High-precision digital orthographic image rapid generation method based on unmanned aerial vehicle aerial photograph

Abstract

This application discloses a method for rapidly generating high-precision digital orthophotos based on UAV aerial images, belonging to the field of UAV image processing technology. The method includes: acquiring multiple UAV aerial images and corresponding flight parameters; adaptively dividing the flight area into multiple blocks based on its geographical extent to obtain multiple block region images; extracting feature point sets from each aerial image using the ORB algorithm; calculating the pixel coordinates of the feature point sets corresponding to their respective block region images; combining flight parameters to calculate the geographical coordinates of the feature points; establishing a one-to-one correspondence between pixel coordinates and geographical coordinates, which serve as ground control points; registering and correcting each region image using the ground control points to generate standardized region images with geographical coordinates; stitching together all standardized region images to obtain a complete initial image; automatically removing imperfections at the stitching points in the complete initial image using a multi-scale fusion algorithm; and outputting a high-precision digital orthophoto, thus improving the accuracy of digital orthophoto generation.

Description

Technical Field

[0001] This application belongs to the field of UAV image processing technology, and in particular relates to a method for rapid generation of high-precision digital orthophotos based on UAV aerial photographs. Background Technology

[0002] Digital orthophotos are remote sensing images with orthophoto projection properties, capable of accurately reflecting the planar position and shape of ground features. They play an irreplaceable role in fields such as land resource surveys, urban planning and management, and emergency disaster relief. Due to its high flexibility, ease of operation, and low cost, drone aerial photography technology has become one of the main means of acquiring high-resolution image data.

[0003] However, existing UAV orthophoto generation technologies still have many shortcomings: First, traditional segmentation methods often use fixed-size divisions without considering differences in terrain complexity, resulting in insufficient segmentation accuracy in complex terrain areas and low segmentation efficiency in flat areas; Second, obtaining ground control points largely relies on manual on-site measurements, which is not only time-consuming and labor-intensive but also difficult to implement in complex or dangerous terrain areas, limiting the timeliness of image generation; Third, defects such as stitching gaps, uneven brightness, and loss of details are prone to occur during image stitching, and existing removal methods mostly rely on manual intervention, resulting in low automation and unstable results; Fourth, the overall process lacks coordination, with a lack of optimized connections between various stages from segmentation and stitching to control point extraction and image mosaicking, making it difficult to balance image generation efficiency and accuracy.

[0004] Therefore, there is an urgent need for an orthophoto generation method that can adapt to terrain features, automatically extract control points, and efficiently remove stitching defects, in order to solve the problems of insufficient accuracy, low efficiency, and low automation in existing technologies. Summary of the Invention

[0005] This application aims to address at least one of the technical problems existing in the prior art. To this end, this application proposes a method for rapid generation of high-precision digital orthophotos based on UAV aerial images. This method achieves rapid generation and improved accuracy of orthophotos through adaptive block division, automated control point extraction, and multi-scale fusion mosaicking.

[0006] To address the aforementioned problems, according to a first aspect of the present invention, a method for rapidly generating high-precision digital orthophotos based on UAV aerial photographs is provided, the method comprising: Acquire multiple drone aerial images and corresponding flight parameters, adaptively divide the flight area into blocks based on the geographical range of the flight area, maintain a preset overlap between blocks, and stitch all aerial images in each block area into a single area image to obtain multiple block area images; The feature point set of each aerial image is extracted based on the ORB algorithm. The pixel coordinates of the feature point set corresponding to the image block region are calculated. The geographic coordinates of the feature points are calculated in combination with the flight parameters. A one-to-one correspondence between the pixel coordinates and the geographic coordinates is established as ground control points. The images of each region are registered and corrected using the ground control points to generate standardized regional images with geographic coordinates; All standardized regional images are stitched together to obtain a complete initial image. A multi-scale fusion algorithm is then used to automatically remove imperfections at the stitching points in the complete initial image, outputting a high-precision digital orthophoto.

[0007] According to one embodiment of this application, the adaptive segmentation based on the geographical extent of the flight area, with pre-defined overlap between segments, includes: Based on the geographical range of the flight area, the geographical coordinate range of the UAV flight trajectory data and aerial images is analyzed to determine the latitude and longitude boundaries and terrain complexity of the flight area; The block size is adaptively adjusted according to the terrain complexity. When the terrain complexity is higher than a preset threshold, the first block size is used, and when the terrain complexity is lower than or equal to the preset threshold, the second block size is used. The first block size is smaller than the second block size. The flight area is divided into grids according to the block size, and the blocks maintain a preset overlap.

[0008] According to one embodiment of this application, the step of stitching together all aerial images within each segmented region into a single region image to obtain multiple segmented region images includes: Distortion correction and grayscale equalization preprocessing are performed on all aerial images within each segmented region to obtain multiple preprocessed aerial images; The SIFT algorithm is used to extract key feature points from each preprocessed aerial image, and the FLANN matcher is used to match feature points between adjacent aerial images. Based on the random sampling consensus algorithm, mismatched feature points are removed, and the homography matrix of adjacent preprocessed aerial images is calculated. Based on the homography matrix, all preprocessed aerial images within the block are projected, transformed, and fused together to obtain multiple block region images.

[0009] According to one embodiment of this application, the step of extracting a set of feature points for each aerial image based on the ORB algorithm, calculating the pixel coordinates of the feature point set corresponding to the image block region, and calculating the geographic coordinates of the feature points in combination with the flight parameters includes: The corner and edge features of each aerial image are extracted based on the ORB algorithm to form a set of feature points; Based on the homography matrix of the preprocessed aerial image, the pixel coordinates of each feature point in the segmented region image are obtained by reverse mapping. Acquire real-time flight parameters during drone aerial photography, including flight altitude, heading angle, pitch angle, roll angle, and GPS positioning coordinates; Based on the real-time flight parameters of the UAV camera, the geodetic coordinates of each feature point are calculated by solving the collinearity equation of photogrammetry, and used as the geographic coordinates of the feature point set.

[0010] According to one embodiment of this application, the step of registering and correcting each area image using the ground control points to generate a standardized area image with geographic coordinates includes: For each region image, the corresponding ground control points are filtered, and abnormal control points with coordinate deviations greater than a preset threshold are removed. A quadratic polynomial fitting algorithm is used to establish a mapping relationship model between regional image pixel coordinates and geographic coordinates; Based on the mapping relationship model, geometric correction is performed on the regional image to eliminate image distortion caused by terrain undulation and flight attitude deviation, and the corrected image is obtained. The corrected image is resampled using a bilinear interpolation algorithm to generate a standardized regional image with regular pixel arrangement and geographic coordinates.

[0011] According to one embodiment of this application, the step of automatically removing defects at the stitching points in the complete initial image using a multi-scale fusion algorithm to output a high-precision digital orthophoto image includes: Obtain the stitching boundary of adjacent regions in the complete initial image, and extract the transition region on both sides of the stitching boundary; Multi-scale decomposition of the transition region yields low-frequency and high-frequency components; The low-frequency components are fused using a weighted average method to obtain the fused low-frequency components; The gradient fusion method is used to fuse the high-frequency components to obtain the fused high-frequency components; The low-frequency and high-frequency components after fusion are inversely transformed to obtain a high-precision digital orthophoto with seamless stitching and uniform brightness.

[0012] According to one embodiment of this application, the method further includes: verifying the accuracy of the output high-precision digital orthophoto image by selecting multiple evenly distributed checkpoints and comparing the image coordinates of the checkpoints with the measured geographic coordinates; If the coordinate deviations are all less than the preset accuracy threshold, the final orthophoto image will be output. If there are checkpoints with coordinate deviations greater than the preset accuracy threshold, return to the ground control point extraction step, supplement the feature points of the area, and recalculate the control point coordinates.

[0013] According to a second aspect of the present invention, a device for rapidly generating high-precision digital orthophotos based on UAV aerial images is provided, the device comprising: The acquisition module is used to acquire multiple drone aerial images and corresponding flight parameters. It adaptively divides the flight area into blocks based on the geographical range of the flight area, maintains a preset overlap between blocks, and stitches together all aerial images in each block area into a single area image to obtain multiple block area images. The control point extraction module is used to extract the feature point set of each aerial image based on the ORB algorithm, calculate the pixel coordinates of the feature point set to the corresponding image block, calculate the geographic coordinates of the feature points in combination with the flight parameters, and establish a one-to-one correspondence between the pixel coordinates and the geographic coordinates as ground control points. The standardized regional image generation module is used to register and correct each regional image using the ground control points to generate a standardized regional image with geographic coordinates. The high-precision digital orthophoto generation module is used to stitch together all standardized area images to obtain a complete initial image. It uses a multi-scale fusion algorithm to automatically remove defects at the stitching points in the complete initial image and output a high-precision digital orthophoto.

[0014] According to a third aspect of the present invention, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method for rapid generation of high-precision digital orthophotos based on UAV aerial images as described in the first aspect above.

[0015] According to a fourth aspect of the present invention, a non-transitory computer-readable storage medium is provided, on which a computer program is stored, wherein when the computer program is executed by a processor, it implements the method for rapid generation of high-precision digital orthophotos based on UAV aerial photographs as described in the first aspect above.

[0016] According to a fifth aspect of the present invention, a chip is provided, the chip including a processor and a communication interface, the communication interface being coupled to the processor, the processor being used to run programs or instructions to implement the method for rapid generation of high-precision digital orthophotos based on UAV aerial images as described in the first aspect.

[0017] According to a sixth aspect of the present invention, a computer program product is provided, comprising a computer program that, when executed by a processor, implements the method for rapid generation of high-precision digital orthophotos based on UAV aerial photographs as described in the first aspect above.

[0018] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application.

[0019] The present invention provides a method for rapidly generating high-precision digital orthophotos based on UAV aerial images, which has the following advantages over existing technologies: (1) This invention achieves dynamic adjustment of block size through terrain complexity adaptive block strategy, which not only ensures the splicing accuracy of complex terrain areas, but also improves the processing efficiency of flat areas. At the same time, the preset block overlap provides sufficient feature matching redundancy for subsequent splicing, effectively solving the problem of accuracy and efficiency imbalance caused by traditional fixed blocks.

[0020] (2) This invention extracts feature points through the ORB algorithm and solves the geographic coordinates by combining flight parameters and photogrammetric collinearity equations, thereby realizing the automated extraction of ground control points. No manual on-site measurement is required, which greatly reduces labor costs and improves the timeliness of image generation. It is especially suitable for image acquisition scenarios in large areas or hard-to-reach areas.

[0021] (3) This invention uses a multi-scale fusion algorithm to refine the splicing transition area, optimize the brightness uniformity of the low-frequency component and the detail integrity of the high-frequency component respectively, and automatically remove defects such as splicing gaps and brightness unevenness. Compared with the traditional manual intervention to remove defects, it is not only more efficient, but also has a more stable fusion effect, which improves the visual quality and data reliability of orthophotos. Attached Figure Description

[0022] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which: Figure 1 This is one of the flowcharts illustrating the rapid generation method for high-precision digital orthophotos based on UAV aerial photographs provided in this application embodiment; Figure 2 This is the second flowchart illustrating the rapid generation method of high-precision digital orthophotos based on UAV aerial photographs provided in this application embodiment; Figure 3 This is a schematic diagram of the structure of the high-precision digital orthophoto rapid generation device based on UAV aerial photography provided in this application embodiment; Figure 4 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0023] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0024] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and are not used to describe a specific order or sequence. It should be understood that such use of data can be interchanged where appropriate to allow embodiments of this application to be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class and the number of objects is not limited; for example, a first object can be one or more. Furthermore, the use of "and" and / or to indicate at least one of the connected objects in the specification and claims, and the character " / ", generally indicates an "or" relationship between the preceding and following objects.

[0025] The following description, in conjunction with the accompanying drawings, details the method for rapid generation of high-precision digital orthophotos based on UAV aerial photographs, the apparatus for rapid generation of high-precision digital orthophotos based on UAV aerial photographs, the electronic equipment, and the readable storage medium provided in this application, through specific embodiments and application scenarios.

[0026] Among them, the method for rapidly generating high-precision digital orthophotos based on UAV aerial images can be applied to terminals, specifically executed by hardware or software within the terminal.

[0027] The terminal includes, but is not limited to, portable communication devices such as mobile phones or tablets with touch-sensitive surfaces (e.g., touchscreen displays and / or touchpads). It should also be understood that, in some embodiments, the terminal may not be a portable communication device, but rather a desktop computer with touch-sensitive surfaces (e.g., touchscreen displays and / or touchpads).

[0028] The following embodiments describe a terminal including a display and a touch-sensitive surface. However, it should be understood that the terminal may include one or more other physical user interface devices such as a physical keyboard, mouse, and joystick.

[0029] The high-precision digital orthophoto generation method based on UAV aerial images provided in this application embodiment can be executed by an electronic device or a functional module or entity within an electronic device that can implement the high-precision digital orthophoto generation method based on UAV aerial images. The electronic devices mentioned in this application embodiment include, but are not limited to, mobile phones, tablets, computers, cameras, and wearable devices. The following description uses an electronic device as an example to illustrate the high-precision digital orthophoto generation method based on UAV aerial images provided in this application embodiment.

[0030] Figure 1 This is one of the flowcharts illustrating the rapid generation method for high-precision digital orthophotos based on UAV aerial photography provided in this application embodiment, such as... Figure 1 As shown, the method for rapidly generating high-precision digital orthophotos based on UAV aerial photography includes steps 110, 120, 130, and 140.

[0031] Step 110: Acquire multiple drone aerial images and corresponding flight parameters, adaptively divide the area into blocks based on the geographical range of the flight area, maintain a preset overlap between blocks, and stitch all the aerial images in each block area into a single area image to obtain multiple block area images. In some embodiments, the adaptive segmentation based on the geographical extent of the flight area, with preset overlap between segments, includes: Based on the geographical range of the flight area, the geographical coordinate range of the UAV flight trajectory data and aerial images is analyzed to determine the latitude and longitude boundaries and terrain complexity of the flight area; The block size is adaptively adjusted according to the terrain complexity. When the terrain complexity is higher than a preset threshold, the first block size is used, and when the terrain complexity is lower than or equal to the preset threshold, the second block size is used. The first block size is smaller than the second block size. The flight area is divided into grids according to the block size, and the blocks maintain a preset overlap.

[0032] In some embodiments, stitching together all aerial images within each segmented region into a single region image to obtain multiple segmented region images includes: Distortion correction and grayscale equalization preprocessing are performed on all aerial images within each segmented region to obtain multiple preprocessed aerial images; The SIFT algorithm is used to extract key feature points from each preprocessed aerial image, and the FLANN matcher is used to match feature points between adjacent aerial images. Based on the random sampling consensus algorithm, mismatched feature points are removed, and the homography matrix of adjacent preprocessed aerial images is calculated. Based on the homography matrix, all preprocessed aerial images within the block are projected, transformed, and fused together to obtain multiple block region images.

[0033] In practical applications, the server receives aerial image data transmitted by the UAV through the UAV ground station. The aerial images are high-resolution RGB images, and the server also obtains the corresponding flight parameters, including the flight altitude, heading angle, pitch angle, roll angle, GPS positioning coordinates, and camera intrinsic parameters (focal length, principal point coordinates, distortion coefficient, etc.) at the time each frame of the aerial image was taken.

[0034] The server analyzes the geographic coordinate range of all aerial images to determine the latitude and longitude boundaries of the flight area. It then calculates the terrain complexity of the area using digital elevation model data. Terrain complexity can be measured using the standard deviation of slope; a larger standard deviation indicates more complex terrain. A terrain complexity threshold of 5 is set. When the standard deviation of slope for a given area is greater than 5, a first block size of 500x500 pixels is used; when the standard deviation of slope is less than or equal to 5, a second block size of 1000x1000 pixels is used.

[0035] The flight area is divided into grids according to the determined block size, with a 20% overlap between adjacent blocks. For example, the flight area's latitude and longitude range is 116.3-116.5 degrees east longitude and 39.8-40.0 degrees north latitude. Areas with complex terrain are divided into 500x500 pixel blocks, with adjacent blocks overlapping by 0.02 in both longitude and latitude directions to ensure redundant feature point matching during block stitching.

[0036] The aerial images within each block are preprocessed. First, radial and tangential distortion are corrected based on camera intrinsic parameters. Then, a histogram equalization algorithm is used for grayscale equalization to eliminate brightness differences under different lighting conditions. The SIFT algorithm is used to extract key feature points from each preprocessed aerial image. Each feature point generates a 128-dimensional feature descriptor. The Euclidean distance between feature descriptors of adjacent aerial images is calculated using a FLANN matcher to achieve preliminary feature point matching.

[0037] False matching feature points were eliminated using a random sampling consistency algorithm, with 1000 iterations and a distance threshold of 2.5. The matching set with the largest number of interior points was selected by calculating the homography matrix to ensure matching accuracy. Based on the homography matrix, all aerial images within the block were projected and transformed, mapping aerial images from different angles and positions to a unified block coordinate system. A linear fusion algorithm was then used to smoothly transition overlapping areas, generating seamless regional images.

[0038] Step 120: Extract the feature point set of each aerial image based on the ORB algorithm, calculate the pixel coordinates of the feature point set to the corresponding image block, calculate the geographic coordinates of the feature points in combination with the flight parameters, and establish a one-to-one correspondence between the pixel coordinates and the geographic coordinates as ground control points. In some embodiments, the step of extracting a set of feature points for each aerial image based on the ORB algorithm, calculating the pixel coordinates of the feature point set corresponding to the image block region, and calculating the geographic coordinates of the feature points in conjunction with the flight parameters includes: The corner and edge features of each aerial image are extracted based on the ORB algorithm to form a set of feature points; Based on the homography matrix of the preprocessed aerial image, the pixel coordinates of each feature point in the segmented region image are obtained by reverse mapping. Acquire real-time flight parameters during drone aerial photography. These parameters include flight altitude, heading angle, pitch angle, roll angle, and GPS positioning coordinates. Based on the real-time flight parameters of the UAV camera, the geodetic coordinates of each feature point are calculated by solving the collinearity equation of photogrammetry, and used as the geographic coordinates of the feature point set.

[0039] The ORB algorithm was used to extract corner and edge features from each aerial image. The maximum number of feature points was set to 2000, and the pyramid layer was set to 8 to ensure that a sufficient number of stable feature points were extracted. Based on the block stitching projection transformation matrix obtained in step 110, the pixel coordinates (u, v) of each feature point in the block region image were calculated through reverse mapping.

[0040] Acquire the flight parameters for each aerial image taken, including flight altitude H, heading angle, pitch angle, roll angle, and GPS positioning coordinates (,,), and combine them with camera intrinsic parameters. The geodetic coordinates of feature points are calculated by solving the collinearity equation through photogrammetry. The collinear equations are as follows: Where x and y are the coordinates of the feature point in the image coordinate system (obtained by pixel coordinate transformation). For the elements of the rotation matrix, The coordinates of the photography center (calculated from GPS positioning coordinates and flight altitude) are given. Here are the geodetic coordinates of the feature point.

[0041] The geographic coordinates of each feature point are obtained by solving the above equations. A one-to-one correspondence between the pixel coordinates of the feature points in the regional image and their geographic coordinates is established, thus forming a set of ground control points for the segmented region.

[0042] Step 130: Use the ground control points to register and correct the images of each region to generate a standardized regional image with geographic coordinates; In some embodiments, the step of registering and correcting each area image using the ground control points to generate a standardized area image with geographic coordinates includes: For each region image, the corresponding ground control points are filtered, and abnormal control points with coordinate deviations greater than a preset threshold are removed. A quadratic polynomial fitting algorithm is used to establish a mapping relationship model between regional image pixel coordinates and geographic coordinates; Based on the mapping relationship model, geometric correction is performed on the regional image to eliminate image distortion caused by terrain undulation and flight attitude deviation, and the corrected image is obtained. The corrected image is resampled using a bilinear interpolation algorithm to generate a standardized regional image with regular pixel arrangement and geographic coordinates.

[0043] The set of ground control points corresponding to each region's image is filtered, and the coordinate deviation of each control point is calculated. Abnormal control points with a deviation greater than 0.5 pixels are removed. A quadratic polynomial fitting algorithm is used to establish a mapping relationship model between pixel coordinates (u, v) and geographic coordinates (X, Y). The quadratic polynomial equation is as follows: in, , The coefficients are polynomials, obtained through least squares fitting. Based on this mapping model, geometric correction is performed on the regional image to eliminate image distortion caused by terrain undulations and flight attitude deviations. A bilinear interpolation algorithm is used to resample the corrected image, arranging the image pixels into a regular grid to generate a standardized regional image with geographic coordinates. The geographic coordinate system adopted is the WGS-84 coordinate system.

[0044] Step 140: Stitch together all standardized area images to obtain a complete initial image. Use a multi-scale fusion algorithm to automatically remove defects at the stitching points in the complete initial image and output a high-precision digital orthophoto.

[0045] In some embodiments, the automated removal of defects at the stitching points in the complete initial image using a multi-scale fusion algorithm to output a high-precision digital orthophoto image includes: Obtain the stitching boundary of adjacent regions in the complete initial image, and extract the transition region on both sides of the stitching boundary; Multi-scale decomposition of the transition region yields low-frequency and high-frequency components; The low-frequency components are fused using a weighted average method to obtain the fused low-frequency components; The gradient fusion method is used to fuse the high-frequency components to obtain the fused high-frequency components; The low-frequency and high-frequency components after fusion are inversely transformed to obtain a high-precision digital orthophoto with seamless stitching and uniform brightness.

[0046] All standardized regional images are stitched together according to their geographic coordinates to obtain a complete initial orthophoto. The stitching boundaries between adjacent regional images in the initial image are identified, and transition regions of 80 pixels on each side of the boundary are extracted. The transition regions are then decomposed into four levels using a Laplacian pyramid to obtain the low-frequency and high-frequency components of each level.

[0047] For low-frequency components, a weighted average method is used for fusion. The pixel weight coefficient changes linearly with the distance of the pixel from the stitching boundary; the closer to the boundary, the smaller the weight, and the farther from the boundary, the larger the weight, ensuring a uniform brightness transition in the transition area. For high-frequency components, a gradient fusion method is used for fusion. The gradient value of each pixel in the transition area is calculated, and the features of pixels with larger gradient values ​​are retained to ensure the integrity of image details.

[0048] The fused multi-scale components are subjected to an inverse Laplacian pyramid transform to obtain a seamless, uniformly bright, complete orthophoto image. Twenty evenly distributed checkpoints are selected, and the image coordinates of the checkpoints are compared with the measured geographic coordinates. If the coordinate deviation of all checkpoints is less than 0.3 meters, the image is output as the final high-precision digital orthophoto image. If there are checkpoints with coordinate deviations greater than 0.3 meters, the process returns to step 120, where feature points in the area of ​​the checkpoint are added and the control point coordinates are recalculated until the accuracy requirements are met.

[0049] In some embodiments, the method further includes: verifying the accuracy of the output high-precision digital orthophoto, selecting multiple evenly distributed checkpoints, and comparing the image coordinates of the checkpoints with the measured geographic coordinates; If the coordinate deviations are all less than the preset accuracy threshold, the final orthophoto image will be output. If there are checkpoints with coordinate deviations greater than the preset accuracy threshold, return to the ground control point extraction step, supplement the feature points of the area, and recalculate the control point coordinates.

[0050] Figure 2 This is the second flowchart illustrating the rapid generation method for high-precision digital orthophotos based on UAV aerial photography provided in this application. Figure 2 As shown, the detailed execution steps of this method are as follows: (1) Data preparation: The UAV performs aerial photography missions, collects high-resolution aerial images of the target area, and records flight parameters (flight altitude, attitude angle, GPS coordinates, etc.) and camera intrinsic parameters simultaneously; the UAV ground station receives the aerial images and flight parameters and transmits them to the server for storage.

[0051] (2) Adaptive segmentation: The server analyzes the geographical range and terrain complexity of the aerial image, determines the segment size and overlap, and adaptively segments the flight area.

[0052] (3) Block preprocessing: Distortion correction and grayscale equalization are performed on the aerial images in each block to eliminate the effects of camera distortion and lighting differences.

[0053] (4) Block stitching: Extract SIFT feature points of aerial images within the block, realize feature matching and mismatch elimination of adjacent aerial images, calculate homography matrix and complete the fusion stitching of aerial images within the block to generate regional images.

[0054] (5) Control point extraction: The ORB algorithm is used to extract the feature points of the aerial image, and the pixel coordinates are obtained by reverse mapping. The geographic coordinates are calculated by combining the flight parameters and the collinearity equation, and a set of ground control points is established.

[0055] (6) Registration and correction: Select ground control points, fit the mapping model between pixel coordinates and geographic coordinates, perform geometric correction and resampling on regional images, and generate standardized regional images.

[0056] (7) Overall stitching: stitch all standardized regional images together according to geographic coordinates to obtain a complete initial image.

[0057] (8) Defect removal: The stitching transition area of ​​the initial image is processed by a multi-scale fusion algorithm to remove defects such as stitching gaps and uneven brightness.

[0058] (9) Accuracy verification: Select checkpoints to verify the image accuracy. If the preset accuracy requirements are met, output the final orthophoto image; otherwise, return to the control point extraction step for optimization.

[0059] In this embodiment, the dynamic adjustment of block size is achieved by using a terrain complexity adaptive block strategy. This ensures the stitching accuracy of complex terrain areas and improves the processing efficiency of flat areas. At the same time, the preset block overlap provides sufficient feature matching redundancy for subsequent stitching, effectively solving the problem of the imbalance between accuracy and efficiency caused by traditional fixed blocks.

[0060] The method for rapidly generating high-precision digital orthophotos based on UAV aerial images provided in this application can be executed by a device for rapidly generating high-precision digital orthophotos based on UAV aerial images. This application uses the example of a device for rapidly generating high-precision digital orthophotos based on UAV aerial images executing the method to illustrate the device provided in this application.

[0061] This application also provides a device for rapidly generating high-precision digital orthophotos based on UAV aerial photographs, such as... Figure 3 As shown, the high-precision digital orthophoto rapid generation device based on UAV aerial photography includes: an acquisition module 310, a control point extraction module 320, a standardized area image generation module 330, and a high-precision digital orthophoto generation module 340.

[0062] The acquisition module 310 is used to acquire multiple drone aerial images and corresponding flight parameters, adaptively divide the flight area into blocks based on the geographical range of the flight area, maintain a preset overlap between blocks, and stitch all aerial images in each block area into a single area image to obtain multiple block area images. The control point extraction module 320 is used to extract the feature point set of each aerial image based on the ORB algorithm, calculate the pixel coordinates of the feature point set to the corresponding image block, calculate the geographic coordinates of the feature points in combination with the flight parameters, and establish a one-to-one correspondence between the pixel coordinates and the geographic coordinates as ground control points. The standardized regional image generation module 330 is used to register and correct each regional image using the ground control points to generate a standardized regional image with geographic coordinates. The high-precision digital orthophoto generation module 340 is used to stitch together all standardized area images to obtain a complete initial image. It uses a multi-scale fusion algorithm to automatically remove defects at the stitching points in the complete initial image and outputs a high-precision digital orthophoto.

[0063] The method for rapidly generating high-precision digital orthophotos based on UAV aerial images provided in this application employs a multi-scale fusion algorithm to refine the stitching transition area, optimizing the brightness uniformity of low-frequency components and the detail integrity of high-frequency components respectively. It automatically removes defects such as stitching gaps and uneven brightness. Compared with the traditional method of manually removing defects, it is not only more efficient but also has a more stable fusion effect, improving the visual quality and data reliability of orthophotos.

[0064] The high-precision digital orthophoto rapid generation device based on UAV aerial photography provided in this application embodiment can achieve… Figures 1-2 The various processes implemented in the embodiment of the method for rapid generation of high-precision digital orthophotos based on UAV aerial images will not be described in detail here to avoid repetition.

[0065] In some embodiments, such as Figure 4As shown, this application embodiment also provides an electronic device 400, including a processor 401, a memory 402, and a computer program stored in the memory 402 and executable on the processor 401. When the program is executed by the processor 401, it implements the various processes of the above-described embodiment of the method for rapid generation of high-precision digital orthophotos based on UAV aerial photographs and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0066] It should be noted that the electronic devices in the embodiments of this application include the mobile electronic devices and non-mobile electronic devices described above.

[0067] This application also provides a non-transitory computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the various processes of the above-described method for rapidly generating high-precision digital orthophotos based on UAV aerial photographs and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0068] The processor is the processor in the electronic device described in the above embodiments. The readable storage medium includes computer-readable storage media, such as computer read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0069] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method for rapidly generating high-precision digital orthophotos based on UAV aerial photographs.

[0070] The processor is the processor in the electronic device described in the above embodiments. The readable storage medium includes computer-readable storage media, such as computer read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0071] This application embodiment also provides a chip, which includes a processor and a communication interface. The communication interface and the processor are coupled. The processor is used to run programs or instructions to implement the various processes of the above-described embodiment of the method for rapid generation of high-precision digital orthophotos based on UAV aerial images, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0072] It should be understood that the chip mentioned in the embodiments of this application may also be referred to as a device-level chip, device chip, chip device, or on-chip device chip, etc.

[0073] It should be noted that, in this document, the terms include, encompass, or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, including an element by a statement does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0074] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a computer software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, or network device, etc.) to execute the high-precision digital orthophoto rapid generation method based on UAV aerial photography of various embodiments of this application.

[0075] In the description of this application, the first feature and the second feature may include one or more of the features.

[0076] In the description of this application, "multiple" means two or more.

[0077] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

[0078] In the description of this specification, the references to the terms "an embodiment," "some embodiments," "illustrative embodiments," "examples," "specific examples," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0079] Although embodiments of this application have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of this application, the scope of which is defined by the claims and their equivalents.

Claims

1. A method for rapidly generating high-precision digital orthophotos based on UAV aerial photography, characterized in that, include: Acquire multiple drone aerial images and corresponding flight parameters, adaptively divide the flight area into blocks based on the geographical range of the flight area, maintain a preset overlap between blocks, and stitch all aerial images in each block area into a single area image to obtain multiple block area images; The feature point set of each aerial image is extracted based on the ORB algorithm. The pixel coordinates of the feature point set corresponding to the image block region are calculated. The geographic coordinates of the feature points are calculated in combination with the flight parameters. A one-to-one correspondence between the pixel coordinates and the geographic coordinates is established as ground control points. The images of each region are registered and corrected using the ground control points to generate standardized regional images with geographic coordinates; All standardized regional images are stitched together to obtain a complete initial image. A multi-scale fusion algorithm is then used to automatically remove imperfections at the stitching points in the complete initial image, outputting a high-precision digital orthophoto.

2. The method for rapid generation of high-precision digital orthophotos based on UAV aerial photography according to claim 1, characterized in that, The adaptive segmentation based on the geographical extent of the flight area, with preset overlap between segments, includes: Based on the geographical range of the flight area, the geographical coordinate range of the UAV flight trajectory data and aerial images is analyzed to determine the latitude and longitude boundaries and terrain complexity of the flight area; The block size is adaptively adjusted according to the terrain complexity. When the terrain complexity is higher than a preset threshold, the first block size is used, and when the terrain complexity is lower than or equal to the preset threshold, the second block size is used. The first block size is smaller than the second block size. The flight area is divided into grids according to the block size, and the blocks maintain a preset overlap.

3. The method for rapid generation of high-precision digital orthophotos based on UAV aerial photographs according to claim 2, characterized in that, The process of stitching together all aerial images within each segmented region to form a single region image, resulting in multiple segmented region images, includes: Distortion correction and grayscale equalization preprocessing are performed on all aerial images within each segmented region to obtain multiple preprocessed aerial images; The SIFT algorithm is used to extract key feature points from each preprocessed aerial image, and the FLANN matcher is used to match feature points between adjacent aerial images. Based on the random sampling consensus algorithm, mismatched feature points are removed, and the homography matrix of adjacent preprocessed aerial images is calculated. Based on the homography matrix, all preprocessed aerial images within the block are projected, transformed, and fused together to obtain multiple block region images.

4. The method for rapid generation of high-precision digital orthophotos based on UAV aerial photographs according to claim 3, characterized in that, The process of extracting feature point sets from each aerial image using the ORB algorithm, calculating the pixel coordinates of the feature point sets corresponding to their respective image blocks, and combining the flight parameters to calculate the geographic coordinates of the feature points includes: The corner and edge features of each aerial image are extracted based on the ORB algorithm to form a set of feature points; Based on the homography matrix of the preprocessed aerial image, the pixel coordinates of each feature point in the segmented region image are obtained by reverse mapping. Acquire real-time flight parameters during drone aerial photography, including flight altitude, heading angle, pitch angle, roll angle, and GPS positioning coordinates; Based on the real-time flight parameters of the UAV camera, the geodetic coordinates of each feature point are calculated by solving the collinearity equation of photogrammetry, and used as the geographic coordinates of the feature point set.

5. The method for rapid generation of high-precision digital orthophotos based on UAV aerial photographs according to claim 4, characterized in that, The process of registering and correcting the images of each region using the ground control points to generate a standardized regional image with geographic coordinates includes: For each region image, the corresponding ground control points are filtered, and abnormal control points with coordinate deviations greater than a preset threshold are removed. A quadratic polynomial fitting algorithm is used to establish a mapping relationship model between regional image pixel coordinates and geographic coordinates; Based on the mapping relationship model, geometric correction is performed on the regional image to eliminate image distortion caused by terrain undulation and flight attitude deviation, and the corrected image is obtained. The corrected image is resampled using a bilinear interpolation algorithm to generate a standardized regional image with regular pixel arrangement and geographic coordinates.

6. The method for rapid generation of high-precision digital orthophotos based on UAV aerial photographs according to claim 5, characterized in that, The method employs a multi-scale fusion algorithm to automatically remove imperfections at the stitching points of the complete initial image, outputting a high-precision digital orthophoto image, including: Obtain the stitching boundary of adjacent regions in the complete initial image, and extract the transition region on both sides of the stitching boundary; Multi-scale decomposition of the transition region yields low-frequency and high-frequency components; The low-frequency components are fused using a weighted average method to obtain the fused low-frequency components; The gradient fusion method is used to fuse the high-frequency components to obtain the fused high-frequency components; The low-frequency and high-frequency components after fusion are inversely transformed to obtain a high-precision digital orthophoto with seamless stitching and uniform brightness.

7. The method for rapid generation of high-precision digital orthophotos based on UAV aerial photographs according to claim 6, characterized in that, The method further includes: verifying the accuracy of the output high-precision digital orthophoto, selecting multiple evenly distributed checkpoints, and comparing the image coordinates of the checkpoints with the measured geographic coordinates; If the coordinate deviations are all less than the preset accuracy threshold, the final orthophoto image will be output. If there are checkpoints with coordinate deviations greater than the preset accuracy threshold, return to the ground control point extraction step, supplement the feature points of the area, and recalculate the control point coordinates.

8. A device for rapidly generating high-precision digital orthophotos based on UAV aerial photographs, implemented using the method for rapidly generating high-precision digital orthophotos based on UAV aerial photographs as described in any one of claims 1 to 7, characterized in that, The device includes: The acquisition module is used to acquire multiple drone aerial images and corresponding flight parameters. It adaptively divides the flight area into blocks based on the geographical range of the flight area, maintains a preset overlap between blocks, and stitches together all aerial images in each block area into a single area image to obtain multiple block area images. The control point extraction module is used to extract the feature point set of each aerial image based on the ORB algorithm, calculate the pixel coordinates of the feature point set to the corresponding image block, calculate the geographic coordinates of the feature points in combination with the flight parameters, and establish a one-to-one correspondence between the pixel coordinates and the geographic coordinates as ground control points. The standardized regional image generation module is used to register and correct each regional image using the ground control points to generate a standardized regional image with geographic coordinates. The high-precision digital orthophoto generation module is used to stitch together all standardized area images to obtain a complete initial image. It uses a multi-scale fusion algorithm to automatically remove defects at the stitching points in the complete initial image and output a high-precision digital orthophoto.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method for rapid generation of high-precision digital orthophotos based on UAV aerial photographs as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the method for rapid generation of high-precision digital orthophotos based on UAV aerial images as described in any one of claims 1 to 7.

Images