Land survey data acquisition method and system based on multi-source data fusion

Abstract

The invention discloses a territorial survey data acquisition method and system based on multi-source data fusion, and relates to the technical field of data acquisition management. The territorial survey data acquisition method based on multi-source data fusion comprises the following steps of coordinate conversion, coordinate deviation evaluation and image fusion evaluation. The method comprises the following steps: acquiring coordinates through projection coordinate equipment, performing coordinate conversion to obtain projection coordinate equipment conversion coordinates, performing coordinate deviation evaluation according to coordinate evaluation data, judging whether to perform coordinate evaluation data optimization or not, and finally fusing territorial acquisition images to obtain a territorial fusion image. Image fusion evaluation is performed according to the image fusion evaluation parameters so as to judge whether to perform territorial acquisition image optimization or not, so that the image fusion accuracy in territorial survey data acquisition and fusion based on different data sources is improved; the problem that in the prior art, difference of coordinate systems is not fully considered in collection and fusion of territorial survey data based on different data sources is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of data collection and management, and in particular to a land survey data collection method and system based on multi-source data fusion. Background Art

[0002] Land surveys are comprehensive, systematic investigations and analyses of land resources, land cover, and land features to support land resource planning, management, and protection. Traditional land surveys rely on ground surveys and manual data collection, but this method is inefficient and subject to significant errors. Modern land surveys collect spatial data through technologies such as remote sensing and drones, combined with high-precision positioning and measurement equipment. They utilize big data, cloud computing, and geographic information system technologies for data analysis, storage, and processing, greatly improving the efficiency and accuracy of land surveys. Using technologies such as machine learning, land features can be automatically extracted from large amounts of remote sensing data for tasks such as change detection and target identification.

[0003] Existing technologies improve the accuracy of collaborative analysis of multi-source data by fusing and registering data from different sources, and then fusing high-resolution panchromatic images with low-resolution multispectral images.

[0004] For example, the patent application with publication number CN119671537A discloses a highway infrastructure assessment and processing method and system, which includes: collecting multimodal data through satellite remote sensing, drone aerial photography, and sensors, performing data preprocessing and feature extraction. A hierarchical fusion method is used to fuse spatial features with detection features, and then secondary fusion is performed with GIS data to establish the association between spatial features and attribute information; an assessment and prediction model based on support vector machines, random forests, gradient boosting decision trees, and logistic regression is constructed; and based on the assessment indicators and scoring criteria, the associated fusion results and the prediction model are combined to output the highway infrastructure assessment grade.

[0005] For example, the invention application with publication number CN119514870A discloses a method and system for determining the spatiotemporal boundaries of the ecological impact of power transmission and transformation projects based on deep fusion and interpretation of multi-source remote sensing information, including: data acquisition, data preprocessing and multi-source data integration, multi-scale spatiotemporal analysis, ecological impact indicator construction, remote sensing ecological index model construction, ecological impact spatiotemporal boundary determination, dynamic monitoring and early warning system platform establishment, and long-term monitoring and evaluation steps.

[0006] However, in the process of implementing the technical solutions of the embodiments of the present application, the present application discovered that the above technology has at least the following technical problems:

[0007] Current land surveys rely on a variety of independent devices for satellite remote sensing, drone aerial photography, and ground mobile surveying. Data collection is fragmented, and the integration and analysis process is cumbersome. This leads to low survey efficiency and information bias, making it difficult to meet the needs of rapid and accurate land surveys. Different data sources have different resolutions and coordinate systems, which requires processing data heterogeneity during data fusion. This complicates data alignment, unified formatting, and fusion, affecting the accuracy and consistency of the fusion results. Furthermore, the collection and fusion of land survey data from different data sources often fail to fully account for differences in coordinate systems. Summary of the Invention

[0008] The embodiments of the present application provide a land survey data collection method and system based on multi-source data fusion, thereby solving the problem in the prior art of not fully considering the differences in coordinate systems in the collection and fusion of land survey data based on different data sources, and achieving improved image fusion accuracy in the collection and fusion of land survey data based on different data sources.

[0009] The embodiment of the present application provides a land survey data acquisition method based on multi-source data fusion, comprising the following steps: S1, determining the corresponding projection coordinate belts according to the acquisition coordinates of the projection coordinate device, and performing coordinate conversion to obtain the projection coordinate device conversion coordinates, the projection coordinate device acquisition coordinates include the flight route coordinates of the UAV and the driving route coordinates of the ground mobile measurement device, the coordinate conversion is used to eliminate the heterogeneity of the coordinate systems of different land survey data acquisition devices, the projection coordinate device conversion coordinates include the UAV conversion coordinates and the ground measurement conversion coordinates, which are used to reflect the spatial position characteristics of the land acquisition image, and the land survey data acquisition equipment includes satellite remote sensing equipment, UAV and ground mobile measurement equipment; S2, according to the coordinates of the projection coordinate device The evaluation data is evaluated for coordinate deviation to determine whether to optimize the coordinate evaluation data. The coordinate evaluation data includes the coordinates collected by the projection coordinate device and the coordinates converted by the projection coordinate device. The optimization of the coordinate evaluation data indicates that the coordinates of the land survey data acquisition device are optimized to improve the data accuracy of the coordinate evaluation data of the projection coordinate device. S3, after the optimization of the coordinate evaluation data, the land acquisition images of the same coordinates after the coordinate conversion are fused to obtain a land fusion image. An image fusion evaluation is performed based on the image fusion evaluation parameters of the land fusion image and the land acquisition image to determine whether to optimize the land acquisition image. The optimization of the land acquisition image indicates that the fusion process of the land fusion image is optimized to eliminate the resolution heterogeneity existing in the land acquisition image.

[0010] The embodiment of the present application provides a land survey data acquisition system based on multi-source data fusion, including: a coordinate conversion module, a coordinate deviation evaluation module and an image fusion evaluation module; wherein the coordinate conversion module is used to determine the corresponding projection coordinate band according to the coordinates collected by the projection coordinate device, and perform coordinate conversion to obtain the projection coordinate device conversion coordinates, the projection coordinate device collection coordinates include the flight route coordinates of the drone and the driving route coordinates of the ground mobile measurement device, the coordinate conversion is used to eliminate the heterogeneity of the coordinate systems of different land survey data acquisition devices, the projection coordinate device conversion coordinates include the drone conversion coordinates and the ground measurement conversion coordinates, which are used to reflect the spatial position characteristics of the land acquisition image, and the land survey data acquisition equipment includes satellite remote sensing equipment, drones and ground mobile measurement equipment; coordinate deviation evaluation module The module is used to evaluate the coordinate deviation based on the coordinate evaluation data of the projection coordinate device and determine whether to optimize the coordinate evaluation data. The coordinate evaluation data includes the acquisition coordinates of the projection coordinate device and the conversion coordinates of the projection coordinate device. The coordinate evaluation data optimization represents the optimization of the coordinates of the land survey data acquisition device to improve the data accuracy of the coordinate evaluation data of the projection coordinate device; the image fusion evaluation module is used to fuse the land acquisition images of the same coordinates after coordinate conversion to obtain a land fusion image after optimizing the coordinate evaluation data, and perform image fusion evaluation based on the image fusion evaluation parameters of the land fusion image and the land acquisition image to determine whether to optimize the land acquisition image. The land acquisition image optimization represents the optimization of the fusion process of the land fusion image to eliminate the resolution heterogeneity in the land acquisition image.

[0011] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0012] 1. Coordinates are collected by projection coordinate equipment and then converted to obtain projection coordinate equipment conversion coordinates. Then, coordinate deviation evaluation is performed based on the coordinate evaluation data to optimize the coordinate evaluation data. Finally, image fusion evaluation is performed based on image fusion evaluation parameters to optimize the land acquisition image, thereby eliminating the heterogeneity of data collected under different coordinate systems, and thus achieving improved image fusion accuracy in the collection and fusion of land survey data based on different data sources, effectively solving the problem in the existing technology that the differences in coordinate systems are not fully considered in the collection and fusion of land survey data based on different data sources.

[0013] 2. The first coordinate comparison coefficient is obtained through the initial flight coordinate deviation coefficient, the initial driving coordinate deviation coefficient and the UAV-ground position deviation coefficient. Then, the initial UAV conversion deviation coefficient is obtained according to the initial UAV conversion coordinate. Then, the initial ground conversion deviation coefficient is obtained according to the initial ground conversion coordinate. Finally, the coordinate deviation evaluation result is obtained for the initial UAV conversion deviation coefficient and the initial ground conversion deviation coefficient. This quantitatively evaluates the coordinate acquisition and conversion accuracy of the UAV and ground mobile measurement equipment, thereby achieving improved accuracy of the coordinates of the land survey equipment.

[0014] 3. The edge amplitude ratio coefficient is obtained by the edge pixel gradient amplitude of the land acquisition image and the edge pixel gradient amplitude of the land fusion image, and then the multi-scale information entropy fusion comparison coefficient is obtained by combining the multi-scale fusion information entropy with the information entropy of the corresponding sub-band obtained by wavelet decomposition in the land fusion image. Finally, the image fusion evaluation result is obtained by the structural similarity index, edge amplitude ratio coefficient, inverse proportional artifact index and multi-scale information entropy fusion comparison coefficient, thereby quantitatively evaluating the fusion effect of the land fusion image and improving the quality of the land fusion image. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A flowchart of a land survey data collection method based on multi-source data fusion provided in an embodiment of the present application;

[0016] Figure 2 A structural diagram of a land survey data acquisition system based on multi-source data fusion provided in an embodiment of the present application. DETAILED DESCRIPTION

[0017] The embodiments of the present application solve the problem in the prior art of insufficient consideration of coordinate system differences in the collection and fusion of land survey data based on different data sources by providing a land survey data collection method and system based on multi-source data fusion. Coordinates are collected by a projection coordinate device and coordinate conversion is performed to obtain projection coordinate device conversion coordinates. Then, coordinate deviation evaluation is performed based on the coordinate evaluation data to determine whether to optimize the coordinate evaluation data. Finally, the land collection images are fused to obtain a land fusion image. Image fusion evaluation is performed based on image fusion evaluation parameters to determine whether to optimize the land collection image. This achieves image fusion accuracy in the collection and fusion of land survey data based on different data sources.

[0018] The technical solution in the embodiment of the present application is to solve the problem that the differences in coordinate systems are not fully considered in the collection and fusion of land survey data based on different data sources. The overall idea is as follows:

[0019] The coordinates are collected by the projection coordinate device and the coordinate conversion is performed to obtain the projection coordinate device conversion coordinates. Then, the coordinate deviation is evaluated based on the coordinate evaluation data to optimize the coordinate evaluation data. Finally, the image fusion evaluation is performed based on the image fusion evaluation parameters to optimize the land collection image, thereby achieving the effect of improving the image fusion accuracy in the collection and fusion of land survey data based on different data sources.

[0020] In order to better understand the above technical solution, the above technical solution will be described in detail below with reference to the accompanying drawings and specific implementation methods.

[0021] like Figure 1 As shown, it is a flowchart of the land survey data collection method based on multi-source data fusion provided in an embodiment of the present application, and the method includes the following steps: S1, coordinate conversion: determine the corresponding projection coordinate belts according to the coordinates collected by the projection coordinate device, and perform coordinate conversion to obtain the projection coordinate device conversion coordinates. The coordinates collected by the projection coordinate device include the flight route coordinates of the UAV and the driving route coordinates of the ground mobile measuring device. The flight route coordinates of the UAV and the driving route coordinates of the ground mobile measuring device represent the coordinates of the UAV ground mobile measuring device and the actual position on the ground; the projection coordinate belt is obtained through the longitude range of the position, for example, longitude 114° is located at belt number 50N; the width of each projection coordinate belt is 6 degrees of longitude.

[0022] The coordinate conversion involved here refers to converting the coordinate system used by the projection coordinate device to collect coordinates into a geographic coordinate system to eliminate the heterogeneity of the coordinate systems of different land survey data collection devices. The coordinate system used by the projection coordinate device to collect coordinates is usually the Universal Transverse Mercator Grid System (UTM coordinate system). The geographic coordinate system refers to the coordinate system used for macroscopic land images collected by satellite remote sensing equipment. The geographic coordinate system is usually the World Geodetic System 1984 (WGS84 coordinate system). The coordinate conversion of the projection coordinate device includes drone conversion coordinates and geodetic conversion coordinates, which are used to reflect the spatial location characteristics of the land collection image. Land survey data collection equipment includes satellite remote sensing equipment, drones, and ground mobile measurement equipment.

[0023] S2, coordinate deviation assessment: coordinate deviation assessment is performed based on the coordinate evaluation data of the projection coordinate device to determine whether coordinate evaluation data optimization is performed. The projection coordinate device includes drones and ground mobile measurement equipment. The coordinate evaluation data includes the coordinates collected by the projection coordinate device and the coordinates converted by the projection coordinate device. Coordinate evaluation data optimization means optimizing the coordinates of the land survey data acquisition device to improve the data accuracy of the coordinate evaluation data of the projection coordinate device.

[0024] S3, image fusion evaluation: After the coordinate evaluation data is optimized, the land acquisition images of the same coordinates after coordinate conversion are fused to obtain a land fusion image. Image fusion evaluation is performed based on the image fusion evaluation parameters of the land fusion image and the land acquisition image to determine whether to optimize the land acquisition image. The land acquisition images include macro land images, meso land images, and micro land images. Land acquisition image optimization means optimizing the fusion process of the land fusion image to eliminate the resolution heterogeneity in the land acquisition images; macro land images are images collected by satellite remote sensing, meso land images are images collected by drone aerial photography, and micro land images are images collected by ground mobile measurement.

[0025] In this example, current land surveys rely on a variety of independent devices for satellite remote sensing, drone aerial photography, and ground mobile surveying. This fragmented data collection and cumbersome integration and analysis processes result in low survey efficiency and information bias, making it difficult to meet the needs of rapid and accurate land surveys. Different data sources have different resolutions and coordinate systems, requiring data fusion to address heterogeneity. This complicates data alignment, unified formatting, and fusion, impacting the accuracy and consistency of the fusion results.

[0026] This application improves the efficiency of land surveys by integrating and analyzing data obtained from satellite remote sensing, drone aerial photography, and ground mobile measurement; through coordinate transformation and deviation evaluation, it eliminates the heterogeneity generated when different devices collect data in different coordinate systems, thereby ensuring that data from different devices can be effectively fused and compared in the same coordinate system, ensuring the data quality and spatial consistency of land surveys; through coordinate evaluation data optimization, it improves the positioning accuracy of drones and ground measurement equipment, reduces possible deviations, and thus improves the accuracy of subsequent data fusion and analysis; through panchromatic image enhancement fusion (Pansharpening) technology, it fuses land acquisition images to eliminate resolution heterogeneity between different images, improve image details and enhance spatial resolution, thereby improving the image fusion accuracy and land survey accuracy in the collection and fusion of land survey data based on different data sources.

[0027] Furthermore, the specific process of performing coordinate conversion to obtain the conversion coordinates of the projection coordinate device is as follows: D1, respectively determine whether the projection coordinate bands corresponding to the coordinates collected by the projection coordinate device are unique. If they are all unique, execute D2, otherwise execute D3; D2, coordinate conversion is performed on the coordinates collected by the projection coordinate device through the Quantum Geographic Information System (QGIS) to obtain the conversion coordinates of the projection coordinate device; D3, if the projection coordinate band corresponding to the flight route coordinates of the UAV is not unique, the flight route coordinates of the UAV in each projection coordinate band are coordinate converted and merged to obtain the UAV conversion coordinates; if the projection coordinate band corresponding to the driving route coordinates of the ground mobile measurement device is not unique, the driving route coordinates of the ground mobile measurement device in each projection coordinate band are coordinate converted and merged to obtain the ground measurement conversion coordinates; after the coordinate conversion, when merging the coordinates, the projection coordinate device conversion coordinates after coordinate conversion are spliced together according to the order in which the projection coordinate device collected the coordinates.

[0028] In this embodiment, since the coordinates collected by the projection coordinate device may span multiple projection coordinate systems, the uniqueness of the projection coordinate band is determined and coordinate conversion is performed to ensure that the coordinates in each projection coordinate band are accurately converted, thereby improving the accuracy of the coordinate conversion of the projection coordinate device.

[0029] Furthermore, coordinate deviation evaluation is performed based on the coordinate evaluation data of the projection coordinate device to determine whether to optimize the coordinate evaluation data. The specific method is as follows:

[0030] The first coordinate deviation adjustment factor is introduced to assign the result of absolute deviation processing between the initial flight coordinates of the UAV flight route and the preset initial flight coordinates obtained from the preset database to obtain the initial flight coordinate deviation coefficient, i.e. FZP. The specific restriction expression is:

[0031]

[0032] Where WZB x Indicates the initial flight horizontal coordinate of the UAV flight route, WZB y Indicates the initial flight ordinate of the UAV’s flight route, Indicates the preset initial flight horizontal coordinate of the drone's preset route, It represents the preset initial flight vertical coordinate of the preset route of the UAV, and Z1 represents the first coordinate deviation adjustment factor; the initial flight coordinates (including the initial flight horizontal coordinate and the initial flight vertical coordinate) are represented by the coordinates of the starting point recorded by the Global Positioning System (GPS device) on the UAV, and the preset initial flight coordinates represent the starting point of the preset route. The preset route is set according to the preset personnel.

[0033] The second coordinate deviation adjustment factor is introduced to assign the result of absolute deviation processing between the initial driving coordinates of the driving route of the ground mobile measurement device and the preset initial driving coordinates obtained from the preset database to obtain the initial driving coordinate deviation coefficient, namely XZP. The specific restriction expression is:

[0034]

[0035] Where, DZB x Indicates the initial travel horizontal coordinate of the travel route of the ground mobile measuring device, DZB y Indicates the initial travel ordinate of the travel route of the ground mobile measurement device, represents the preset initial travel abscissa of the preset route of the ground mobile measuring device, It represents the preset initial driving vertical coordinate of the preset route of the ground mobile measuring device, Z2 represents the second coordinate deviation adjustment factor, and the initial driving coordinates (including the initial driving horizontal coordinate and the initial driving vertical coordinate) are represented by the coordinates of the starting point of the driving path recorded by the ground mobile test device; the preset initial driving coordinates represent the starting point of the specified route, and the specified route is set according to the preset personnel.

[0036] The third coordinate deviation adjustment factor is introduced to assign a value to the absolute deviation processing result of the initial flight coordinates of the UAV flight route and the initial driving coordinates of the ground mobile measurement device’s driving route, and the UAV-ground measurement position deviation coefficient, namely WDP, is obtained. The specific restriction expression is:

[0037] WDP=Z3×(|WZB x -DZB x |+|WZB y -DZB y |);

[0038] Where WZB x Indicates the initial flight horizontal coordinate of the UAV flight route, WZB y Indicates the initial flight ordinate of the UAV flight route, DZB x Indicates the initial travel horizontal coordinate of the travel route of the ground mobile measuring device, DZB y It represents the initial ordinate of the travel route of the ground mobile measuring device, and Z3 represents the third coordinate deviation adjustment factor.

[0039] The first coordinate comparison coefficient is obtained by coupling the initial flight coordinate deviation coefficient, the initial driving coordinate deviation coefficient, and the UAV-ground position deviation coefficient: BDX = FZP + XZP + WDP. Here, BDX represents the first coordinate comparison coefficient.

[0040] According to the absolute deviation comparison processing of the initial UAV conversion coordinates and the preset initial UAV conversion coordinates obtained from the preset database, the initial UAV conversion deviation coefficient is obtained, that is, Where WZB x' Indicates the initial UAV conversion horizontal coordinate, WZB y' Indicates the initial UAV transformation ordinate, Indicates the preset initial drone conversion horizontal coordinate, Indicates the preset initial UAV conversion vertical coordinate; the initial UAV conversion coordinate (including the initial UAV conversion horizontal coordinate and the initial UAV conversion vertical coordinate) is obtained by coordinate conversion of the initial flight coordinate. The preset initial UAV conversion coordinate represents the coordinate of the preset initial flight coordinate in the geographic coordinate system, which is set according to the preset personnel.

[0041] The absolute deviation comparison process is performed based on the initial geodetic conversion coordinates and the preset initial geodetic conversion coordinates obtained from the preset database to obtain the initial geodetic conversion deviation coefficient, that is, Where, DZB x' Indicates the initial geodetic transformation horizontal coordinate, DZB y' represents the initial geodetic transformation ordinate, Indicates the preset initial geodetic transformation horizontal coordinate, Indicates the preset initial geodetic conversion vertical coordinate; the initial geodetic conversion coordinate (including the initial geodetic conversion horizontal coordinate and the initial geodetic conversion vertical coordinate) is obtained by coordinate conversion of the initial driving coordinate. The preset initial geodetic conversion coordinate represents the coordinate of the preset initial driving coordinate in the geographic coordinate system and is set according to the preset personnel.

[0042] The fourth and fifth coordinate deviation adjustment factors are introduced to assign the initial UAV conversion deviation coefficients and the initial ground measurement conversion deviation coefficients. Combined with the first coordinate comparison coefficient, a coupling process is performed to obtain the coordinate deviation evaluation result. The coordinate deviation evaluation result is used to quantitatively evaluate the acquisition and conversion accuracy of the flight coordinates of the UAV flight route and the driving coordinates of the ground mobile measurement equipment. The specific restriction expression is:

[0043]

[0044] Where ZP represents the coordinate deviation evaluation result, Z4 represents the fourth coordinate deviation adjustment factor, and Z5 represents the fifth coordinate deviation adjustment factor.

[0045] The coordinate deviation adjustment factors involved are obtained from a preset database. The first coordinate deviation adjustment factor represents the influence of the initial flight coordinates and the preset initial flight coordinates on the coordinate deviation evaluation result. The second coordinate deviation adjustment factor represents the influence of the initial driving coordinates and the preset initial driving coordinates on the coordinate deviation evaluation result. The third coordinate deviation adjustment factor represents the influence of the initial flight coordinates and the initial driving coordinates on the coordinate deviation evaluation result. The fourth coordinate deviation adjustment factor represents the influence of the initial UAV conversion coordinates and the preset initial UAV conversion coordinates on the coordinate deviation evaluation result. The fifth coordinate deviation adjustment factor represents the influence of the initial ground-measured conversion coordinates and the preset initial ground-measured conversion coordinates on the coordinate deviation evaluation result. The sum of the five is 1. For example, the initial flight coordinates, the preset initial flight coordinates and the preset first coordinate deviation adjustment factor form a mapping set. The real-time initial flight coordinates and the preset initial flight coordinates are input into the mapping set to obtain the corresponding first coordinate deviation adjustment factor. The initial driving coordinates , the preset initial driving coordinates and the preset second coordinate deviation adjustment factor form a mapping set, and the real-time initial driving coordinates and the preset initial driving coordinates are input into the mapping set to obtain the corresponding second coordinate deviation adjustment factor; the initial flight coordinates, the initial driving coordinates and the preset third coordinate deviation adjustment factor form a mapping set, and the real-time initial flight coordinates and the initial driving coordinates are input into the mapping set to obtain the corresponding third coordinate deviation adjustment factor; the initial UAV conversion coordinates, the preset initial UAV conversion coordinates and the preset fourth coordinate deviation adjustment factor form a mapping set, and the real-time initial UAV conversion coordinates and the preset initial UAV conversion coordinates are input into the mapping set to obtain the corresponding fourth coordinate deviation adjustment factor; the initial geodetic conversion coordinates, the preset initial geodetic conversion coordinates and the preset fifth coordinate deviation adjustment factor form a mapping set, and the real-time initial geodetic conversion coordinates and the preset initial geodetic conversion coordinates are input into the mapping set to obtain the corresponding fifth coordinate deviation adjustment factor; the mapping relationship can be one-to-one or many-to-one.

[0046] The specific process involved in determining whether to optimize coordinate evaluation data is as follows:

[0047] A1, determine whether the coordinate deviation evaluation result is greater than the first coordinate deviation threshold obtained from the preset database. If the coordinate deviation evaluation result is greater than the first coordinate deviation threshold obtained from the preset database, optimize the first coordinate evaluation data, otherwise execute A2; the first coordinate deviation threshold is represented by the sum of the average value and three times the standard deviation of the coordinate deviation evaluation result in the historical time period.

[0048] A2, determine whether the coordinate deviation evaluation result is greater than the second coordinate deviation threshold obtained from the preset database. If the coordinate deviation evaluation result is greater than the second coordinate deviation threshold obtained from the preset database, optimize the second coordinate evaluation data; the second coordinate deviation threshold is represented by the average value of the coordinate deviation evaluation results in the historical time period.

[0049] The first coordinate evaluation data optimization includes coordinate evaluation data noise filtering, land survey data acquisition equipment position calibration and image re-acquisition. The first coordinate evaluation data optimization means performing noise filtering and image re-acquisition on the coordinate evaluation data to eliminate errors caused by equipment errors or environmental influences during data acquisition; performing noise filtering through Kalman filtering; and equipment position calibration, which is to adjust the position of the equipment through known reference points (such as preset initial flight coordinates, preset initial driving coordinates) to ensure that the coordinate data it collects is aligned with the actual position.

[0050] The second coordinate evaluation data optimization means adjusting the projection coordinate device conversion coordinates to ensure data consistency. The specific process is as follows:

[0051] The initial UAV conversion coordinates and the preset initial UAV conversion coordinates are processed by the least square method to obtain the rotation adjustment angle and translation adjustment distance of the initial UAV conversion coordinates. The scale adjustment factor of the initial UAV conversion coordinates is obtained by processing the square of the deviation between the initial UAV conversion coordinates and the preset initial UAV conversion coordinates. The specific formula for the lateral translation adjustment distance is: Among them, Δx is the horizontal translation adjustment distance, WZB x' Indicates the initial UAV transformation horizontal coordinate, Indicates the preset initial drone conversion horizontal coordinate; the specific formula for the longitudinal translation adjustment distance is: Among them, Δy is the longitudinal translation adjustment distance, WZB y' Indicates the initial UAV transformation ordinate, Indicates the preset initial drone transformation vertical coordinate; the rotation matrix is: The sum of squared rotation errors is: Where θ is the rotation angle, J θ is the rotation error, X' represents the horizontal coordinate obtained after rotation, and Y' represents the vertical coordinate obtained after rotation. The rotation adjustment angle is obtained by taking the derivative of the sum of squares of the rotation error; the scale adjustment factor is: Where γ is the scale adjustment factor.

[0052] The initial geodesic transformation coordinates and the preset initial geodesic transformation coordinates are subjected to square deviation processing by the least squares method to obtain the rotation adjustment angle and translation adjustment distance of the initial geodesic transformation coordinates. The scale adjustment factor of the initial geodesic transformation coordinates is obtained according to the translation adjustment distance of the initial geodesic transformation coordinates, the initial geodesic transformation coordinates and the preset initial geodesic transformation coordinates. Similarly, the rotation adjustment angle, translation adjustment distance and scale adjustment factor of the initial geodesic transformation coordinates are obtained.

[0053] The coordinate deviation evaluation result and the second coordinate deviation threshold obtained from the preset database are compared to obtain the adjustment reliability. The adjustment reliability is used to quantify the accuracy of the coordinate adjustment parameters. The specific formula of the adjustment reliability is: Wherein, kxd represents the adjustment credibility, ZP represents the coordinate deviation evaluation result, and ZP0 represents the second coordinate deviation threshold.

[0054] The adjustment credibility and coordinate adjustment parameters are input into the adjustment amount mapping set to obtain the coordinate adjustment amount. The coordinate adjustment parameters include the rotation adjustment angle, translation adjustment distance, and scale adjustment factor of the initial UAV conversion coordinates and the rotation adjustment angle, translation adjustment distance, and scale adjustment factor of the initial ground measurement conversion coordinates. The adjustment amount mapping set is a set of mapping relationships between the adjustment credibility, coordinate adjustment parameters, and corresponding coordinate adjustment amounts obtained from a preset database.

[0055] Adjust the initial UAV conversion coordinates and the initial ground measurement conversion coordinates one by one according to the coordinate adjustment amount until the termination condition of the second coordinate evaluation data optimization is reached; first rotate the initial UAV conversion coordinates and the initial ground measurement conversion coordinates, and then perform scale adjustment and translation adjustment. The termination condition of the second coordinate evaluation data optimization involved is:

[0056] B1, judge whether the coordinate deviation evaluation result after each coordinate adjustment is greater than the second coordinate deviation threshold. If the coordinate deviation evaluation result after each coordinate adjustment is greater than the second coordinate deviation threshold, execute B2, otherwise terminate the second coordinate evaluation data optimization; B2, judge whether the total coordinate adjustment amount is less than the coordinate adjustment parameter. If the total coordinate adjustment amount is less than the coordinate adjustment parameter, continue to adjust the drone conversion coordinates and the ground measurement conversion coordinates one by one, otherwise terminate the second coordinate evaluation data optimization; the total coordinate adjustment amount is obtained by summing the coordinate adjustment amounts under all the adjusted times.

[0057] In this embodiment, since the device of the present application is equipped with a foldable satellite signal receiving antenna, a retractable drone cabin and a vehicle body equipped with a ground mobile measurement component, the drone can automatically take off and land from the equipment cabin to achieve rapid operation; therefore, when the coordinates do not consider the altitude, the initial flight coordinates of the drone and the initial driving coordinates of the ground mobile measurement device may be the same. Therefore, the larger the drone-ground position deviation coefficient, the lower the measurement accuracy of the device coordinates, which leads to a larger initial flight coordinate deviation coefficient and an initial driving coordinate deviation coefficient, and then leads to a larger initial drone conversion deviation coefficient and an initial ground measurement conversion deviation coefficient; the larger the initial drone conversion deviation coefficient, the more likely it is that there is an error in the coordinate conversion, which may lead to a larger initial ground measurement conversion deviation coefficient.

[0058] Through the above steps, the acquisition and conversion accuracy of the flight coordinates of the UAV flight route and the driving coordinates of the ground mobile measurement equipment were quantitatively evaluated, effectively reducing the coordinate deviation caused by equipment errors, environmental factors and differences between different equipment, improving the accuracy and consistency of the coordinates, and enhancing the overall efficiency of land survey work.

[0059] Noise filtering effectively removes unnecessary interference, ensuring more accurate device coordinate data for subsequent image fusion. Device position calibration and image recapture eliminate coordinate deviations caused by device errors, avoiding image acquisition errors due to coordinate measurement errors and improving the accuracy of data integration. The coordinates of drones and ground mobile measurement devices are adjusted to ensure data consistency and accuracy.

[0060] By optimizing the coordinates through the least squares method, errors caused by measurement errors and equipment accuracy differences can be effectively reduced, so that the coordinate systems of coordinate data from different data sources can be aligned to ensure consistent results during data fusion. By setting the adjustment confidence level for gradual adjustment, the optimization can be flexibly stopped to avoid excessive adjustment, thereby improving the accuracy of the equipment coordinates and ensuring the data quality in land surveys.

[0061] Furthermore, an image fusion evaluation is performed based on the image fusion evaluation parameters of the land fusion image and the land acquisition image to determine whether to optimize the land acquisition image. The specific method is as follows:

[0062] In the first step, the edge pixel gradient amplitude of the land acquisition image and the edge pixel gradient amplitude of the land fusion image are compared to obtain the edge amplitude ratio coefficient, that is, Where BYQ iIt represents the edge pixel gradient amplitude of the i-th land acquisition image, BYQ0 represents the edge pixel gradient amplitude of the land fusion image, and the edge pixel gradient amplitude of the land acquisition image and the land fusion image is obtained by performing the square root operation on the square sum of the horizontal gradient and vertical gradient obtained by Canny edge detection.

[0063] In the second step, the inverse proportional operation is performed on the artifact index to obtain the inverse proportional artifact index, that is, In the formula, WEY i Represents the artifact index of the i-th type of land acquisition image; the artifact index is obtained by Fourier spectrum transform to detect the high-frequency artifact energy and the ratio of all frequency energies. High frequency refers to the frequency in the spectrum greater than the cutoff frequency, and the cutoff frequency is set by the preset personnel; Where f0 is the cutoff frequency and F(u, v) is the frequency domain signal.

[0064] In the third step, the information entropy of the corresponding sub-bands obtained by wavelet decomposition of all land acquisition images is coupled to obtain multi-scale fusion information entropy, namely DXN n ; Among them, the specific limiting expression of multi-scale fusion information entropy is:

[0065]

[0066] Where, XIN i.n Represents the information entropy of the nth sub-band in the i-th land acquisition image; perform 3-layer wavelet decomposition on the land acquisition image and calculate the information entropy of each sub-band.

[0067] The fourth step is to compare the multi-scale fusion information entropy with the information entropy of the corresponding sub-band obtained by wavelet decomposition in the land fusion image, and then perform coupled average processing on the result, and perform inverse proportional operation to obtain the multi-scale information entropy fusion comparison coefficient, that is, Where, DXN n represents the multi-scale fusion information entropy of the n-th layer sub-band, XIN 0.n Represents the information entropy of the nth sub-band in the land fusion image; perform 3-layer wavelet decomposition on the land fusion image and calculate the information entropy of each sub-band.

[0068] In the fifth step, the image fusion adjustment factor is introduced to assign values to the structural similarity index, edge amplitude ratio coefficient, inverse proportional artifact index and multi-scale information entropy fusion contrast coefficient, and then coupled average processing is performed to obtain the image fusion evaluation result. The image fusion evaluation result is used to quantitatively evaluate the fusion effect of the land fusion image; the structural similarity index between the land fusion image and the land collection image is calculated separately through OpenCV.

[0069] The specific restriction expression of the image fusion evaluation result is:

[0070]

[0071] Where i represents the type number of the land acquisition image, i = 1 represents the macroscopic land image, i = 2 represents the mesoscopic land image, i = 3 represents the microscopic land image, n represents the subband number obtained by wavelet decomposition, n = 1 represents the first subband of wavelet decomposition, n = 2 represents the second subband of wavelet decomposition, n = 3 represents the third subband of wavelet decomposition, n = 4 represents the fourth subband of wavelet decomposition, TR represents the image fusion evaluation result, SSM i represents the structural similarity index of the i-th type of land acquisition image, T1 represents the first image fusion adjustment factor, T2 represents the second image fusion adjustment factor, T3 represents the third image fusion adjustment factor, and T4 represents the fourth image fusion adjustment factor.

[0072] The image fusion adjustment factors involved are obtained from a preset database. The first image fusion adjustment factor represents the degree of influence of the structural similarity index on the image fusion evaluation result. The second image fusion adjustment factor represents the degree of influence of the edge pixel gradient amplitude and the edge pixel gradient amplitude of the land fusion image on the image fusion evaluation result. The third image fusion adjustment factor represents the degree of influence of the artifact index on the image fusion evaluation result. The fourth image fusion adjustment factor represents the degree of influence of the artifact index and the information entropy in the land fusion image on the image fusion evaluation result. The sum of the four is 1. For example, the structural similarity index and the preset first image fusion adjustment factor form a mapping set, and the real-time structural similarity index is input into the mapping set to obtain the corresponding first image fusion adjustment factor. The edge pixel gradient amplitude, the edge pixel gradient amplitude of the land fusion image and the preset second image fusion adjustment factor form a mapping set, and the real-time edge pixel gradient amplitude and the edge pixel gradient amplitude of the land fusion image are input into the mapping set to obtain the corresponding second image fusion adjustment factor; the artifact index and the preset third image fusion adjustment factor form a mapping set, and the real-time artifact index is input into the mapping set to obtain the corresponding third image fusion adjustment factor; the multi-scale fusion information entropy, the information entropy in the land fusion image and the preset fourth image fusion adjustment factor form a mapping set, and the real-time multi-scale fusion information entropy and the information entropy in the land fusion image are input into the mapping set to obtain the corresponding fourth image fusion adjustment factor; the mapping relationship can be one-to-one or many-to-one.

[0073] The specific process involved in determining whether to optimize land acquisition images is as follows:

[0074] C1, determines whether the image fusion evaluation result is greater than the preset image fusion threshold obtained from the preset database. If the image fusion evaluation result is not greater than the preset image fusion threshold obtained from the preset database, image quality optimization is performed and then C2 is executed. Otherwise, the land fusion image is stored in the land information dataset. Image quality optimization includes image denoising and image histogram equalization. Image histogram equalization is achieved through contrast-limited adaptive histogram equalization (CLAHE); the preset image fusion threshold is represented by the average value of the image fusion evaluation results in the historical time period; Gaussian filtering is used to eliminate noise in the land fusion image; contrast-limited adaptive histogram equalization performs histogram equalization in a local area through an adaptive method, and limits contrast enhancement to avoid noise or artifacts caused by excessive enhancement. The contrast-limited threshold is set according to the preset personnel, for example, it can be set to 0.02.

[0075] C2, judge whether the image fusion evaluation result after image quality optimization is greater than the preset image fusion threshold obtained from the preset database. If the image fusion evaluation result after image quality optimization is not greater than the preset image fusion threshold obtained from the preset database, perform registration optimization and then execute C3, otherwise terminate the land acquisition image optimization.

[0076] C3, determines whether the image fusion evaluation result after registration optimization is greater than the preset image fusion threshold obtained from the preset database. If the image fusion evaluation result after registration optimization is not greater than the preset image fusion threshold obtained from the preset database, C4 is executed after removing the artifacts, otherwise the land acquisition image optimization is terminated. The land acquisition image is decomposed by discrete wavelet transform to remove artifacts, and the artifacts are reduced by separating the low-frequency and high-frequency parts. The sub-band with a frequency greater than the preset frequency is set as the high-frequency part, and the sub-band with a frequency not greater than the preset frequency is set as the low-frequency part; the preset frequency is set according to the preset personnel.

[0077] C4, determines whether the image fusion evaluation result after removing artifacts is greater than the preset image fusion threshold obtained from the preset database. If the image fusion evaluation result after removing artifacts is not greater than the preset image fusion threshold obtained from the preset database, C5 is executed after the image information entropy is optimized, otherwise the land acquisition image optimization is terminated.

[0078] C5 determines whether the image fusion evaluation result after image information entropy optimization is greater than the preset image fusion threshold obtained from the preset database. If the image fusion evaluation result after image information entropy optimization is not greater than the preset image fusion threshold obtained from the preset database, feedback is provided; otherwise, the land acquisition image optimization is terminated. Land acquisition image optimization includes image quality optimization, registration optimization, artifact removal, and image information entropy optimization.

[0079] Specifically, the registration optimization process is as follows: The land acquisition images are coarsely registered using feature point matching and the RANdomSAmple Consensus (RANSAC) algorithm. If the image fusion evaluation result after coarse registration is not greater than a preset image fusion threshold obtained from a preset database, the land acquisition images after coarse registration are finely registered using the Levenberg-Marquardt Algorithm (LM algorithm). Otherwise, the land acquisition images are fused. Fine registration and coarse registration refer to registering the land acquisition images using different algorithms to improve image registration accuracy. Fusion of the land acquisition images refers to decomposing land acquisition images of different resolutions into different scales using wavelet transform, and then performing weighted averaging fusion of subbands of a preset frequency in combination with a subband fusion control factor. The subband fusion control factor is obtained by inputting the image fusion evaluation result and the structural similarity index into a subband fusion mapping set. The subband fusion mapping set is a set of mapping relationships obtained from a preset database that represents the image fusion evaluation result, the structural similarity index, and the subband fusion control factor.

[0080] Specifically, the specific process of image information entropy optimization is as follows: determine whether the information entropy corresponding to the subband obtained by wavelet transform is greater than the preset information entropy threshold obtained from the preset database; if the information entropy corresponding to the subband obtained by wavelet transform is greater than the preset information entropy threshold obtained from the preset database, retain the corresponding subband; otherwise, interpolate and reconstruct the corresponding subband through the adjacent subbands of the corresponding subband; perform interpolation and reconstruction through spline interpolation to improve the accuracy of interpolation and reconstruction.

[0081] In this embodiment, the higher the structural similarity index, the higher the degree of structural information retention in the land acquisition image, the larger the edge amplitude ratio coefficient, and the smaller the artifact index; the larger the edge amplitude ratio coefficient, the clearer the edge information of the land acquisition image, thereby improving the structural similarity index and reducing the artifact index; the closer the multi-scale information entropy fusion contrast coefficient is to 0, the better the fusion effect of the land acquisition image, the higher the structural similarity index and the edge amplitude ratio coefficient, and the smaller the artifact index.

[0082] Through the above steps, the fusion effect of the national land fusion image was quantitatively evaluated, which provided a basis for the subsequent optimization of the national land fusion image and improved the quality of the national land fusion image.

[0083] Through denoising processing and image histogram equalization, the noise in the land fusion image is removed, and the details of the land fusion image are enhanced, making the land fusion image clearer and richer in details; through registration optimization, the spatial alignment between the land acquisition images is ensured, and the accuracy of the land fusion image is improved; through artifact removal, the distortion in the land acquisition image is reduced, and the authenticity of the land fusion image is improved; through information entropy optimization, the complexity and detail level of the land fusion image are improved, and the information retention degree of the land fusion image is improved.

[0084] Furthermore, the method further includes: determining whether the registration time of the land acquisition image is greater than the preset registration time obtained from the preset database: if the registration time of the land acquisition image is greater than the preset registration time obtained from the preset database, optimizing the registration time using distributed computing; if the registration time of the land acquisition image is not greater than the preset registration time obtained from the preset database, continuing the operation. The registration time is calculated by the time difference between the timestamp of the recorded land acquisition image completion time and the land acquisition image registration completion time; the preset registration time is set by the preset personnel.

[0085] In this embodiment, through distributed computing, the registration task is assigned to multiple computing nodes for parallel processing, which shortens the registration time and improves the registration efficiency of land acquisition images, thereby achieving the collection efficiency of land survey data.

[0086] like Figure 2As shown, it is a structural diagram of a land survey data acquisition system based on multi-source data fusion provided by an embodiment of the present application. The land survey data acquisition system based on multi-source data fusion provided by an embodiment of the present application includes: the land survey data acquisition system based on multi-source data fusion includes: a coordinate conversion module, a coordinate deviation evaluation module and an image fusion evaluation module; wherein the coordinate conversion module is used to determine the corresponding projection coordinate band according to the coordinates collected by the projection coordinate device, and perform coordinate conversion to obtain the projection coordinate device conversion coordinates. The projection coordinate device collection coordinates include the flight route coordinates of the drone and the driving route coordinates of the ground mobile measurement device. The coordinate conversion means converting the coordinate system used for the projection coordinate device collection coordinates into a geographic coordinate system to eliminate the heterogeneity of the coordinate systems of different land survey data acquisition devices. The geographic coordinate system refers to the coordinate system used for the macroscopic land image collected by satellite remote sensing equipment. The projection coordinate device conversion coordinates include the drone conversion coordinates and the ground measurement conversion coordinates, which are used to reflect the spatial position characteristics of the land collection image. The land survey data collection The equipment includes satellite remote sensing equipment, unmanned aerial vehicles and ground mobile measuring equipment; the coordinate deviation evaluation module is used to perform coordinate deviation evaluation based on the coordinate evaluation data of the projection coordinate equipment to determine whether to optimize the coordinate evaluation data. The projection coordinate equipment includes unmanned aerial vehicles and ground mobile measuring equipment. The coordinate evaluation data includes the acquisition coordinates of the projection coordinate equipment and the conversion coordinates of the projection coordinate equipment. The coordinate evaluation data optimization represents the optimization of the coordinates of the land survey data acquisition equipment to improve the data accuracy of the coordinate evaluation data of the projection coordinate equipment; the image fusion evaluation module is used to fuse the land acquisition images of the same coordinates after coordinate conversion to obtain a land fusion image after the coordinate evaluation data optimization. The image fusion evaluation is performed based on the image fusion evaluation parameters of the land fusion image and the land acquisition image to determine whether to optimize the land acquisition image. The land acquisition images include macro land images, meso land images and micro land images. The land acquisition image optimization represents the optimization of the fusion process of the land fusion image to eliminate the resolution heterogeneity in the land acquisition image.

[0087] This application uses a high-resolution satellite signal receiving device that can receive multi-band satellite data to obtain large-scale macroscopic images of the territory and monitor changes in land use types and large-scale topographic features. The drone is equipped with an optical camera, a thermal infrared camera, and a multispectral camera, capable of acquiring high-resolution images from low altitudes and conducting detailed mapping of key areas, such as monitoring farmland planting conditions and the distribution of small buildings. During ground travel, it collects three-dimensional ground data and ground feature images in real time to supplement local details such as street facilities and small plot boundaries. Data from different sources is first preprocessed to unify the coordinate system and data format. Then, through feature extraction and matching, the macroscopic information from satellite remote sensing, the mesoscopic imagery from drone aerial photography, and the microscopic data from ground mobile measurements are integrated to generate a comprehensive and accurate land information dataset.

[0088] In this embodiment, the coordinate conversion module is used to eliminate the consistency of the coordinate systems of different devices, ensuring that the data collected by different devices are spatially docked and fused; the coordinate deviation evaluation module is used to facilitate timely discovery and correction of device coordinate errors, thereby ensuring the accuracy of the coordinates and improving the accuracy of subsequent image fusion and analysis; the image fusion evaluation module is used to eliminate the differences in resolution of images obtained by different devices, ensuring the quality of the fusion image of the land.

[0089] To sum up, the embodiment of the present application collects coordinates through a projection coordinate device and performs coordinate conversion to obtain projection coordinate device conversion coordinates, then performs coordinate deviation evaluation based on the coordinate evaluation data to optimize the coordinate evaluation data, and finally performs image fusion evaluation based on image fusion evaluation parameters to optimize the land acquisition image, thereby eliminating the heterogeneity of data collected under different coordinate systems, and further achieving improved image fusion accuracy in the collection and fusion of land survey data based on different data sources, effectively solving the problem in the existing technology that the differences in coordinate systems are not fully considered in the collection and fusion of land survey data based on different data sources.

[0090] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0091] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0092] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0093] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0094] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0095] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. The land survey data collection method based on multi-source data fusion is characterized by: The following steps are involved: S1, determining corresponding projection coordinate zones based on coordinates collected by a projection coordinate device, and performing coordinate conversion to obtain projection coordinate device conversion coordinates, wherein the projection coordinate device collection coordinates include the flight route coordinates of an unmanned aerial vehicle (UAV) and the travel route coordinates of a ground mobile measurement device, and the coordinate conversion is used to eliminate the heterogeneity of coordinate systems of different land survey data collection devices, and the projection coordinate device conversion coordinates include UAV conversion coordinates and ground measurement conversion coordinates, which are used to reflect the spatial position characteristics of land collection images, and the land survey data collection devices include satellite remote sensing equipment, UAVs, and ground mobile measurement equipment; S2, performing coordinate deviation evaluation based on coordinate evaluation data of the projection coordinate device, and determining whether to optimize the coordinate evaluation data, wherein the coordinate evaluation data includes coordinates collected by the projection coordinate device and coordinates converted by the projection coordinate device, and the coordinate evaluation data optimization indicates optimizing the coordinates of the land survey data collection device to improve the data accuracy of the coordinate evaluation data of the projection coordinate device; S3, after the coordinate evaluation data is optimized, the land acquisition images of the same coordinates after coordinate conversion are fused to obtain a land fusion image, and image fusion evaluation is performed based on the image fusion evaluation parameters of the land fusion image and the land acquisition image to determine whether to perform land acquisition image optimization. The land acquisition image optimization means optimizing the fusion process of the land fusion image to eliminate the resolution heterogeneity existing in the land acquisition image.

2. The land survey data collection method based on multi-source data fusion according to claim 1 is characterized in that: The specific process of performing coordinate conversion to obtain the projection coordinate device conversion coordinates is as follows: D1, respectively determine whether the projection coordinate bands corresponding to the coordinates collected by the projection coordinate device are unique. If they are both unique, execute D2; otherwise, execute D3; D2, performing coordinate conversion according to the coordinates collected by the projection coordinate device to obtain the projection coordinate device conversion coordinates; D3. If the projection coordinate band corresponding to the UAV's flight path coordinates is not unique, the UAV's flight path coordinates within each projection coordinate band are transformed and then merged to obtain the UAV's transformed coordinates. If the projection coordinate band corresponding to the ground mobile measurement device's driving route coordinates is not unique, the ground mobile measurement device's driving route coordinates within each projection coordinate band are transformed and then merged to obtain the ground measurement transformed coordinates.

3. The land survey data collection method based on multi-source data fusion according to claim 1 is characterized in that: The specific method for evaluating coordinate deviation based on the coordinate evaluation data of the projection coordinate device is as follows: The first coordinate deviation adjustment factor is introduced to assign a value to the result of absolute deviation processing between the initial flight coordinates of the UAV flight route and the preset initial flight coordinates obtained from the preset database to obtain the initial flight coordinate deviation coefficient; A second coordinate deviation adjustment factor is introduced to assign a value to the result of absolute deviation processing between the initial driving coordinates of the driving route of the ground mobile measurement device and the preset initial driving coordinates obtained from the preset database to obtain an initial driving coordinate deviation coefficient; The third coordinate deviation adjustment factor is introduced to assign a value to the absolute deviation processing result of the initial flight coordinates of the UAV flight route and the initial driving coordinates of the ground mobile measurement device's driving route, and the UAV-ground measurement position deviation coefficient is obtained; The first coordinate comparison coefficient is obtained by coupling processing based on the initial flight coordinate deviation coefficient, the initial driving coordinate deviation coefficient and the UAV-ground measurement position deviation coefficient; Perform absolute deviation comparison processing on the initial UAV conversion coordinates and the preset initial UAV conversion coordinates obtained from the preset database to obtain the initial UAV conversion deviation coefficient; Performing absolute deviation comparison processing on the initial geodetic conversion coordinates and the preset initial geodetic conversion coordinates obtained from a preset database to obtain an initial geodetic conversion deviation coefficient; The fourth coordinate deviation adjustment factor and the fifth coordinate deviation adjustment factor are introduced to assign the initial UAV conversion deviation coefficient and the initial ground measurement conversion deviation coefficient, and the first coordinate comparison coefficient is combined for coupling processing to obtain the coordinate deviation evaluation result. The coordinate deviation evaluation result is used to quantitatively evaluate the acquisition and conversion accuracy of the flight coordinates of the UAV flight route and the driving coordinates of the driving route of the ground mobile measurement equipment.

4. The land survey data collection method based on multi-source data fusion according to claim 3 is characterized by: The specific process of determining whether to optimize the coordinate evaluation data is as follows: A1: If the coordinate deviation evaluation result is greater than the first coordinate deviation threshold obtained from the preset database, the first coordinate evaluation data is optimized; otherwise, A2 is executed; A2: If the coordinate deviation evaluation result is greater than the second coordinate deviation threshold obtained from the preset database, the second coordinate evaluation data is optimized; The first coordinate assessment data optimization includes coordinate assessment data noise filtering, land survey data acquisition equipment position calibration and image re-acquisition; The second coordinate evaluation data optimization represents adjusting the projection coordinate device conversion coordinates to ensure data consistency.

5. The land survey data collection method based on multi-source data fusion according to claim 4 is characterized in that: The specific process of optimizing the second coordinate evaluation data is as follows: Performing square processing on the deviations of the initial UAV conversion coordinates and the preset initial UAV conversion coordinates to obtain the rotation adjustment angle and translation adjustment distance of the initial UAV conversion coordinates, and performing square processing on the deviations of the initial UAV conversion coordinates and the preset initial UAV conversion coordinates to obtain the scale adjustment factor of the initial UAV conversion coordinates; Performing square processing on the deviations of the initial geodesic conversion coordinates and the preset initial geodesic conversion coordinates to obtain a rotation adjustment angle and a translation adjustment distance of the initial geodesic conversion coordinates, and performing square processing on the deviations of the initial geodesic conversion coordinates and the preset initial geodesic conversion coordinates to obtain a scale adjustment factor of the initial geodesic conversion coordinates according to the translation adjustment distance of the initial geodesic conversion coordinates; Performing deviation comparison processing based on the coordinate deviation evaluation result and a second coordinate deviation threshold obtained from a preset database to obtain an adjustment reliability, wherein the adjustment reliability is used to quantify the accuracy of the coordinate adjustment parameters; Inputting the adjustment credibility and the coordinate adjustment parameters into the adjustment amount mapping set to obtain the coordinate adjustment amount, wherein the coordinate adjustment parameters include the rotation adjustment angle, translation adjustment distance, and scale adjustment factor of the initial UAV conversion coordinates and the rotation adjustment angle, translation adjustment distance, and scale adjustment factor of the initial ground measurement conversion coordinates; The initial UAV conversion coordinates and the initial ground measurement conversion coordinates are adjusted successively according to the coordinate adjustment amount until the termination condition of the second coordinate evaluation data optimization is reached; The termination condition for the optimization of the second coordinate evaluation data is: B1: If the coordinate deviation evaluation result after each coordinate adjustment is greater than the second coordinate deviation threshold, execute B2; otherwise, terminate the optimization of the second coordinate evaluation data; B2: If the total amount of coordinate adjustment is less than the coordinate adjustment parameter, continue to adjust the UAV conversion coordinates and the ground measurement conversion coordinates one by one, otherwise terminate the optimization of the second coordinate evaluation data.

6. The land survey data collection method based on multi-source data fusion according to claim 1 is characterized in that: The specific method for performing image fusion evaluation based on the image fusion evaluation parameters of the land fusion image and the land acquisition image is as follows: Compare the edge pixel gradient amplitude of the land acquisition image and the edge pixel gradient amplitude of the land fusion image to obtain the edge amplitude ratio coefficient; Performing an inverse proportional operation on the artifact index to obtain an inverse proportional artifact index; The information entropy of the corresponding sub-bands obtained by wavelet decomposition of all land acquisition images is coupled to obtain multi-scale fusion information entropy; The multi-scale fusion information entropy is compared with the information entropy of the corresponding sub-band obtained by wavelet decomposition in the land fusion image, and then the result of coupled averaging is processed and the inverse proportional operation is performed to obtain the multi-scale information entropy fusion comparison coefficient; An image fusion adjustment factor is introduced to assign values to the structural similarity index, edge amplitude ratio coefficient, inverse proportional artifact index and multi-scale information entropy fusion contrast coefficient, and then coupled average processing is performed to obtain the image fusion evaluation result, which is used to quantitatively evaluate the fusion effect of the national land fusion image.

7. The land survey data collection method based on multi-source data fusion according to claim 1 is characterized in that: The specific process of determining whether to perform land acquisition image optimization is as follows: C1: If the image fusion evaluation result is not greater than the preset image fusion threshold obtained from the preset database, then perform image quality optimization and then execute C2; otherwise, store the land fusion image in the land information dataset. The image quality optimization includes image denoising and image histogram equalization. C2: If the image fusion evaluation result after image quality optimization is not greater than the preset image fusion threshold obtained from the preset database, then C3 is executed after registration optimization, otherwise the land acquisition image optimization is terminated; C3, if the image fusion evaluation result after registration optimization is not greater than the preset image fusion threshold obtained from the preset database, then C4 is executed after removing the artifacts, otherwise the land acquisition image optimization is terminated. The artifact removal is performed by decomposing the land acquisition image through discrete wavelet transform, and reducing the artifacts by separating the low-frequency and high-frequency parts; C4: If the image fusion evaluation result after removing artifacts is not greater than the preset image fusion threshold obtained from the preset database, C5 is executed after the image information entropy is optimized, otherwise the land acquisition image optimization is terminated; C5, if the image fusion evaluation result after image information entropy optimization is not greater than the preset image fusion threshold obtained from the preset database, feedback is given, otherwise the land acquisition image optimization is terminated; The land acquisition image optimization includes image quality optimization, registration optimization, artifact removal and image information entropy optimization.

8. The land survey data collection method based on multi-source data fusion according to claim 7 is characterized by: The specific process of the registration optimization is as follows: The land acquisition images are coarsely registered. If the image fusion evaluation result after coarse registration is not greater than the preset image fusion threshold obtained from the preset database, the land acquisition images after coarse registration are finely registered. Otherwise, the land acquisition images are fused. Fine registration and coarse registration mean that different algorithms are used to register the land acquisition images to improve the image registration accuracy. The fusion of the land acquisition images means decomposing the land acquisition images of different resolutions into different scales through wavelet transform, and performing weighted average fusion on sub-bands of preset frequencies in combination with sub-band fusion control factors, wherein the sub-band fusion control factors are obtained by mapping the image fusion evaluation results and the structural similarity index in a preset database; The specific process of image information entropy optimization is as follows: It is determined whether the information entropy corresponding to the sub-band obtained by the wavelet transform is greater than the preset information entropy threshold obtained from the preset database. If the information entropy corresponding to the sub-band obtained by the wavelet transform is greater than the preset information entropy threshold obtained from the preset database, the corresponding sub-band is retained; otherwise, the corresponding sub-band is reconstructed by interpolation through the adjacent sub-bands of the corresponding sub-band.

9. The land survey data collection method based on multi-source data fusion according to claim 1, characterized in that: Also includes: Determine whether the registration time of the land acquisition image is greater than the preset registration time obtained from the preset database: If the registration time of the land acquisition image is greater than the preset registration time obtained from the preset database, distributed computing is used to optimize the registration time; If the registration time of the land acquisition image is not greater than the preset registration time obtained from the preset database, continue running.

10. The land survey data acquisition system based on multi-source data fusion is characterized by: include: Coordinate transformation module, coordinate deviation evaluation module and image fusion evaluation module; Among them, the coordinate conversion module is used to determine the corresponding projection coordinate belt according to the coordinates collected by the projection coordinate device, and perform coordinate conversion to obtain the projection coordinate device conversion coordinates. The projection coordinate device collection coordinates include the flight route coordinates of the drone and the driving route coordinates of the ground mobile measurement device. The coordinate conversion is used to eliminate the heterogeneity of the coordinate systems of different land survey data collection devices. The projection coordinate device conversion coordinates include the drone conversion coordinates and the ground measurement conversion coordinates, which are used to reflect the spatial position characteristics of the land collection image. The land survey data collection equipment includes satellite remote sensing equipment, drones and ground mobile measurement equipment; The coordinate deviation evaluation module is used to perform coordinate deviation evaluation based on the coordinate evaluation data of the projection coordinate device, and determine whether to optimize the coordinate evaluation data, wherein the coordinate evaluation data includes the coordinates collected by the projection coordinate device and the coordinates converted by the projection coordinate device, and the coordinate evaluation data optimization represents optimizing the coordinates of the land survey data collection device to improve the data accuracy of the coordinate evaluation data of the projection coordinate device; The image fusion evaluation module is used to optimize the coordinate evaluation data, fuse the land acquisition images of the same coordinates after coordinate conversion to obtain a land fusion image, and perform image fusion evaluation based on the image fusion evaluation parameters of the land fusion image and the land acquisition image to determine whether to optimize the land acquisition image. The land acquisition image optimization means optimizing the fusion process of the land fusion image to eliminate the resolution heterogeneity in the land acquisition image.

Images