes are merged into independent areas. For example, multiple consecutive grids dominated by pine trees and with the same soil type are merged into an independent pine tree planting area. For grids with small areas or special attributes, according to their similarity to the surrounding areas, they are incorporated into adjacent independent areas as close as possible to avoid the appearance of overly small or isolated division units.

[0023] After dividing the independent areas, a unique identification code (such as "RZ-001") is assigned to each independent area, and key attribute information is marked, including the area name (such as "Coniferous Forest Area of XX Mountains"), the administrative division to which it belongs, the main vegetation type, soil type, average altitude, etc., to form an area attribute file. Then, forest resource statistical indicators of each area are calculated, such as total area, total standing volume of trees, average canopy density, etc., to provide basic data for subsequent growth model establishment and data conflict analysis.

[0024] Associate the forest resource data (including spatial data, attribute data, monitoring data, etc.) within each independent area with the identification code of that area to ensure the integrity and traceability of the data.

[0025] Specifically, in S3, for the forest resource data of each independent area, a dedicated forest resource growth quantification model is established. The forest resource growth quantification model is specifically as follows: ; y(t) is the forest resource growth index, and y(t) specifically represents breast height diameter or tree height or biomass; K is the forest resource growth limit value, indicating the maximum value that the forest resource growth index y(t) (such as breast height diameter or tree height or biomass) can reach under the current site conditions, reflecting the limitation of the environment on forest growth.

[0026] r is the intrinsic growth rate, indicating the inherent growth rate of the forest resource growth index under the condition of no environmental limitation, reflecting the growth potential of the tree species itself.

[0027] Calculate the average growth rate of the forest resource growth index in different time periods. Through multiple fittings and adjustments, determine the r value that can most accurately describe the growth trend of this tree species. For example, through fitting calculation, the intrinsic growth rate r of poplar trees in a certain area is 0.2 per year, which means that under ideal conditions, the growth index of poplar trees increases by 20% per year.

[0028] t 0 is the growth inflection point time, which is the time point when the growth rate of forest resources reaches the maximum value. Before t 0 , the growth rate gradually increases; after t 0 , the growth rate gradually slows down.

[0029] Preferably, the site factors (elevation, slope, aspect, soil type, soil fertility, etc.) are further introduced to correct the parameters of the forest resources growth quantification model: establish the regression relationship between the site factors and the parameters of the forest resources growth quantification model. For example, as the elevation increases, the growth limit value K of a certain tree species will decrease, and the higher the soil organic matter content, the larger the K value. Through a large amount of data statistics and analysis, the specific regression equation parameters are determined. For the intrinsic growth rate r, the slope and aspect will affect the conditions for trees to obtain light and water, and thus affect the growth rate. Through the comparative analysis of the data of plots with different slopes and aspects, the correction formula of r with the slope and aspect is established, so that the model can more accurately reflect the growth of forest resources under different site conditions.

[0030] Specifically, for the correction of the growth limit value K: taking K as the dependent variable, elevation (x 1 ), soil fertility index (x 2 ), vegetation type (x 3 , dummy variable) as the independent variables, construct a linear regression equation, and determine the correction relationship of the growth limit value K by fitting the parameters of the linear regression equation through the least squares method. For example, K (the maximum value that the growth tree height of forest resources can reach) = 60 - 0.04x 1 + 15x 2 + x 3 , (x 1 unit is meter, x 2 is the fertility index, x 3 takes a fixed value), which means that for every 100m increase in elevation, the K value decreases by 4%; for every 0.1 increase in the fertility index, the K value increases by 1.5%.

[0031] Specifically, for the correction of the intrinsic growth rate r: for example, for the influence of slope (y 1 ) and aspect (y 2 , sunny slope = 1, shady slope = 0), construct a non-linear correction model: r = r 0 * (1 - 0.015y 1 ) * (1 + 0.1y 2 ), where r 0 is the basic growth rate. After being calibrated by the plot data, for every 10° increase in slope, the r value decreases by 15%; the r value in the sunny slope area is 10% higher than that in the shady slope.

[0032] Specifically, for the forestry data management system, it detects whether there are conflicts in forest resource data before data fusion. Such conflicts refer to inconsistent growth indicators of forest resources in the same area collected at different time points. For the same monitoring point, if the growth rate increase between adjacent years exceeds 1.5 times the maximum growth rate predicted by the forest resource growth quantification model, it is determined that there are data conflicts in the time dimension. For example, if the model predicts that the maximum annual growth rate of a certain tree species, such as poplar, is 5 cm, when the actual growth rate increase between adjacent years exceeds 7.5 cm, the conflict detection is triggered. Although tree growth is affected by multiple factors, there will be no drastic abnormal growth in a short period. Data exceeding this threshold is very likely to have problems.

[0033] Specifically, based on a pre-constructed regular grid of 100m×100m, each grid is assigned a unique GridID. Data detection will take the grid as the basic unit and conduct centralized analysis of the forest resource data within the grid.

[0034] Specifically, when conducting conflict detection, the system sets the current grid and its ±1 adjacent grids as the spatial detection range, which can comprehensively consider the influence of adjacent areas and avoid misjudgment caused by boundary errors or local anomalies. In the time dimension, with the current time point as the center, it extends 30 days forward and backward as the time detection window. This is because the growth changes of forest resources are relatively continuous in a short period (30 days). If there are data inconsistencies within this time window, it is more likely to be a data conflict rather than a real growth mutation.

[0035] For the selected data, the system compares and calculates the actual measured forest resource growth indicators (such as diameter at breast height, tree height, biomass, etc.) with the predicted values based on the forest resource growth quantification model. During the calculation process, it fully considers the influence of model parameters (such as the growth limit value K, the intrinsic growth rate r, the growth inflection point time t 0 ), as well as site factors (altitude, slope, soil type, etc.) on the predicted values, so as to obtain the deviation between the measured value and the predicted value.

[0036] Compare the calculated deviation with the pre-set detection standard. If the deviation exceeds the standard range, it is determined that there is a conflict in this data, and the conflict data is marked.

[0037] Specifically, in S5, when there are conflicts in the forest resource data of a certain independent region, the conflict data is corrected through the forest resource growth quantification model of this independent region; after detecting data conflicts, the system first determines the specific types of the conflict data, such as tree diameter at breast height, tree height, and biomass growth index data. At the same time, based on the previous grid division and detection area setting, the grid range where the conflict data is located is accurately positioned, its spatial position within the independent region is clarified, the parameters of the forest resource growth quantification model are called, and the conflict data correction operation is executed.

[0038] If the conflict data are growth indicators such as tree diameter at breast height, tree height, and biomass, directly substitute the t value at this moment, as well as the model parameters K, r, and t 0 into the forest resource growth quantification model to calculate the predicted value. For example, the measured annual increase in the tree diameter at breast height of a 20-year-old pine tree is 6 cm, exceeding the upper limit of the model prediction by 5 cm. Substitute t = 20, as well as K = 50, r = 0.15, and t 0 = 15 into the formula, and we can get y(20) = 45.5 cm. The tree diameter at breast height in the previous year was 40 cm, and it is predicted to increase by 5.5 cm this year. Use this predicted value to replace the measured conflict value to complete the data correction.

[0039] Specifically, in S6, when there are conflicts in the forest resource data of multiple independent regions, the conflict data is corrected jointly through the forest resource growth quantification models of these multiple independent regions; when it is detected that there are conflicts in the forest resource data of multiple independent regions, the system constructs a joint correction by integrating the forest resource growth quantification models of multiple regions. The specific implementation steps are as follows: Collect all the parameters of the forest resource growth quantification models involved in the conflict areas, including the growth limit value K (representing the maximum value that forest resource growth indicators such as tree diameter at breast height, tree height, or biomass can reach under the current site conditions, reflecting the environmental restrictions on forest growth), the intrinsic growth rate r (indicating the inherent growth rate of forest resource growth indicators under the condition of no environmental restrictions, reflecting the growth potential of tree species), and the growth inflection point time t 0 (i.e., the time point when the forest resource growth rate reaches the maximum value. Before t 0 the growth rate gradually increases, and then gradually slows down). At the same time, summarize the site factor parameters of each region, such as altitude, slope, aspect, soil type, soil fertility, etc. These site factors will correct the model parameters through regression relationships to adapt to the growth environments of different regions.

[0040] Weights are set for each region participating in the joint correction based on factors such as the spatial distance between the region and the region where the conflict data is located, the similarity of the ecological environment, and the data reliability. For example, regions adjacent to the conflict region and with similar ecological environments have higher weights; regions with high data collection frequencies and high-precision monitoring equipment also have correspondingly increased weights. The weight value range is 0 - 1, and the sum of the weights of all regions is 1, which is used to balance the influence of each region's model during subsequent joint calculations.

[0041] Based on the 100m×100m regular grid and the regional division results already constructed in the system, the specific grid where the conflict data is located and the multiple independent regions to which it belongs are determined. Combining the spatial detection range set during conflict detection (the current grid and its ±1 adjacent grids), the boundaries of the affected regions and the data scope are clarified to ensure comprehensive coverage of forest resource data that may be related.

[0042] For each independent region participating in the joint correction, substitute the time t corresponding to the conflict data into its forest resource growth quantification model to calculate their respective predicted values. Then, according to the pre-set regional weights, the predicted values of each region are weighted and summed to obtain the joint predicted value. This joint predicted value comprehensively considers the growth laws and site conditions of multiple regions and can better reflect the true growth trend of the region where the conflict data is located.

[0043] Take the joint predicted value as the correction result of the conflict data to replace the original conflict data. For example, if the conflict data is the abnormal annual growth value of the measured breast diameter of a cross-regional forest, and the predicted growth value obtained through joint calculation is a more reasonable value, then use this predicted value to update the original data.

[0044] In addition, preferably: Due to the spatial correlation of data in multiple regions, after the conflict data is corrected, other regional data associated with the data is automatically checked and corrected. If it is found that other regional data may have new conflicts or logical contradictions due to this correction (such as unreasonable differences in tree growth amounts in adjacent regions), then local secondary detection and adjustment are initiated to ensure the overall consistency and rationality of multi-regional data. Specifically: Set the difference threshold for growth indicators in adjacent regions. When the difference in the same type of growth indicators (such as breast diameter, tree height) between the corrected region and the adjacent independent region exceeds 15% of the predicted value of the region's model, detection is triggered.

[0045] For example, after correction, the annual increase in DBH of a certain grid in area A is 5.5 cm, and the predicted value of the same type of tree model in the adjacent area B is 4.5 cm. If the absolute value of the difference > 0.675 cm (4.5 cm × 15%), it is determined that there is a spatial association conflict. For adjacent grids with similar site factors (altitude difference < 50 m, slope difference < 10°, soil fertility index difference < 10%), if the difference in measured growth exceeds 20% of the corrected value of the intrinsic growth rate (for example, the r values of both areas are 0.15, and the allowable difference threshold is 0.03), it is marked as a potential logical contradiction. Centered on the grid where the conflict data is located, it automatically expands to ±2 adjacent grids (forming a 3×3 grid detection area) to cover the range that may be affected by the spatial conduction effect. If the conflict area crosses the boundary of an independent area (such as a forest farm boundary, a vegetation type dividing line), all related grids in the adjacent area are synchronously included to ensure the detection of data consistency across the area boundary.

[0046] Specifically, the secondary detection and adjustment include: Using GIS overlay analysis, compare the forest resource boundaries, vegetation type distributions, and site factor gradient changes between the conflict grid and adjacent grids. It is required that the changes in growth indicators across the area boundary conform to the ecological transition law (for example, for every 100 m increase in altitude, the predicted value of DBH growth should decrease by 3% - 5%). When the mutation amplitude exceeds 2 times this law, an alarm is triggered. Extract the growth data of the affected area in the same period for the past 3 years, calculate the standard deviation (σ) of the annual growth rate. If the corrected data causes the current growth rate fluctuation to exceed the historical mean ±2σ (for example, the historical standard deviation is 0.5 cm, and the current fluctuation > 1 cm), it is determined to be abnormal in the time dimension.

[0047] Retrieve the site factor data (altitude, slope, soil fertility, etc.) of the conflict area and adjacent areas, and inversely deduce the model parameters through the established regression relationship. For example, if the corrected K value (growth limit value) in area A is 50 cm, and the K value of the same tree species in the adjacent area B is 40 cm, but the soil fertility difference between the two places is only 5% and the altitude difference < 30 m, an alarm for abnormal K value is triggered, indicating that there may be a conflict in model parameters.

[0048] It can also quantify the influence weight of site factors on growth indicators (for example, the influence weight of altitude on the K value is 0.4, and that of soil fertility is 0.3). If the corrected data causes a contradiction in the weight contribution (such as the K value in the low-altitude area is lower than that in the high-altitude area), it is determined to be a model fitting deviation.

[0049] For adjacent areas where conflicts are detected, merge the corrected data with the historical monitoring data and refit the parameters of the forest resource growth quantification model. For example, if the deviation of the r value in area B is > 10% due to the correction in area A, extract the data of the same tree species in the two areas in the past 5 years and recalculate the r value through the non-linear least squares method to make the cross-regional model parameters conform to the gradient change of site factors (for example, when the slope increases by 10°, the decline range of the r value should be within 10% - 20%). Adjust the regional weight distribution of the joint correction model, increase the weight of adjacent areas to 0.6 - 0.8 (original weight ≤ 0.5), and reduce the influence of distant areas to ensure that the correction results conform to local ecological consistency.

[0050] For the grids across the regional boundary, use the inverse distance weighted interpolation (IDW) algorithm to generate transition values. The weight factor combines the spatial distance (< 50m weight 0.7, 50 - 100m weight 0.3) and the ecological similarity (same vegetation type weight 0.2, same soil type weight 0.1). For example, after the correction in area A, the diameter at breast height is 45cm, and the uncorrected value in the adjacent area B is 40cm. After interpolation, the boundary grid takes 43cm (45×0.7 + 40×0.3 when the distance is 50m). For the time series mutation caused by the corrected data, use cubic spline interpolation to supplement the intermediate values to ensure the continuity of the growth curve. For example, after the correction in area A in 2023, the growth is 5cm, and in area B in the same period, the growth without correction is 4cm. The boundary grid generates 4.5cm through interpolation to control the annual growth rate fluctuation within ±10% of the historical average.

[0051] S7. After correcting the conflict data, complete the data fusion; specifically, the data fusion levels include attribute data fusion, spatial data fusion, and time series fusion.

[0052] Attribute data fusion includes standardizing the mapping of attribute fields from different data sources according to the "Forestry Data Element Standard" (LY / T 1662 - 2006). For example, fields such as "tree variety" and "tree species name" are unified and standardized as "tree species", and encoded using the GB / T 14396 - 2008 standard to ensure the consistency of semantics and encoding for the same attribute.

[0053] For multiple numerical records of the same attribute (such as the diameter at breast height of forest trees recorded by different monitoring devices in the same plot), perform weighted averaging according to the data credibility weights. The credibility weights are determined according to the accuracy and collection frequency of the data collection device. For example, the weight of data from high-precision sensors is 0.8, and the weight of manual measurement data is 0.5, so as to eliminate data differences and generate a unique and accurate attribute value.

[0054] Spatial data fusion includes, for vector data, precisely matching forest resource boundaries and monitoring points from different sources through spatial topological relationships (such as intersection, inclusion). For example, overlapping the forest farm zoning boundary with the forest boundary interpreted from satellite remote sensing to correct the deviated boundary line so that the spatial error between the two does not exceed one grid cell (100m×100m).

[0055] For raster data such as satellite remote sensing images, based on the UTM projection zone, seamless mosaicking is used for splicing. During the mosaicking process, the brightness and color differences of adjacent images are adjusted through histogram matching to ensure the consistent tone of the fused image. At the same time, resampling technology is used to unify images with different resolutions to a 10-meter resolution to ensure the consistency of spatial details.

[0056] Time series fusion includes aligning data with different time resolutions (such as minute-level sensor data, annual inventory data) to the daily scale based on a unified UTC time (accurate to seconds). For data with time offsets, calibration is performed according to its acquisition period and pattern. For example, if a sensor's data is lagged by 2 hours due to clock error, by calculating its fixed acquisition interval, all data is shifted forward by 2 hours as a whole to achieve precise alignment of the time series.

[0057] Based on the forest resource growth quantification model, data at different time points are concatenated into a continuous growth curve. For missing time period data, interpolation prediction is performed using the model to ensure the integrity of the time series. For example, if a certain area's trees are missing half a year's growth data between 2022 and 2023, by inputting the site factors (elevation, slope, etc.) and existing growth parameters (K, r, t 0 ) of this area into the model, the growth indicators for the missing period are predicted so that the entire time series can accurately reflect the dynamic change process of forest resources.

[0058] Obviously, one or more steps of the method of the present invention can be implemented by a computer program. The computer program for implementing the method of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to the processor of a general-purpose computer, a special-purpose computer, or other programmable data processing devices, so that when the computer program is executed by the processor, the functions / operations specified in the flowchart and / or block diagram are implemented. The computer program can be executed entirely on the machine, partially on the machine, partially on the machine as an independent software package and partially on a remote machine, or entirely on a remote machine or server.

[0059] Therefore, it can be understood that the present invention discloses an electronic device, including: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute one or more steps of the above method.

[0060] Finally, it should be noted that the above are only the preferred embodiments of the present application and are not used to limit the present application. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A data fusion method for a forestry data management system, characterized in that: The method includes the following steps: unifying the data format, purifying the data quality, normalizing the spatial coordinates, and standardizing the time series of the real-time collected data of the forestry data management system; Divide the forest resource data of the target area into different independent areas, and establish a special forest resource growth quantitative model for the forest resource data of each independent area; The forestry data management system detects whether there is a conflict in forest resource data before data fusion; the forest resource data conflict refers to the inconsistency of forest resource growth indicators collected in the same area at different time points; When there is a conflict in the forest resource data of a certain independent area, the conflicting data will be corrected through the quantitative model of forest resource growth in the independent area; When there is a conflict in the forest resource data of multiple independent regions, the conflicting data is corrected through the quantitative growth model of forest resources in these multiple independent regions; Data fusion is completed after correcting the conflicting data.

2. The data fusion method of a forestry data management system according to claim 1 is characterized in that: Purify the real-time data collected by the forestry data management system, including: data integrity check, data consistency processing, and abnormal value and noise data processing; Normalize the spatial coordinates of the real-time collected data of the forestry data management system, including unifying all forestry data into the national geodetic coordinate system, using Gauss-Krüger projection for plane coordinates, dividing the study area into regular grids, assigning a unique ID to each grid, and associating data of different resolutions to corresponding grids through spatial overlay analysis; The real-time collected data of the forestry data management system is standardized in time series, including: all data timestamps are unified in UTC time, satellite remote sensing data, ground sensor data, and forest resource inventory data are uniformly converted to daily scale, for high-frequency ground sensor data, aggregation and downsampling are performed by calculating the mean, and minute-level data are converted to daily-scale data; for low-frequency forest resource inventory data, the cubic spline interpolation method is used to expand the annual data into a continuous time series.

3. The data fusion method of a forestry data management system according to claim 1 is characterized in that: In the forestry data management system, the forest resource data of the target area is divided into different independent areas, including: integrating remote sensing images, terrain data, soil type maps, and vegetation distribution map spatial data to identify areas with similar characteristics, and adjusting the initially divided areas based on administrative boundaries and forest farm boundary vector data; the division method includes clustering algorithm assistance and grid division optimization to merge adjacent grids with similar attributes into independent areas; after dividing the independent areas, each independent area is assigned a unique identification code and marked with key attribute information, including the area name, administrative division, main vegetation type, soil type, and average altitude, to form a regional attribute file; the forest resource data in each independent area is associated with the identification code of the area.

4. The data fusion method of a forestry data management system according to claim 1 is characterized in that: The specific quantitative model of forest resource growth is as follows: , y(t) is the forest resource growth index; When y(t) specifically represents the DBH, K is the DBH limit of forest resources, which means the maximum value of the DBH that can be achieved under the current site conditions, r is the intrinsic growth rate of DBH, which means the inherent growth rate of the DBH of forest resources without environmental restrictions, t0 is the inflection point time of DBH growth, before t0, the DBH growth rate gradually increases; after t0, the DBH growth rate gradually slows down; When y(t) specifically represents the tree height, K is the limit value of the tree height of forest resources, which means the maximum value that the tree height of forest resources can reach under the current site conditions, r is the intrinsic growth rate of tree height, which means the inherent growth rate of tree height of forest resources without environmental restrictions, t0 is the turning point time of tree height growth, before t0, the tree height growth rate gradually increases; after t0, the tree height growth rate gradually slows down; When y(t) specifically represents biomass, K is the limit value of forest resource biomass growth, indicating the maximum value of forest resource biomass growth that can be reached under the current site conditions, r is the intrinsic growth rate of biomass, indicating the inherent growth rate of forest resource biomass growth under no environmental restrictions, t0 is the inflection point time of biomass growth, before t0, the biomass growth rate gradually increases; after t0, the biomass growth rate gradually slows down.

5. The data fusion method of a forestry data management system according to claim 6 is characterized in that: The regression relationship between site factors and the parameters of the quantitative model of forest resource growth was established, and the site factors were introduced to correct the parameters of the quantitative model of forest resource growth.

6. The data fusion method of a forestry data management system according to claim 1, characterized in that: Detection of whether there is a forest resource data conflict specifically includes: for the same monitoring point, if the growth increase in adjacent years exceeds the threshold of the maximum growth rate predicted by the forest resource growth quantitative model, it is determined that there is a data conflict in the time dimension.

7. The data fusion method of a forestry data management system according to claim 6 is characterized in that: When a conflict occurs in the forest resource data of a certain independent area, the conflict data is corrected through the forest resource growth quantification model of the independent area, including: first determining the specific type of the conflict data, which includes: tree breast diameter, tree height, and biomass growth index data; at the same time, based on the grid division and detection area setting, locating the grid range where the conflict data is located, clarifying its spatial position in the independent area, and calling the forest resource growth quantification model parameters; executing the conflict data correction includes calculating the predicted value through the forest resource growth quantification model, replacing the measured conflict value with the predicted value, and completing the data correction.

8. The data fusion method of a forestry data management system according to claim 6 is characterized in that: By integrating the quantitative growth models of forest resources in multiple regions, a joint correction is constructed, which specifically includes: collecting all the quantitative growth model parameters of forest resources involved in the conflict area, summarizing the site factor parameters of each region, and correcting the model parameters through regression relationship to adapt to the growth environment of different regions; setting weights for each region participating in the joint correction according to the spatial distance between each region and the region where the conflict data is located, the similarity of the ecological environment, and the data reliability factor; determining the specific grid where the conflict data is located and the multiple independent regions to which it belongs, and determining the affected regional boundaries and data ranges based on the spatial detection range set during conflict detection; for each independent region participating in the joint correction, substituting the time t corresponding to the conflict data into its forest resource growth quantitative model to calculate their respective predicted values; and then, according to the pre-set regional weights, performing weighted summation on the predicted values ​​of each region to obtain a joint predicted value, and using the joint predicted value as the correction result of the conflict data to replace the original conflict data.

9. The data fusion method of a forestry data management system according to claim 1, characterized in that: After completing the correction of conflicting data, automatically check other regional data associated with the corrected data. If it is found that other regional data may have new conflicts or logical contradictions due to this correction, initiate local secondary detection and adjustment to ensure the overall consistency and rationality of multi-regional data.

10. The data fusion method of a forestry data management system according to claim 1, characterized in that: The data fusion levels include attribute data fusion level, spatial data fusion level, and time series fusion level.

11. An electronic device, characterized in that: include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, wherein the instructions are executed by the at least one processor so that the at least one processor can execute the data fusion method of the forestry data management system of claim 1.

Images