Geographic surveying and mapping method and system based on unmanned aerial vehicle

Abstract

The invention discloses a geographic surveying and mapping method and system based on an unmanned aerial vehicle, and relates to the technical field of geographic surveying and mapping. Performing region division on the satellite image to obtain a plurality of local regions; for each local area, performing area division on the local area through a Voronoi graph to obtain a plurality of target sub-areas, and determining the route plan of the unmanned aerial vehicle in each target sub-area; positioning each unmanned aerial vehicle according to the satellite image, acquiring an image acquired by each unmanned aerial vehicle after positioning, and splicing all the acquired images to obtain a local graph; and fusing all the local graphs to the satellite image to obtain a surveying and mapping graph corresponding to the target area. According to the method, local area division is carried out on the satellite image, and then area division is carried out through the Voronoi graph, so that each unmanned aerial vehicle is only responsible for one target sub-area, the overall surveying and mapping speed is improved, splicing of local images and fusion of the satellite image are improved, the accuracy and consistency of final results are ensured, and the accuracy and efficiency of the overall surveying and mapping result are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of geographic surveying and mapping, and in particular relates to a geographic surveying and mapping method and system based on an unmanned aerial vehicle. Background Art

[0002] With the rapid development of geographic information science and remote sensing technology, traditional geographic surveying and mapping methods mainly rely on manual measurement and ground equipment. These methods are not only inefficient, but also often limited by the working environment, with limited measurement accuracy and range. Although current surveying and mapping methods have introduced new technologies such as satellite positioning systems, lidar, and remote sensing images, they still have problems such as insufficient measurement accuracy and poor adaptability to complex terrain.

[0003] With the development of drone technology, drones have been widely used in agriculture, environmental monitoring, geographic surveying and mapping and other fields. Using drones for surveying and mapping has the advantages of high efficiency, accuracy and flexibility. It can collect geographic data efficiently in complex terrain and dangerous environments. Existing technologies have begun to use satellites plus drones to achieve geographic surveying and mapping.

[0004] However, existing technologies usually use a single drone to collect images. In this case, due to the unified surveying and mapping standards of drones, data is easier to align and fuse, but this will also greatly increase the surveying and mapping time, so multiple drones are required to work together. However, when multiple drones collect images, due to inconsistent standards, the accuracy of the final surveying and mapping results is affected. Summary of the invention

[0005] The purpose of the present invention is to solve the above-mentioned problems and to propose a geographic surveying and mapping method and system based on drones.

[0006] In a first aspect of the present invention, a geographic surveying and mapping method based on an unmanned aerial vehicle is first proposed, the method comprising: Acquire a satellite image of a target area, and divide the satellite image into multiple local areas; For each local area, the Voronoi diagram is used to divide the local area into multiple target sub-areas, and the route planning of the UAV in each target sub-area is determined; Positioning each drone according to the satellite image, and obtaining images collected by each drone after positioning; For each image collected by the drone, all collected images are stitched together to obtain a local image; All local images are fused with the satellite image to obtain a surveying map corresponding to the target area.

[0007] Optionally, dividing the satellite image into multiple local areas includes: Preprocessing the satellite image to obtain an initial satellite image, and segmenting the initial satellite image according to a preset grid to obtain a sub-grid image set; For each sub-grid image in the sub-grid image set, substituting the sub-grid image into a preset model to extract the image type and obtain a type label; Adjacent sub-grid images are fused according to type labels to obtain multiple local regions.

[0008] Optionally, positioning each drone according to the satellite image includes: Obtain an image set collected by the drone at the starting point of route planning, and substitute a feature point detection model for each image in the image set to extract feature points to obtain a plurality of first feature point sets; Substituting the satellite image into the feature point detection model to extract feature points to obtain a second feature point set; For each first feature point set, performing feature point matching between the first feature point set and the second feature point set by using a feature matching algorithm; If the feature point matching result is a successful match, the coordinate space of the image captured by the drone is converted to the coordinate space of the satellite image based on the matching point pair to obtain the corresponding position of the drone in the satellite image; the matching point pair is the first feature point and the second feature point that are successfully matched.

[0009] Optionally, for each image collected by the drone, all collected images are stitched together to obtain a local image including: Obtain the images collected by each UAV in time series, and perform binarization processing on each image to obtain a binary image set; For each image in the binary image set, geometric analysis is performed on the image to obtain a boundary contour line; Aligning the binary images adjacent to the image according to the boundary contour lines, repeating this step until all the binary images in the binary image set are aligned to obtain an initial local image region; The initial local image is dynamically spliced ​​and adjusted according to a preset region of interest to obtain a local image region, a valid region corresponding to each acquired image is determined according to the local image region, and all valid regions are spliced ​​to obtain a local image.

[0010] Optionally, fusing all local images into the satellite image to obtain a surveying map corresponding to the target area includes: Extracting the corresponding area in the satellite image according to the valid areas corresponding to all the acquired images to obtain an effective area image pair; the effective area image pair includes the acquired image and the satellite image; For each valid region image pair, performing Gaussian pyramid decomposition on the valid region image pair to obtain a first Gaussian image set and a second Gaussian image set; For the first Gaussian image set, differences between Gaussian images at adjacent levels are calculated to obtain a first Laplacian pyramid; for the second Gaussian image set, differences between Gaussian images at adjacent levels are calculated to obtain a second Laplacian pyramid; For the first Laplacian pyramid and the second Laplacian pyramid, weighted fusion is performed on features at the same level to obtain a fused Laplacian pyramid; Image reconstruction is performed according to the fused Laplacian pyramid to obtain a local survey map, and all local survey maps are acquired to obtain a survey map corresponding to the target area.

[0011] In a second aspect of the present invention, a geographic surveying and mapping system based on an unmanned aerial vehicle is provided, comprising: A local area determination module is used to obtain a satellite image of a target area and divide the satellite image into multiple local areas; The route planning module is used to divide each local area into multiple target sub-areas by using the Voronoi diagram, and determine the route planning of the UAV in each target sub-area; A UAV positioning module is used to locate each UAV according to the satellite image and obtain the image collected by each UAV after positioning; An image stitching module is used to stitch all the images collected by each drone to obtain a local image; The surveying and mapping map determination module is used to fuse all local maps into the satellite image to obtain a surveying and mapping map corresponding to the target area.

[0012] Optionally, the local area determination module includes: A preprocessing module, used for preprocessing the satellite image to obtain an initial satellite image, and segmenting the initial satellite image according to a preset grid to obtain a sub-grid image set; A type extraction module, for each sub-grid image in the sub-grid image set, substituting the sub-grid image into a preset model to extract the image type and obtain a type label; The grid image fusion module is used to fuse adjacent sub-grid images according to type labels to obtain multiple local areas.

[0013] Optionally, the drone positioning module includes: A first feature point extraction module is used to obtain an image set collected by the drone at the starting point of the route planning, and substitute a feature point detection model into each image in the image set to extract feature points to obtain multiple first feature point sets; A second feature point extraction module, used for substituting the satellite image into the feature point detection model to extract feature points to obtain a second feature point set; A feature point matching module, used for performing feature point matching on each first feature point set and the second feature point set by using a feature matching algorithm; The drone position determination module is used to convert the coordinate space of the drone-captured image into the coordinate space of the satellite image based on the matching point pair if the feature point matching result is a successful match; the matching point pair is the first feature point and the second feature point that are successfully matched.

[0014] Optionally, the image stitching module includes: A binary processing module is used to obtain images collected by each UAV in a time series, and perform a binary processing on each image to obtain a binary image set; A geometric analysis module, for performing geometric analysis on each image in the binary image set to obtain a boundary contour line; An image alignment module is used to align the binary images adjacent to the image according to the boundary contour lines, and repeat this step until all the binary images in the binary image set are aligned to obtain an initial local image region; The effective area stitching module is used to dynamically stitch and adjust the initial local image according to the preset area of ​​interest to obtain a local image area, determine the effective area corresponding to each collected image according to the local image area, and stitch all the effective areas to obtain a local image.

[0015] Optionally, the surveying and mapping determination module includes: An effective area image pair determination module is used to extract the corresponding area in the satellite image according to the effective areas corresponding to all the acquired images to obtain an effective area image pair; the effective area image pair includes an acquired image and a satellite image; A Gaussian pyramid decomposition module is used to perform Gaussian pyramid decomposition on each valid region image pair to obtain a first Gaussian image set and a second Gaussian image set; a Laplacian pyramid calculation module, configured to calculate the difference between Gaussian images at adjacent levels for the first Gaussian image set to obtain a first Laplacian pyramid, and to calculate the difference between Gaussian images at adjacent levels for the second Gaussian image set to obtain a second Laplacian pyramid; A feature fusion module, configured to perform weighted fusion on features at the same level of the first Laplacian pyramid and the second Laplacian pyramid to obtain a fused Laplacian pyramid; An image reconstruction module is used to perform image reconstruction according to the fused Laplacian pyramid to obtain a local survey map, and obtain all local survey maps to obtain a survey map corresponding to the target area.

[0016] Beneficial effects of the present invention: The invention proposes a geographic surveying and mapping method based on a drone, which obtains a satellite image of a target area, divides the satellite image into regions to obtain a plurality of local regions; for each local region, divides the local region into regions through a Voronoi diagram to obtain a plurality of target sub-regions, and determines the route planning of the drone in each target sub-region; locates each drone according to the satellite image, and obtains the image collected by each drone after positioning; for the image collected by each drone, splices all the collected images to obtain a local map; and fuses all the local maps to the satellite image to obtain a surveying and mapping map corresponding to the target area. By dividing the satellite image into local regions, determining which areas to use what type of drone according to the category, the surveying and mapping efficiency is improved, and then dividing the regions through the Voronoi diagram so that each drone is responsible for only one target sub-region, thereby improving the operation efficiency, and avoiding the drone from performing repeated operations on the same area, improving the overall surveying and mapping speed, and then by accurately locating the drone before collecting data and planning the flight path, the flight synchronization error can be reduced, and the alignment accuracy of the data can be improved. Finally, by splicing the local images and fusing the satellite images, the accuracy and consistency of the final result are ensured, and the accuracy and efficiency of the overall surveying and mapping result are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The present invention will be further described below in conjunction with the accompanying drawings.

[0018] Figure 1 A flowchart of a geographic surveying and mapping method based on a drone is provided for an embodiment of the present invention; Figure 2 A framework diagram of a geographic mapping system based on a drone is provided for an embodiment of the present invention. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. The term "and / or" herein is only a description of the association relationship of the associated objects, indicating that there can be three relationships. For example, A and B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, in the present invention, the description of "first", "second", etc. is only used for descriptive purposes, and cannot be understood as indicating or implying its relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" can explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but it must be based on the ability of ordinary technicians in the field to implement. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0020] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative work shall fall within the scope of protection of the present invention.

[0021] The embodiment of the present invention provides a geographic surveying and mapping method based on an unmanned aerial vehicle. Figure 1 , Figure 1 A flowchart of a geographic mapping method based on an unmanned aerial vehicle is provided in an embodiment of the present invention. The method comprises the following steps: S101, obtaining a satellite image of a target area, and dividing the satellite image into multiple local areas; S102, for each local area, dividing the local area into multiple target sub-areas through the Voronoi diagram, and determining the route planning of the drone in each target sub-area; S103, positioning each UAV according to the satellite image, and obtaining images collected by each UAV after positioning; S104, for each image collected by the drone, stitching all collected images to obtain a local image; S105, fusing all local images with the satellite image to obtain a surveying map corresponding to the target area.

[0022] A geographic mapping method based on a drone is provided according to an embodiment of the present invention. By dividing satellite images into local areas, it is determined according to categories which areas should use which type of drones, thereby improving the mapping efficiency. Then, the Voronoi diagram is used to divide the areas so that each drone is responsible for only one target sub-area, thereby improving the work efficiency and avoiding repeated operations of the drone in the same area, thereby improving the overall mapping speed. Then, by accurately positioning the drone before collecting data and planning the flight path, the flight synchronization error can be reduced and the alignment accuracy of the data can be improved. Finally, by splicing the local images and fusing them with the satellite images, the accuracy and consistency of the final results are ensured, thereby improving the accuracy and efficiency of the overall mapping results.

[0023] In one implementation, satellite images can be used to quickly obtain basic information about the target area over a large area, reducing the time and cost of manual measurement. Through the regional division of the Voronoi diagram, the working range and path planning of each drone can be accurately determined to ensure that the drones do not interfere with each other when performing tasks, thereby improving the efficiency and accuracy of surveying and mapping operations. By accurately positioning each drone, the accuracy of image acquisition can be ensured, and areas that are not captured in detail in satellite images can be covered.

[0024] In one implementation, multiple drones perform tasks in different sub-areas, and multiple areas can be scanned in detail at the same time, reducing the overall surveying and mapping time. The local images collected by the drones can be stitched and fused to obtain higher resolution and more perspective information. By fusing multiple local images into satellite images, more detailed and comprehensive surveying and mapping can be obtained.

[0025] In one implementation, the Voronoi diagram is used for region division, which is a classic geometric segmentation method that can effectively avoid overlapping operations between drones. The division of each sub-region enables each drone to be responsible for only one target sub-region, thereby improving operation efficiency and avoiding interference between drones. It helps to improve the planning accuracy of the drone flight path and ensure that each drone can complete the image acquisition task according to the predetermined route.

[0026] In one implementation, in order to ensure the accuracy of the shooting when the drone is taking pictures, the images taken often contain overlapping parts. Therefore, the stitching algorithm needs to process the data of these overlapping areas to ensure that the final stitched images are accurate and improve the quality of the final survey map.

[0027] In one implementation, before each drone collects images, positioning the drone through satellite images is a key step in solving positioning errors. High-precision positioning can ensure that the drone can accurately cover the target area during the mission execution, avoid error accumulation, and reduce data fusion errors caused by positioning problems.

[0028] In one embodiment, dividing the satellite image into multiple local areas includes: Preprocessing the satellite image to obtain an initial satellite image, and segmenting the initial satellite image according to a preset grid to obtain a sub-grid image set; For each sub-grid image in the sub-grid image set, substitute the sub-grid image into a preset model to extract the image type and obtain a type label; Adjacent sub-grid images are fused according to type labels to obtain multiple local regions.

[0029] In one implementation, the satellite image is preprocessed to remove noise and irrelevant information, and then the image is segmented according to a preset grid, so that the image processing is more systematic and standardized. The size and resolution of each sub-grid can be adjusted according to specific needs to balance processing accuracy and efficiency; preprocessing is a common satellite image processing method such as denoising and image enhancement to enhance image clarity; the preset grid size is determined by the technician.

[0030] In one implementation, the image types include building areas, road areas, water areas and vegetation areas, among which building areas and road areas are defined as first type labels, and water areas and vegetation areas are defined as second type labels; the area scenes corresponding to the first type labels are often more complex, so multi-rotor drones are used, and the area scenes corresponding to the second type labels are basically water and vegetation areas, and fixed-wing drones are usually simply used; the preset model can be U-Net, DeepLabV3+, SegNet and other common area type recognition models on the market.

[0031] In one implementation, by identifying type labels of adjacent sub-grid images, regional fusion can be performed more intelligently. For example, adjacent sub-grids of different types, such as buildings and roads, can be identified and merged into a large local area to facilitate more efficient subsequent analysis and application, providing a smoother and more reasonable regional division. By reasonably merging adjacent sub-grids, a local area that is more in line with the actual geographical environment can be obtained.

[0032] In one embodiment, locating each drone based on satellite images includes: Obtain an image set collected by the drone at the starting point of route planning, and substitute a feature point detection model into each image in the image set to extract feature points to obtain a plurality of first feature point sets; Substituting the satellite image into the feature point detection model to extract feature points to obtain a second feature point set; For each first feature point set, performing feature point matching on the first feature point set and the second feature point set by using a feature matching algorithm; If the feature point matching result is a successful match, the coordinate space of the image collected by the drone is converted to the coordinate space of the satellite image based on the matching point pair to obtain the corresponding position of the drone in the satellite image; the matching point pair is the first feature point and the second feature point that are successfully matched.

[0033] In one implementation, by matching the image captured by the drone with the feature points in the satellite image, the image coordinates of the drone can be converted to the coordinate system of the satellite image, thereby achieving high-precision positioning and providing a basis for subsequent image stitching and fusion; the matching process of feature points can be automated to reduce human intervention, especially in complex and dangerous environments. The drone can autonomously locate and adjust the path during the mission execution, improving the automation level of mission completion.

[0034] In one implementation, before each image in the image set is substituted into the feature point detection model and the satellite image is substituted into the feature point detection model, the scale of the satellite image and each image in the image set is standardized. In order to ensure accuracy, the size of each image in the image set is usually used as the standard.

[0035] In one implementation, the image set collected by the drone at the starting point of route planning is the image collected by the drone at the starting point of the route planning. The drone can realize the coordinate mapping with the satellite image based on these collected images. The feature point detection model can be SIFT, SURF, ORB, etc. The feature matching algorithm can be FLANN, SuperGlue, RANSAC, etc. The feature point matching is based on the feature points of each image in the image set. For example, when there is an image with 100 feature points in the image set, then the feature points are matched with the feature points in the satellite image. If 80 feature points are successfully matched, there is a threshold for successful matching. The threshold is the proportion of successful matching of all feature points in the current image. If the proportion is greater than or equal to 80%, the coordinate space of this image can be converted to the coordinate space of the satellite image. For each unmanned driving area, a coordinate system will be established with the starting point as the origin and the initial driving direction as the positive direction of the Y axis. The coordinate space of the satellite image is a coordinate system established based on the satellite image. When the corresponding position of the drone in the satellite image is determined, the corresponding area of ​​the drone-shot image in the satellite image can be determined according to the driving path of the drone, providing a basis for subsequent image fusion.

[0036] In one embodiment, for each image collected by the drone, stitching all collected images to obtain a local image includes: Obtain the images collected by each UAV in time series, and perform binarization processing on each image to obtain a binary image set; For each image in the binary image set, a geometric analysis is performed on the image to obtain a boundary contour line; Align the binary images adjacent to the image according to the boundary contour lines, and repeat this step until all binary images in the binary image set are aligned to obtain the initial local image area; The initial local image is dynamically spliced ​​and adjusted according to the preset region of interest to obtain a local image region, the valid region corresponding to each acquired image is determined according to the local image region, and all valid regions are spliced ​​to obtain a local image.

[0037] In one implementation, binarization can simplify the complexity of image processing and convert the image into only black and white, thereby removing noise, background and irrelevant information in the image. Through binarization, key target areas such as object boundaries and contours are more prominent, which is helpful for subsequent image analysis and feature extraction; geometric analysis can extract edges and contours in the image. The extraction of boundary contour lines helps to understand the shape characteristics of objects or scenes in the image, and can provide necessary structural information for subsequent alignment and stitching, ensuring that the edges, angles and other features between images can be effectively matched.

[0038] In one implementation method, the valid area corresponding to each acquired image is determined according to the local image area. Specifically, the overlapping part of each acquired image and other images after dynamic stitching adjustment is obtained, and the proportion of the overlapping part in the current image is determined. If the proportion exceeds a preset ratio, the overlapping part in the current image is removed, and the remaining part is recorded as the valid area. If the proportion exceeds the preset ratio, the overlapping part is retained, and the retained image area is recorded as the valid area.

[0039] In one implementation, images are aligned by boundary contours to ensure that adjacent images are accurately aligned in space, thereby avoiding stitching errors and improving the quality and accuracy of image stitching; dynamic stitching adjustment can optimize local images according to a preset region of interest (ROI). The region of interest usually refers to the area in the image that needs to be focused on. Dynamic adjustment can avoid interference from irrelevant areas, ensuring that the stitched local image area accurately reflects the details of the target area, while reducing distortion caused by stitching.

[0040] In one implementation, geometric analysis is performed on the image to obtain boundary contour lines, specifically by extracting contours from areas with the same pixel values ​​in the image. Each contour is a closed path composed of multiple points, indicating the boundary of an object in the image. In this way, the contours of the object can be identified from the image and used as contour lines for subsequent processing. The extracted contours can be multiple discrete point sets, indicating the boundary of an object or area. The contours separate different areas in the image. Because the satellite images are often supplemented by drones for mapping and mapping, the objects usually supplemented are some regular graphics. The preset area of ​​interest is a place that needs to be identified in detail in the satellite image, such as a house building area. The specific preset area of ​​interest is determined by a technician. The dynamic splicing is adjusted according to the overlapping area. If the image overlapping area is larger than the preset first area, a wider area of ​​interest is usually required. At this time, the width of the original area of ​​interest will be changed to 1.2 times the original, so as to better smooth the exposure difference and the color tone change. If the image overlapping area is smaller than the preset second area, the width of the area of ​​interest is usually required to be reduced. At this time, the width of the original area of ​​interest will be changed to 0.8 times the original, so as to improve the splicing efficiency and reduce unnecessary calculations.

[0041] In one implementation, multiple images can be combined into a larger, more detailed image by stitching together local image regions, and image data from multiple time points can be used to obtain more comprehensive regional information, thereby forming a seamless overall image that can provide higher accuracy and coverage.

[0042] In one embodiment, fusing all local images into satellite images to obtain a surveying map corresponding to the target area includes: Extracting corresponding areas in satellite images according to valid areas corresponding to all acquired images to obtain valid area image pairs; the valid area image pairs include acquired images and satellite images; For each valid region image pair, performing Gaussian pyramid decomposition on the valid region image pair to obtain a first Gaussian image set and a second Gaussian image set; For the first Gaussian image set, the difference between Gaussian images at adjacent levels is calculated to obtain a first Laplacian pyramid, and for the second Gaussian image set, the difference between Gaussian images at adjacent levels is calculated to obtain a second Laplacian pyramid; For the first Laplacian pyramid and the second Laplacian pyramid, weighted fusion is performed on the features at the same level to obtain a fused Laplacian pyramid; The local survey map is obtained by image reconstruction based on the fused Laplacian pyramid, and all the local survey maps are obtained to obtain the survey map corresponding to the target area.

[0043] In one implementation, the image can be decomposed into different spatial frequency bands from low frequency to high frequency by performing Gaussian pyramid and Laplacian pyramid decomposition on the image, so that when fusion is performed at different levels, low-frequency structural information such as large-scale background and geographical features can be better retained and high-frequency details such as texture and feature points can be highlighted. It can flexibly perform adaptive processing according to the characteristics of different images such as the resolution and exposure differences of drone images and satellite images, avoiding detail loss or over-smoothing caused by direct weighted averaging.

[0044] In one implementation, each layer of the Laplacian pyramid includes a low-frequency layer and a high-frequency layer, wherein the Laplacian pyramid obtained by the Gaussian pyramid of the first preset number of layers is the high-frequency layer, and the Laplacian pyramid obtained by the Gaussian pyramid after the preset number of layers is the low-frequency layer, and the weight corresponding to the first Laplacian pyramid of the Laplacian pyramid of the low-frequency layer is 0.7, and the weight corresponding to the second Laplacian pyramid is 0.3; the weight corresponding to the first Laplacian pyramid of the Laplacian pyramid of the high-frequency layer is 0.2, and the weight corresponding to the second Laplacian pyramid is 0.8.

[0045] In one implementation, satellite images usually contain a larger field of view but have a lower resolution; while drone images provide a higher resolution but cover a smaller area. By calculating the difference between multi-resolution pyramid decomposition and Laplacian pyramid, local details of the image can be effectively extracted and enhanced, making the features of the target area clearer and helping to improve the matching accuracy of the image.

[0046] In one implementation, the local survey map is obtained by image reconstruction based on the fused Laplacian pyramid, specifically, starting from the bottom layer of the fused Laplacian pyramid and reconstructing upward layer by layer, upsampling each layer of the image to expand it to the same size as the previous layer, and then superimposing the two images until the top layer image is finally obtained.

[0047] In one implementation, during the image fusion process, through the layer-by-layer fusion of the Laplacian pyramid, a smooth transition can be achieved at the seams, avoiding obvious splicing marks due to brightness or texture differences between different images, making the final generated local survey map more natural and seamless, especially when the drone image and the satellite image are spliced, the sense of incongruity caused by the image seams can be effectively eliminated; by decomposing the image into multiple local areas and processing them layer by layer, the difficulty of processing large images can be effectively reduced, and by performing Gaussian pyramid decomposition on each effective area image pair, the features of each local area can be balanced and retained at the low-frequency and high-frequency levels, and the final generated local survey map has higher quality. After the local survey maps are merged, the survey map of the overall target area is obtained, which can not only maintain accuracy but also process larger geographical areas.

[0048] Based on the same inventive concept, the embodiment of the present invention also provides a geographic surveying and mapping system based on a drone. Figure 2 , Figure 2 A framework diagram of a geographic surveying and mapping system based on an unmanned aerial vehicle provided in an embodiment of the present invention includes: A local area determination module is used to obtain a satellite image of a target area and divide the satellite image into multiple local areas; The route planning module is used to divide each local area into multiple target sub-areas by using the Voronoi diagram, and determine the route planning of the UAV in each target sub-area; The UAV positioning module is used to locate each UAV based on satellite images and obtain the images collected by each UAV after positioning; An image stitching module is used to stitch all the images collected by each drone to obtain a local image; The surveying and mapping map determination module is used to fuse all local maps into the satellite image to obtain the surveying and mapping map corresponding to the target area.

[0049] A geographic mapping system based on a drone provided by an embodiment of the present invention improves the mapping efficiency by dividing satellite images into local areas and determining which areas to use which type of drones according to categories. The Voronoi diagram is then used to divide the areas so that each drone is responsible for only one target sub-area, thereby improving the operating efficiency and avoiding repeated operations of the drone on the same area, thereby improving the overall mapping speed. The precise positioning of the drone before data collection and flight path planning can reduce flight synchronization errors and improve data alignment accuracy. Finally, the local image stitching and satellite image fusion can ensure the accuracy and consistency of the final result, thereby improving the accuracy and efficiency of the overall mapping result.

[0050] In one embodiment, the local area determination module includes: A preprocessing module is used to preprocess the satellite image to obtain an initial satellite image, and to segment the initial satellite image according to a preset grid to obtain a sub-grid image set; A type extraction module is used to substitute each sub-grid image in the sub-grid image set into a preset model to extract the image type and obtain a type label; The grid image fusion module is used to fuse adjacent sub-grid images according to type labels to obtain multiple local areas.

[0051] In one embodiment, the drone positioning module includes: A first feature point extraction module is used to obtain an image set collected by the drone at the starting point of the route planning, and substitute a feature point detection model for each image in the image set to extract feature points to obtain multiple first feature point sets; A second feature point extraction module is used to substitute the satellite image into the feature point detection model to extract feature points to obtain a second feature point set; A feature point matching module, used for performing feature point matching between the first feature point set and the second feature point set by using a feature matching algorithm for each first feature point set; The drone position determination module is used to convert the coordinate space of the drone-captured image into the coordinate space of the satellite image based on the matching point pair if the feature point matching result is a successful match; the matching point pair is the first feature point and the second feature point that are successfully matched.

[0052] In one embodiment, the image stitching module includes: A binary processing module is used to obtain images collected by each UAV in a time series, and perform a binary processing on each image to obtain a binary image set; A geometric analysis module, for performing geometric analysis on each image in the binary image set to obtain a boundary contour line; An image alignment module is used to align the binary images adjacent to the image according to the boundary contour lines, and repeat this step until all the binary images in the binary image set are aligned to obtain an initial local image region; The effective area stitching module is used to dynamically stitch and adjust the initial local image according to the preset area of ​​interest to obtain the local image area, determine the effective area corresponding to each collected image according to the local image area, and stitch all the effective areas to obtain the local image.

[0053] In one embodiment, the map determination module includes: An effective area image pair determination module is used to extract the corresponding area in the satellite image according to the effective areas corresponding to all the acquired images to obtain an effective area image pair; the effective area image pair includes the acquired image and the satellite image; A Gaussian pyramid decomposition module is used to perform Gaussian pyramid decomposition on each valid region image pair to obtain a first Gaussian image set and a second Gaussian image set; A Laplacian pyramid calculation module, used for calculating the difference between Gaussian images at adjacent levels for the first Gaussian image set to obtain a first Laplacian pyramid, and for calculating the difference between Gaussian images at adjacent levels for the second Gaussian image set to obtain a second Laplacian pyramid; A feature fusion module, used for performing weighted fusion on features at the same level of the first Laplacian pyramid and the second Laplacian pyramid to obtain a fused Laplacian pyramid; The image reconstruction module is used to reconstruct the image according to the fused Laplace pyramid to obtain a local survey map, and obtain all the local survey maps to obtain a survey map corresponding to the target area.

[0054] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A geographic surveying and mapping method based on an unmanned aerial vehicle, characterized in that: The method comprises: Acquire a satellite image of a target area, and divide the satellite image into multiple local areas; For each local area, the Voronoi diagram is used to divide the local area into multiple target sub-areas, and the route planning of the UAV in each target sub-area is determined; Positioning each drone according to the satellite image, and obtaining images collected by each drone after positioning; For each image collected by the drone, all collected images are stitched together to obtain a local image; All local images are fused with the satellite image to obtain a surveying map corresponding to the target area.

2. A geographic surveying and mapping method based on an unmanned aerial vehicle according to claim 1, characterized in that: The satellite image is divided into multiple local areas including: Preprocessing the satellite image to obtain an initial satellite image, and segmenting the initial satellite image according to a preset grid to obtain a sub-grid image set; For each sub-grid image in the sub-grid image set, substituting the sub-grid image into a preset model to extract the image type and obtain a type label; Adjacent sub-grid images are fused according to type labels to obtain multiple local regions.

3. A geographic surveying and mapping method based on an unmanned aerial vehicle according to claim 1, characterized in that: Positioning each drone based on the satellite image includes: Obtain an image set collected by the drone at the starting point of route planning, and substitute a feature point detection model for each image in the image set to extract feature points to obtain a plurality of first feature point sets; Substituting the satellite image into the feature point detection model to extract feature points to obtain a second feature point set; For each first feature point set, performing feature point matching between the first feature point set and the second feature point set by using a feature matching algorithm; If the feature point matching result is a successful match, the coordinate space of the image captured by the drone is converted to the coordinate space of the satellite image based on the matching point pair to obtain the corresponding position of the drone in the satellite image; the matching point pair is the first feature point and the second feature point that are successfully matched.

4. The method for geographical surveying and mapping based on an unmanned aerial vehicle according to claim 1, characterized in that: For each image collected by the drone, all the collected images are stitched together to obtain a local map including: Obtain the images collected by each UAV in time series, and perform binarization processing on each image to obtain a binary image set; For each image in the binary image set, geometric analysis is performed on the image to obtain a boundary contour line; Aligning the binary images adjacent to the image according to the boundary contour lines, repeating this step until all the binary images in the binary image set are aligned to obtain an initial local image region; The initial local image is dynamically spliced ​​and adjusted according to a preset region of interest to obtain a local image region, a valid region corresponding to each acquired image is determined according to the local image region, and all valid regions are spliced ​​to obtain a local image.

5. A geographic surveying and mapping method based on an unmanned aerial vehicle according to claim 4, characterized in that: The mapping map corresponding to the target area is obtained by fusing all local images into the satellite image, including: Extracting the corresponding area in the satellite image according to the valid areas corresponding to all the acquired images to obtain an effective area image pair; the effective area image pair includes the acquired image and the satellite image; For each valid region image pair, performing Gaussian pyramid decomposition on the valid region image pair to obtain a first Gaussian image set and a second Gaussian image set; For the first Gaussian image set, differences between Gaussian images at adjacent levels are calculated to obtain a first Laplacian pyramid; for the second Gaussian image set, differences between Gaussian images at adjacent levels are calculated to obtain a second Laplacian pyramid; For the first Laplacian pyramid and the second Laplacian pyramid, weighted fusion is performed on features at the same level to obtain a fused Laplacian pyramid; Image reconstruction is performed according to the fused Laplacian pyramid to obtain a local survey map, and all local survey maps are acquired to obtain a survey map corresponding to the target area.

6. A geographic surveying and mapping system based on drones, characterized in that: The system comprises: A local area determination module is used to obtain a satellite image of a target area and divide the satellite image into multiple local areas; The route planning module is used to divide each local area into multiple target sub-areas by using the Voronoi diagram, and determine the route planning of the UAV in each target sub-area; A UAV positioning module is used to locate each UAV according to the satellite image and obtain the image collected by each UAV after positioning; An image stitching module is used to stitch all the images collected by each drone to obtain a local image; The surveying and mapping map determination module is used to fuse all local maps into the satellite image to obtain a surveying and mapping map corresponding to the target area.

7. A geographic surveying and mapping system based on an unmanned aerial vehicle according to claim 6, characterized in that: The local area determination module comprises: A preprocessing module, used for preprocessing the satellite image to obtain an initial satellite image, and segmenting the initial satellite image according to a preset grid to obtain a sub-grid image set; A type extraction module, for each sub-grid image in the sub-grid image set, substituting the sub-grid image into a preset model to extract the image type and obtain a type label; The grid image fusion module is used to fuse adjacent sub-grid images according to type labels to obtain multiple local areas.

8. The geographic surveying and mapping system based on an unmanned aerial vehicle according to claim 6, characterized in that: The UAV positioning module includes: A first feature point extraction module is used to obtain an image set collected by the drone at the starting point of the route planning, and substitute a feature point detection model into each image in the image set to extract feature points to obtain multiple first feature point sets; A second feature point extraction module, used for substituting the satellite image into the feature point detection model to extract feature points to obtain a second feature point set; A feature point matching module, used for performing feature point matching on each first feature point set and the second feature point set by using a feature matching algorithm; The drone position determination module is used to convert the coordinate space of the drone-captured image into the coordinate space of the satellite image based on the matching point pair if the feature point matching result is a successful match; the matching point pair is the first feature point and the second feature point that are successfully matched.

9. The geographic surveying and mapping system based on an unmanned aerial vehicle according to claim 6, characterized in that: The image stitching module comprises: A binary processing module is used to obtain images collected by each UAV in a time series, and perform a binary processing on each image to obtain a binary image set; A geometric analysis module, for performing geometric analysis on each image in the binary image set to obtain a boundary contour line; An image alignment module is used to align the binary images adjacent to the image according to the boundary contour lines, and repeat this step until all the binary images in the binary image set are aligned to obtain an initial local image region; The effective area stitching module is used to dynamically stitch and adjust the initial local image according to the preset area of ​​interest to obtain a local image area, determine the effective area corresponding to each collected image according to the local image area, and stitch all the effective areas to obtain a local image.

10. A geographic surveying and mapping system based on an unmanned aerial vehicle according to claim 9, characterized in that: The surveying and mapping determination module comprises: An effective area image pair determination module is used to extract the corresponding area in the satellite image according to the effective areas corresponding to all the acquired images to obtain an effective area image pair; the effective area image pair includes an acquired image and a satellite image; A Gaussian pyramid decomposition module is used to perform Gaussian pyramid decomposition on each valid region image pair to obtain a first Gaussian image set and a second Gaussian image set; a Laplacian pyramid calculation module, configured to calculate the difference between Gaussian images at adjacent levels for the first Gaussian image set to obtain a first Laplacian pyramid, and to calculate the difference between Gaussian images at adjacent levels for the second Gaussian image set to obtain a second Laplacian pyramid; A feature fusion module, configured to perform weighted fusion on features at the same level of the first Laplacian pyramid and the second Laplacian pyramid to obtain a fused Laplacian pyramid; An image reconstruction module is used to perform image reconstruction according to the fused Laplacian pyramid to obtain a local survey map, and obtain all local survey maps to obtain a survey map corresponding to the target area.

Images