Control point parallel manual measurement method for image aerial triangulation

Abstract

The invention discloses a control point parallel manual measurement method for image aerial triangulation. The method comprises the following steps: firstly, performing aerial triangulation on original image data with initial positioning parameters under the condition of no ground control, and optimizing geometric positioning parameters; secondly, dividing the complete measurement area into a plurality of subareas, predicting an image where a control point is located and an image point position of the image according to coordinates of the control point and image positioning parameters, cutting a local image block by taking the position as a center, and storing the local image block into a plurality of integrated image point data files according to the subareas; then, distributing the image point data file to a plurality of operators for parallelized control point measurement; and finally, the measured control point image coordinates are exported and recovered to an original measurement area project, and parallel measurement can be completed. According to the invention, the data portability is improved, the single-point measurement efficiency is improved, and parallel measurement is supported, so that the efficiency of large-data-volume control point measurement is improved to the maximum extent.

Description

Technical Field

[0001] The present invention relates to the field of photogrammetry and remote sensing, and in particular to a control point parallel manual measurement method for image aerial triangulation, wherein a parallel measurement scheme for local image cutting, management, reading and distribution based on control point coordinate projection is a key technology of the present invention. Background Art

[0002] Aerial triangulation (Aerial Triangulation) is a key technology in photogrammetric geometric processing. Its main task is to determine the precise geometric positioning parameters of all images in the area and the ground coordinates of the points to be determined. Aerial triangulation is the basis and prerequisite for the production of photogrammetric 4D products. It is widely used in the fields of geographic information systems, remote sensing image processing, and map making, and has a decisive influence on the accuracy and efficiency of the entire surveying and mapping project. In the process of aerial triangulation of images, control points are crucial to determining the absolute positioning accuracy of the image area network. The ground three-dimensional coordinates of the control points are known, and as geometric constraints, they can be used to correct and optimize the positioning parameters of the image, thereby ensuring the accuracy of the image data. Specifically, the control points accurately align the image area network with the actual ground position by providing geographic references, so that it meets the accuracy requirements of the relevant scale surveying and mapping specifications. For example, in the production of large-scale topographic maps, accurate control points can significantly improve the quality of the final product.

[0003] With the widespread application of POS (Position and Orientation System) and the development of aerial triangulation processing technology, the bottleneck problems faced by current image aerial triangulation mainly include huge data volume (the existing aerial triangulation image data is as high as tens of thousands or even hundreds of thousands), low processing efficiency, and difficulty in ensuring accuracy. Among them, the problem of control point measurement is particularly prominent. With the increase in image resolution and number, the manual measurement workload is huge and inefficient, and it is difficult to meet the needs of fast processing. After importing control points, current photogrammetry processing software can automatically predict the approximate position of control points on the image based on the ground coordinates of the control points and the initial positioning parameters of the image. However, the precise position of the control points on the image still needs to be accurately adjusted through manual measurement. The current manual measurement of control points mainly has the following two problems: (1) Large workload. With the increasing richness of data sources and the continuous increase in data volume, the number of control points required also increases. A survey area usually contains hundreds, thousands or even tens of thousands of images, and there is overlap between multiple images. Therefore, each control point may be located on multiple images, resulting in a huge workload for manual measurement. Existing software usually only allows one person to measure control points in a survey area, and it cannot be parallelized, resulting in low work efficiency; (2) Slow speed. As image resolution continues to increase, the size of original image files is getting larger and larger, especially satellite images, each of which is usually more than 1GB. When a control point is observed on multiple images, it is necessary to open and read the image data containing the control point one by one; and after adjusting the position of the control point, the image needs to be re-read according to the position. This one-by-one processing method not only has a slow reading speed, but also leads to slow measurement speed, and a lot of time is wasted on data reading and waiting. For example, if a control point is observed on 30 images, and the processing time for each image is 10 seconds, then the processing time for a single control point is 6 minutes. If the survey area contains hundreds of control points, the overall time consumption will be very huge. Obviously, the single-person measurement method is no longer suitable for the current needs of fast aerial triangulation processing of large-scale image data.

[0004] Therefore, it is necessary to develop a measurement method with fast measurement speed, high measurement efficiency and the ability to meet the needs of rapid aerial triangulation processing of large-scale image data. Summary of the invention

[0005] The purpose of the present invention is to provide a control point parallel manual measurement method for image aerial triangulation, which is a method of image clipping and parallel measurement based on control points as the center, with fast measurement speed and high measurement efficiency, and can meet the needs of fast aerial triangulation processing of large-scale image data. The method predicts the position of the control point on all related images, clips the local image with the predicted control point as the center, and then integrates all local images of all control points into a separate file, copies and distributes them, realizes image clipping and optimized preservation, retains image texture feature information, significantly reduces the image file size, and maximizes the image reading speed and measurement efficiency; at the same time, the clipped image file is distributed according to the work task, and any operator can edit and measure the distributed image, and multiple operators can edit and measure at the same time, which minimizes the workload of a single person, thereby greatly improving the control point measurement efficiency, reducing the workload of a single person, and greatly improving the control point measurement speed and efficiency; solves the problem of large workload of single-person measurement in traditional aerial triangulation, slow manual measurement speed and low efficiency of a large number of control points.

[0006] In order to achieve the above object, the technical solution of the present invention is: a control point parallel manual measurement method for image aerial triangulation, characterized in that it includes the following steps:

[0007] Step 1, obtaining the image positioning parameters of the survey area;

[0008] Step 2: Partition the survey area: partition the survey area image data and divide the work tasks in advance;

[0009] Step 3: predicting the position of the control point; predicting the position of the control point on the image based on the image positioning parameters and the control point coordinates in step 1;

[0010] Step 4, taking the control point of the predicted point as the center, perform local image cropping and output it to the integrated image point data file;

[0011] Step 5, performing parallel control point measurement according to the partition division in step 2;

[0012] Step 6: Export the measurement results, obtain the precise image point coordinates of the control points, and restore the control point positions.

[0013] In the above technical solution, in step 1, the method for obtaining the positioning parameters of the survey area image includes but is not limited to performing non-control aerial triangulation on the survey area image to obtain the image positioning parameters, obtaining the initial positioning parameters of the original image of the survey area as the image positioning parameters, and directly using the initial positioning parameters of the original image of the survey area to predict the control point position;

[0014] Uncontrolled aerial triangulation is to perform aerial triangulation of the survey area images without control conditions, assisted by initial positioning parameters, to optimize image positioning parameters.

[0015] In the above technical solution, the specific method of non-controlled aerial triangulation is:

[0016] Using the initial positioning parameters of the survey area image, aerial triangulation is performed without ground control points to optimize the image positioning parameters and improve the accuracy and consistency of control point position prediction.

[0017] In the above technical solution, in step 2, the entire survey area is divided into several sub-areas according to the geographical scope of the survey area and the coverage of the image data, and then the control point measurement task of each sub-area is assigned to different operators.

[0018] In the above technical solution, in step 3, according to the initial positioning parameters of the image and the coordinates of the control points, the image of the survey area where the control points are located and the image point coordinates of the control points on each image are predicted through projection calculation, and the projection is realized by the following formula:

[0019] (x,y)=f k (X,Y,Z) (1)

[0020] Among them, (X, Y, Z) is the ground coordinates of the control point, (x, y) is the projection function f through image n k (*) Predicted image coordinates of control points.

[0021] In the above technical solution, the latitude and longitude range in the metadata of the original image is used to determine whether a control point is located on the image, so that the control point and the image are retrieved and paired before projection to improve the calculation efficiency. The determination method is as follows:

[0022] Assume that the longitude and latitude coordinates of control point i are (Lat i ,Lon i ), for the longitude and latitude range of the kth image If the control point satisfies the following formula (2), the control point is on the image; otherwise, the control point is not on the image and is deleted.

[0023]

[0024] In the above technical solution, in step 4, after predicting the position of the control point, the survey area image is cropped and saved as a small picture with the control point of the predicted point as the center. For the cropped image, the following parameter information should be recorded: the image plane coordinates of the control point (x, y); the image plane starting point coordinates of the cropping window (x0, y0); the cropped image size r×c, where r and c are the number of rows and columns of the cropped image respectively; the above parameter information will be used to restore the position of the point in the original complete image.

[0025] The above parameter information is integrated and output into an integrated image point data file, which includes the following data contents: ground coordinates of control points, cropped local image blocks, original image metadata and parameter information of the cropped image recorded therein.

[0026] In the above technical solution, in step 5, during the manual editing process, infinite scaling technology is used to ensure the accuracy of image analysis and avoid errors introduced by image scaling. At the same time, one image can be selected as a reference image, its image point position can be edited, and then the point can be automatically matched with the same-name point on other images to further improve efficiency. In addition, when the image meets the binocular stereo conditions, stereo measurement of control points in a stereo environment is supported.

[0027] In the above technical solution, in step 6, the precise image point coordinates of the exported control points are independent of the original image data, do not include the image file, and only retain the position data.

[0028] The present invention has the following advantages:

[0029] (1) The editing efficiency of a single control point is significantly improved; the present invention significantly reduces the amount of data read from the original large image to only read a partial image, so the reading efficiency is high and the zooming speed is fast; it overcomes the shortcomings of the prior art that the single-person measurement method requires reading the entire image, resulting in a large amount of data read, low reading efficiency, and slow zooming speed;

[0030] (2) It supports multi-person parallel measurement, improves the overall editing efficiency of multiple control points, and reduces the workload of a single person; by decomposing tasks, the editing tasks of all control points in the entire survey area are assigned to multiple people for parallel editing. Compared with single-person editing, the editing efficiency is improved several times (the editing efficiency of the present invention is approximately equal to the number of people editing in parallel), with fast measurement speed and high measurement efficiency, and can meet the needs of fast aerial triangulation processing of large-scale image data (such as tens of thousands or even hundreds of thousands of images); it overcomes the problem that the prior art usually only allows one person to measure control points in a survey area, and cannot be parallelized, the workload of a single person is large, the measurement speed is slow, the measurement operation efficiency is low, and the needs of fast aerial triangulation processing of large-scale image data cannot be met;

[0031] (3) It is highly independent and does not rely on the overall aerial triangulation survey area data containing numerous files. It only needs to integrate the image point data files. It has strong mobility and can perform parallel measurement work on different computers or even different platforms, which effectively improves the portability of data and simplifies the data management process, making it convenient for staff to perform real-time measurement operations anytime and anywhere. It overcomes the problem that the existing technology relies on the overall aerial triangulation survey area data containing numerous files, requires measurement work to be performed on a designated computer, cannot transplant and carry data, and cannot realize real-time measurement operations of staff. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is the overall technical flow chart of the present invention;

[0033] Figure 2 This is a schematic diagram of the control point distribution and measurement area division of the present invention;

[0034] Figure 3 This is a schematic diagram of cutting control point image blocks according to the present invention;

[0035] Figure 4 This is a schematic diagram of the control point position measurement of the present invention;

[0036] Figure 5 It is a schematic diagram of the three-dimensional measurement of the present invention. DETAILED DESCRIPTION

[0037] The following is a detailed description of the implementation of the present invention in conjunction with the accompanying drawings, but they do not constitute a limitation of the present invention and are only given as examples. At the same time, the advantages of the present invention are made clearer and easier to understand through the description.

[0038] The present invention discloses a control point parallel manual measurement method for image aerial triangulation. The core is to export the local image data required for control point measurement in photogrammetry aerial triangulation operation into independent files, and then perform parallel image point measurement according to the partitions to improve the control measurement efficiency. With the original image data with initial positioning parameters and control point data as input, firstly, the original image data with initial positioning parameters is subjected to aerial triangulation without ground control conditions, and the geometric positioning parameters are optimized, or the initial positioning parameters of the original image are directly used to predict the control point positioning. Secondly, the image data is partitioned according to the operation requirements, and the complete survey area is divided into multiple partitions. According to the control point coordinates and the image initial positioning parameters, the control points are predicted in which images and their image point positions, and the local image blocks are cropped with the position as the center, and saved to multiple integrated image point data files according to the partitions; then, the image point data files are distributed to multiple operators for parallel control point measurement; finally, the measured control point image coordinates are exported and restored to the original survey area project, and the parallel measurement can be completed. Different from the traditional single-person measurement mode, the method of the present invention is no longer limited to the single operation mode of raw data and engineering, but maximizes the efficiency of large-scale data control point measurement by improving data portability, improving single-point measurement efficiency and supporting parallel measurement.

[0039] Referring to the attached drawings, it can be seen that the present invention is mainly implemented through the following technical solutions: a method for parallel manual measurement of control points for image aerial triangulation. The method takes the original image data with initial positioning parameters and control point data as input, first performs uncontrolled aerial triangulation on the original image data, and partitions the data according to the operation requirements, and predicts the point position on the image based on the initial positioning parameters of the image and the coordinates of the control points, and performs local image cropping, saves it to a unified file, and distributes it for measurement. Its core process includes the following steps:

[0040] Step 1: Uncontrolled aerial triangulation. Perform aerial triangulation on the survey area images with the aid of initial positioning parameters under uncontrolled conditions to optimize the image positioning parameters;

[0041] Step 2: Partition the survey area. Partition the survey area data and divide the work tasks in advance;

[0042] Step 3: Predict the position of the control point. Predict the position of the control point on the image based on the image positioning parameters and the coordinates of the control point;

[0043] Step 4, cropping the local image and outputting it to an integrated image point data file;

[0044] Step 5, perform parallel control point measurement according to the partition division;

[0045] Step 6: Export the measurement results and restore the control point positions.

[0046] Furthermore, in step 1, the initial positioning parameters of the survey area image are used to perform aerial triangulation without ground control points, aiming to optimize the positioning parameters of the image. Different types of image data usually have different forms of initial positioning parameters. For example, satellite images usually use RPC (Rational Polynomial Coefficients) parameters; aerial and drone images use the central projection of the pinhole imaging principle, and their parameters include the camera's internal orientation elements (focal length, image principal point coordinates, objective lens distortion coefficient, etc.) and external orientation elements (including position and attitude). Through uncontrolled aerial triangulation, these initial positioning parameters can be used to further optimize the geometric positioning of the image, eliminate systematic errors caused by factors such as sensors, platform motion, and external interference, reduce model differences, and thus improve the relative positioning accuracy of the image and accurately predict the point position. This is crucial for the subsequent prediction of the point position of the control point on the image, and the efficiency of manual measurement is improved by significantly improving the accuracy of the predicted position of the control point on the image.

[0047] It should be emphasized that the step of uncontrolled aerial triangulation is not absolutely necessary. When the initial positioning parameters of the image are relatively accurate, the initial positioning parameters of the image can be used directly to predict the position of the control points. However, since these original parameters have not been optimized, there may be large systematic errors between images, resulting in large deviations in the predicted positions of the control points on different images. Therefore, in order to improve the prediction accuracy and consistency of the control points, it is usually recommended to perform uncontrolled aerial triangulation before processing, so as to optimize the initial positioning parameters, ensure that the predicted position is close to the actual point position, and improve the efficiency of manual measurement.

[0048] Further, in step 2, first, according to the geographical scope of the survey area and the coverage of the image data, the entire survey area is divided into Figure 2 The process needs to consider the distribution density of control points and the coverage of images. For example, data partitioning can be optimized based on the number of images around each control point to ensure a balanced ratio of images and control points in each partition, thereby improving the efficiency and accuracy of subsequent processing. Secondly, after the partitioning is completed, specific work tasks will be divided according to the data volume and complexity of each partition and assigned to different processing personnel to achieve an efficient parallel workflow.

[0049] Furthermore, in step 3, according to the initial positioning parameters of the image and the coordinates of the control points, through projection calculation, it is predicted which images the control points are located in and the coordinates of the image points on each image, which can be achieved by projection through the following formula:

[0050] (x,y)=f k (X,Y,Z) (1)

[0051] Among them, (X, Y, Z) is the ground coordinates of the control point, (x, y) is the projection function f through image n k (*) Predicted image coordinates of control points.

[0052] Considering that there are many control points and each control point may correspond to multiple images, in order to improve the prediction efficiency, before predicting the control points, it is necessary to retrieve the image number corresponding to the control point in advance. To this end, the latitude and longitude range in the metadata of the original image can be used to determine whether a control point is located on the image. Assume that the longitude and latitude coordinates of the control point i are (Lat i ,Lon i ), for the kth image with a latitude and longitude range Then when the control point satisfies the following formula (2), the control point is on the image;

[0053]

[0054] Further, in step 4, after predicting the position of the control point, the image is cropped with the control point as the center and saved as shown in the attached figure. Figure 3 The small local picture shown ( Figure 3 The image segmentation method of the present invention is shown. By segmenting, the storage space can be fully reduced, the portability can be improved, and the image reading efficiency can be improved. For the cropped image, the following cropping parameter information should be recorded: the image coordinates of the control point (x, y); the image coordinates of the starting point of the cropping window (x0, y0); the cropped image size r×c, where r and c are the number of rows and columns of the cropped image respectively. The above parameter information will be used to restore the position of the point in the original complete image.

[0055] The above information and related image data are integrated and output into a data file, which is called the integrated image point data file in the present invention. The file includes the following data content: the ground coordinates of the control point, the local image block obtained by cutting, the original image metadata and its recorded cutting parameters. The file effectively improves the portability of the data by storing each control point and its local image block on each image together. Each staff member can directly copy the file to a personal machine device for reading and writing operations and measurement work, which simplifies the data management process.

[0056] Furthermore, in step 5, the data file generated in step 4 is used to perform the control point measurement. This step is carried out in parallel in different partitions, and each operator performs accurate measurement of the control points according to the assigned task. During the manual editing process, the infinite scaling technology is used to ensure the accuracy of image analysis and avoid errors introduced by image scaling. The measurement operation must be completed at high resolution to ensure the accuracy and reliability of the control point position. Single point measurement is shown in the attached figure. Figure 4 As shown ( Figure 4 The measurement effect of the image block of the present invention is shown. Figure 4 It can be seen that the image analysis of the present invention has a high accuracy, which proves the feasibility of using image blocks for measurement. It can be seen that the present invention has the function of simultaneously measuring single-point images with different widths only through image blocks); at the same time, Figure 4 The measurement process of the control point to be measured on multiple image blocks is shown; the multiple image blocks are cut from different scene images using the method of the present invention. When n control points are to be measured in a measurement project, the project area covers m scene images. For each control point, it is not necessarily on only one image, but may exist on multiple (such as 2, 3, 4, etc.) images; Figure 4 The above shows the measurement process of the control point to be measured on two image blocks; when the control point is on a larger number of images (such as 5, 6, etc.), the method of the present invention can also be used to measure the control point.

[0057] To further improve measurement efficiency, operators can use image matching to assist in image point editing. Specifically, you can select an image as a reference image, edit its image point position, and then automatically match the point with the same name on other images. This method is particularly effective when the control point involves multiple images. For most control points with obvious features, this method can achieve one-click control point measurement, and the operator only needs to quickly visually check. In addition, when the image meets the binocular stereo conditions, it can also be used in the following Figure 5 Stereo measurement of control points in the stereo environment shown in the figure ( Figure 5 It is an effect diagram of the superposition of stereoscopic images, showing the three-dimensional effect of the stereoscopic image. The present invention can measure the control points on the three-dimensional basis of the stereoscopic image); Figure 5 It is shown that when the image meets the stereo condition, additional control points (including plane and elevation) can be directly measured. It can be seen that the present invention has a stereo measurement function, improves the control point measurement effect, and improves the applicability of the present invention.

[0058] Furthermore, in step 6, the precise image point coordinates of the manually edited control points are exported. These coordinates are independent of the original image data, do not contain the image file, and only retain the position data. The original control point positions are updated according to the measured data to ensure that each control point is a manually measured result.

[0059] Example

[0060] The present invention is now described in detail by taking the application of the present invention to a certain province for image aerial triangulation as an example, which can also provide guidance for the application of the present invention to image data aerial triangulation in other regions.

[0061] This example uses satellite images covering the entire province for illustration. There are 145 scenes of Resource 3 three-line array images in the province, with a data size of about 224GB and a total of 722 field control points. After the project is established, the control point data is imported.

[0062] This embodiment adopts the single-person measurement method of the prior art for measurement, which has a large amount of data reading, low reading efficiency, slow scaling speed, and low measurement efficiency. Specifically, it is reflected in the following aspects: First, 224GB of data is not suitable and inconvenient to distribute, and it takes more than several hours to copy each time. Secondly, when a control point is observed on multiple images, assuming that a point is on 7 images, then the measurement of a single point requires opening about 11gb of images at the same time, which consumes a lot of memory. The computer is very demanding during measurement. At the same time, when such a large image is opened on the computer, even if the memory is large enough, the rendering time is very long, and the scaling and movement will be stuck. Finally, it takes about 5 minutes for a skilled operator to measure one control point (data reading time + manual measurement and adjustment time). Based on the prior art, since the data can only be measured by one person, the overall measurement time of this example is about 3610 minutes (about 60 hours), the measurement workload is large, and the measurement efficiency is low.

[0063] The flowchart of this example using the method of the present invention to perform parallel editing of control points is shown in the attached figure. Figure 1 As shown, the following steps are included:

[0064] Step 1: Perform aerial triangulation on the satellite image of the survey area without ground control points to optimize the satellite image RPC parameters. In this embodiment, the survey area data uses satellite image data, so the image positioning parameters here are RPC model parameters, as shown in the following formula:

[0065]

[0066] Among them, (P n ,L n ,H n ) is the regularized ground coordinate, (r n ,c n) are the regularized image coordinates.

[0067] Step 2: Partition the measurement area data. In this embodiment, the measurement area data is divided per person according to three testers (it can also be distributed to other numbers of operators as needed).

[0068] Step 3: Predict control point positions.

[0069] Import all control points that need to be measured. After the import is completed, use the point prediction function to calculate and predict the satellite images and image point coordinates of the control points through RPC projection.

[0070] Step 4: Output the integrated pixel data file.

[0071] Based on step 3, the control point parallel editing data is exported. The exported file contains small pictures and cropping parameters of each control point on all images and metadata of the original image. After the export is completed, you can see that the integrated file size of the exported small picture is 3.9GB. Compared with the original image, the data size is reduced by 98.26%, the zoom becomes faster, and the speed of image loading is effectively improved.

[0072] Step 5: Measure control points according to the partition division.

[0073] The file exported in the previous step is copied to the work computers of all operators. The operators can open the files on other devices according to the task division in step 2 to start measurement. It has strong independence and only needs to integrate the image point data files. It has strong mobility and can perform parallel measurement work on different computers or even different platforms. Since the present invention supports multi-person parallel measurement, the overall editing efficiency of multiple control points is improved and the workload of a single person is reduced. Through task decomposition, the editing tasks of all control points in the entire measurement area are assigned to multiple people for parallel editing. Compared with single-person editing, the editing efficiency is improved several times.

[0074] Step 6: Export the measurement results and restore the control point positions.

[0075] After the measurement is completed, the measured image point positions are exported as image point coordinate files. This file does not contain image data and is relatively small. Use the recovery function to import the measured control point image point coordinates into the measurement area. After all the control point positions of the partition files are restored, the control point measurement work is completed.

[0076] This embodiment adopts the method of the present invention to distribute the cut and packaged data to three people for simultaneous measurement. Even if the zoom speed is not taken into account, the efficiency is improved by 3 times. Taking the zoom speed into account, the speed of image loading is effectively improved, and the time for manual measurement of one point can be reduced to less than 1 minute. The overall measurement time of this embodiment is about 12 hours, and the overall measurement time is reduced to about 1 / 5 of the existing single-person measurement method, and the overall measurement efficiency is greatly improved.

[0077] The present invention mainly solves the problem of slow control point measurement speed and low efficiency in traditional aerial triangulation, and proposes a method of image cropping and parallel measurement based on control points. This method crops images based on control points and integrates them into separate files for distribution. While retaining image texture feature information, it significantly reduces the image file size and maximizes the image reading speed. For the distributed images, each participant can edit and measure at the same time, thereby greatly improving the measurement efficiency, reducing the workload of a single person, and greatly improving the control point measurement efficiency.

[0078] The specific embodiments described herein are merely examples of the spirit of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in similar ways, but they will not deviate from the spirit of the present invention or exceed the scope defined by the appended claims.

[0079] Other parts not described belong to the prior art.

Claims

1. A control point parallel manual measurement method for image aerial triangulation, characterized by: The following steps are included: Step 1, obtaining the image positioning parameters of the survey area; Step 2: Partition the survey area: partition the survey area data and divide the work tasks in advance; Step 3: predicting the position of control points: predicting the position of control points on the image based on the image positioning parameters and the coordinates of the control points; Step 4, cropping the local image and outputting the integrated image point data file; Step 5, perform parallel control point measurement according to the partition division; Step 6: Export the measurement results and restore the control point positions.

2. The control point parallel manual measurement method for image aerial triangulation according to claim 1, characterized in that: In step 1, the method of obtaining the positioning parameters of the survey area image includes but is not limited to performing uncontrolled aerial triangulation on the survey area image to obtain the image positioning parameters, and obtaining the initial positioning parameters of the original image of the survey area as the image positioning parameters.

3. The control point parallel manual measurement method for image aerial triangulation according to claim 2, characterized in that: The specific method of non-controlled aerial triangulation is as follows: Using the initial positioning parameters of the survey area images, aerial triangulation is performed without ground control points.

4. The control point parallel manual measurement method for image aerial triangulation according to claim 2, characterized in that: In step 2, the entire survey area is divided into several sub-areas according to the geographical scope of the survey area and the coverage of the image data, and then the control point measurement tasks of each sub-area are assigned to different operators.

5. The control point parallel manual measurement method for image aerial triangulation according to claim 4, characterized in that: In step 3, according to the initial positioning parameters of the image and the coordinates of the control points, the image of the survey area where the control points are located and the image point coordinates of the control points on each image are predicted through projection calculation, and the projection is realized by the following formula: (x,y)=f k (X,Y,Z) (1) Among them, (X, Y, Z) is the ground coordinates of the control point, (x, y) is the projection function f through image n k (*) Predicted image coordinates of control points.

6. The control point parallel manual measurement method for image aerotriangulation according to claim 5, characterized in that: The latitude and longitude range in the metadata of the original image is used to determine whether a control point is located on the image, so that the control point and the image are retrieved and paired before projection. The determination method is as follows: Assume that the longitude and latitude coordinates of control point i are (Lat i ,Lon i ), for the longitude and latitude range of the kth image If the control point satisfies the following formula (2), the control point is on the image; otherwise, the control point is not on the image and is deleted.

7. The control point parallel manual measurement method for image aerotriangulation according to claim 6, characterized in that: In step 4, the image of the survey area is cropped and saved as a small picture with the control point of the predicted point as the center. For the cropped image, the following parameter information is recorded: the image coordinates of the control point (x, y); the coordinates of the starting point of the cropping window (x0, y0); the cropped image size r×c, where r and c are the number of rows and columns of the cropped image respectively; The above parameter information is integrated and output into an integrated image point data file, which includes the following data contents: ground coordinates of control points, cropped local image blocks, original image metadata and parameter information of the cropped image recorded therein.

8. The control point parallel manual measurement method for image aerotriangulation according to claim 7, characterized in that: In step 5, during the manual editing process, the infinite scaling technique is used to ensure the accuracy of the image analysis. At the same time, one image is selected as the reference image, its image point position is edited, and then the point is automatically matched with the same name point on other images.

9. The control point parallel manual measurement method for image aerial triangulation according to claim 8, characterized in that: In step 5, when the image meets the binocular stereo condition, stereo measurement of control points in a stereo environment is supported.

10. The control point parallel manual measurement method for image aerial triangulation according to claim 9, characterized in that: In step 6, the precise image point coordinates of the derived control points are independent of the original image data, and only the position data is retained.

Images