Mobile camera positioning correction method and device based on long-distance shooting

Abstract

This invention provides a method and apparatus for positioning and correcting a mobile camera based on long-distance shooting. The method automatically obtains feature point sets from two sets of image samples through a feature point model, and obtains the corresponding transformation matrix based on the feature point sets. Then, the transformation matrix is ​​used as the basis for adjusting the position coordinates of the mobile camera. Through feedback adjustment, the deviation between the adjusted position of the mobile camera and the target position coordinates can be effectively reduced. In addition, the mobile camera is adjusted according to the transformation matrix, the first position coordinates, the first gimbal angle, the second position coordinates, and the second gimbal angle until the target reaches the preset position, ensuring the reliability of the mobile camera positioning and correction results.

Description

Technical Field

[0001] This invention relates to the field of computer vision, and more particularly to a method and apparatus for positioning and correcting a mobile camera based on long-distance shooting. Background Technology

[0002] The widespread application of computer vision technology has brought about an industrial technology revolution. In industrial environments, especially in the aerospace field, long-distance shooting technology has many application scenarios.

[0003] However, when current robots carry cameras to move to a target point to take pictures, especially when shooting distant targets, there will be some deviation due to positioning accuracy. In other words, according to the pre-set gimbal angle, magnification, and focal length, the target cannot appear in the preset position, making it impossible to process the target further.

[0004] Therefore, in the existing technology, when using a moving camera to take pictures of a target at a distance, there is a problem that the moving camera is not accurately positioned, resulting in the target not being in the preset position. Summary of the Invention

[0005] In view of this, it is necessary to provide a method, device, electronic device and storage medium for positioning correction of a moving camera based on long-distance shooting, so as to solve the problem in the prior art that the positioning of the moving camera is inaccurate when shooting a target from a distance, resulting in the target not being in the preset position.

[0006] To address the aforementioned problems, this invention provides a mobile camera positioning correction method based on long-distance shooting, comprising:

[0007] The first position coordinates and the first gimbal angle of the moving camera are obtained, the target is photographed, the first image sample set is obtained, and the first feature point set of the first image sample set is obtained according to the first feature point model.

[0008] Adjust the moving camera to the second position coordinates and the second gimbal angle, capture the target, obtain the second image sample set, and obtain the second feature point set of the second image sample set according to the second feature point model;

[0009] Determine the transformation matrix based on the first feature point set and the second feature point set;

[0010] Based on the transformation matrix, the first position coordinates, the first gimbal angle, the second position coordinates, and the second gimbal angle, adjust the moving camera to the target position and the target gimbal angle.

[0011] Further, the first position coordinates of the moving camera and the first gimbal angle are obtained, the target is photographed, and a first image sample set is obtained, including:

[0012] Based on the first position coordinates of the moving camera and the first gimbal angle, the target is photographed sequentially from the maximum magnification to the minimum magnification to obtain the first image sample set.

[0013] Furthermore, the first feature point model and the second feature point model have the same structure, both including convolutional layers, feature learning layers, and classification layers.

[0014] Further, obtaining the first feature point set / second feature point set of the first image sample set / second image sample set based on the first feature point model / second feature point model includes:

[0015] The first image sample set / second image sample set is input into the convolutional layer, and a convolution operation is performed to obtain the first image feature vector / second image feature vector.

[0016] The first image feature vector / the second image feature vector are input into the feature learning layer to obtain the first image feature matrix / the second image feature matrix;

[0017] The first image feature matrix and the second image feature matrix are input into the classification layer to determine the first feature point set and the second feature point set.

[0018] Further, based on the first feature point set and the second feature point set, the transformation matrix is ​​determined, including:

[0019] Based on the first and second feature point sets, the two most similar matching feature points are determined by the random sample consensus algorithm.

[0020] The transformation matrix is ​​determined based on the two sets of most similar matching feature points.

[0021] Furthermore, based on the transformation matrix, the first position coordinates, the first gimbal angle, the second position coordinates, and the second gimbal angle, the camera is moved to the target position and the target gimbal angle is adjusted, including:

[0022] Based on the transformation matrix, the camera is moved to the third position coordinate and the third gimbal angle is adjusted, and the target is photographed to obtain the third image sample set of the target. The third feature point set of the third image sample set is obtained based on the third feature point model.

[0023] Based on whether the third feature point set has reached the preset position, determine whether the moving camera has reached the target position and the target gimbal angle. If yes, then the third position coordinates and the third gimbal angle are determined to be the target position and the target gimbal angle. If not, then the third feature point set is regarded as the second feature point set, the transformation matrix is ​​updated to the second transformation matrix, and the moving camera is repeatedly adjusted according to the second transformation matrix until the final feature point set reaches the preset position. Then, the target position and the target gimbal angle are determined accordingly based on the final feature point set.

[0024] Furthermore, the second and third position coordinates can be the same as the first position coordinates.

[0025] To address the aforementioned problems, the present invention also provides a mobile camera positioning correction device based on long-distance shooting, comprising:

[0026] The first feature point set acquisition module is used to acquire the first position coordinates and the first gimbal angle of the moving camera, capture the target, obtain the first image sample set, and acquire the first feature point set of the first image sample set according to the first feature point model.

[0027] The second feature point set acquisition module is used to adjust the moving camera to the second position coordinates and the second gimbal angle, capture the target, obtain the second image sample set, and obtain the second feature point set of the second image sample set according to the second feature point model;

[0028] The transformation matrix acquisition module is used to determine the transformation matrix based on the first feature point set and the second feature point set.

[0029] The positioning module is used to adjust the moving camera to the target position and the target gimbal angle based on the transformation matrix, the first position coordinates, the first gimbal angle, the second position coordinates, and the second gimbal angle.

[0030] To address the aforementioned problems, the present invention also provides an electronic device, including a processor and a memory, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, it implements the mobile camera positioning correction method based on long-distance shooting as described above.

[0031] To address the aforementioned problems, the present invention also provides a computer-readable storage medium storing computer program instructions. When the computer program instructions are executed by a computer, the computer performs the mobile camera positioning correction method based on long-distance shooting as described above.

[0032] The beneficial effects of adopting the above technical solution are as follows: This invention provides a method, device, electronic device, and storage medium for positioning and correction of a mobile camera based on long-distance shooting. The method automatically obtains feature point sets from two sets of image samples through a feature point model, and obtains the corresponding transformation matrix based on the feature point sets. Then, the transformation matrix is ​​used as the basis for adjusting the position coordinates of the mobile camera. Through feedback adjustment, the deviation between the adjusted position of the mobile camera and the target position coordinates can be effectively reduced. In addition, the mobile camera is adjusted according to the transformation matrix, the first position coordinates, the first gimbal angle, the second position coordinates, and the second gimbal angle until the target reaches the preset position, ensuring the reliability of the positioning and correction results of the mobile camera. Attached Figure Description

[0033] Figure 1 This is a flowchart illustrating an embodiment of the mobile camera positioning correction method based on long-distance shooting provided by the present invention.

[0034] Figure 2 This is a flowchart illustrating an embodiment of determining the transformation matrix provided by the present invention;

[0035] Figure 3 This is a flowchart illustrating an embodiment of adjusting a moving camera provided by the present invention;

[0036] Figure 4 A schematic diagram of an embodiment of the mobile camera positioning and correction device based on long-distance shooting provided by the present invention;

[0037] Figure 5 A structural block diagram of an embodiment of the electronic device provided by the present invention. Detailed Implementation

[0038] Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, which form part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not intended to limit the scope of the present invention.

[0039] Currently, in the field of long-distance imaging, especially in aerospace, the complex external environment makes aligning a moving camera difficult, and a one-to-one correspondence between the moving camera and the target is not always present. Therefore, when using a moving camera to photograph a target at a distance, there is a problem that the target may not be in the preset position.

[0040] To address the aforementioned problems, this invention provides a method, apparatus, electronic device, and storage medium for positioning and correction of a mobile camera based on long-distance shooting, which will be described in detail below.

[0041] like Figure 1 As shown, Figure 1 A flowchart illustrating an embodiment of the mobile camera positioning correction method based on long-distance shooting provided by the present invention includes:

[0042] Step S101: Obtain the first position coordinates and the first gimbal angle of the moving camera, capture the target, obtain the first image sample set, and obtain the first feature point set of the first image sample set according to the first feature point model.

[0043] Step S102: Adjust the moving camera to the second position coordinates and the second gimbal angle, shoot the target, obtain the second image sample set, and obtain the second feature point set of the second image sample set according to the second feature point model.

[0044] Step S103: Determine the transformation matrix based on the first feature point set and the second feature point set.

[0045] Step S104: Based on the transformation matrix, the first position coordinates, the first gimbal angle, the second position coordinates, and the second gimbal angle, adjust the moving camera to the target position and the target gimbal angle.

[0046] In this embodiment, firstly, two sets of image sample sets are acquired, and the corresponding shooting position coordinates and gimbal angles are recorded; then, feature point sets of the two sets of image sample sets are acquired respectively through a feature point model; next, based on the feature point sets of the two sets of image sample sets, a transformation matrix between the two sets of image sample sets is determined; finally, based on the transformation matrix and its corresponding position coordinates and gimbal angles, the position coordinates and gimbal angles of the moving camera are adjusted, so that the target image can be accurately positioned at the target position of the moving camera.

[0047] The feature point model automatically obtains the feature point set in two sets of image samples, and obtains the corresponding transformation matrix based on the feature point set. The transformation matrix is ​​used as the basis for adjusting the position coordinates of the moving camera. Through feedback adjustment, the deviation between the adjusted position of the moving camera and the target position coordinates can be effectively reduced.

[0048] In a preferred embodiment, in step S101, since the distance between the target and the moving camera is not constant, and the settings for the target are not fixed, after determining the position coordinates of the moving camera and the angle of the gimbal, in order to effectively extract the target, it is necessary to adjust the magnification of the moving camera, decreasing it from the maximum magnification to the minimum magnification, and then shooting the target in sequence to obtain the first image sample set.

[0049] In other embodiments, after determining the position coordinates of the moving camera and the angle of the gimbal, the target can be photographed sequentially from the minimum magnification to the maximum magnification to obtain a first image sample set; or the first image sample set can be obtained by photographing in other orders.

[0050] Furthermore, after obtaining the first image sample set, it is also necessary to collect the corresponding first feature point set through the first feature point model.

[0051] In one specific embodiment, the first feature point model is a deep learning model, including a convolutional layer, a feature learning layer, and a classification layer. First, a first image sample set is input into the convolutional layer for convolution operations, and a supercomputing algorithm is used to extract feature points from the first image sample set to obtain a first image feature vector. Then, the first image feature vector is input into the feature learning layer to obtain a first image feature matrix. Finally, the first image feature matrix is ​​input into the classification layer to determine the first feature point set.

[0052] In a preferred embodiment, after obtaining the first feature point set in step S102, the first feature point set generally cannot reach the preset position; that is, the first position coordinates and the first gimbal angle cannot meet the shooting requirements of the moving camera. Therefore, it is also necessary to obtain a second feature point set accordingly.

[0053] In one specific embodiment, based on the position of the target in the first image sample set, the position coordinates of the moving camera and the gimbal angle are organically adjusted accordingly, resulting in the second position coordinates and the second gimbal angle. To facilitate comparison with the first image sample set, the differences caused by adjusting the position coordinates and gimbal angle of the moving camera are magnified. The magnification of the moving camera is adjusted in a certain order to capture target images, obtaining the second image sample set. Then, the second feature point set of the second image sample set is obtained based on the second feature point model.

[0054] In one specific embodiment, the second feature point model has the same attributes and structure as the first feature point model, and the process of processing the second image sample set is also exactly the same as that of the first image sample set.

[0055] By identifying the first image sample set and the second image sample set using the first feature point model and the second feature point model structure, the corresponding first feature point set and second feature point set can be directly extracted, which improves the efficiency of obtaining the feature point set.

[0056] Furthermore, after obtaining the first and second feature point sets, in order to adjust the position and angle of the moving camera, it is also necessary to calculate the transformation matrix as the basis for adjustment. In step S103, as follows... Figure 2 As shown, Figure 2 A flowchart illustrating an embodiment of determining a transformation matrix provided by the present invention includes:

[0057] Step S131: Based on the first feature point set and the second feature point set, determine the two most similar matching feature points using the random sample consensus algorithm.

[0058] Step S132: Determine the transformation matrix based on the two sets of most similar matching feature points.

[0059] The Random Sample Consensus (RSC) algorithm is an iterative method used to estimate the parameters of a mathematical model from a set of observations containing outliers, where the outliers do not affect the estimates. Therefore, it can also be considered an outlier detection method. In a sense, it is a nondeterministic algorithm, producing reasonable results with only a certain probability, which increases with the number of iterations.

[0060] In this embodiment, the feature points in the first feature point set and the second feature point set are traversed and examined. The most similar matching feature points in the two sets of feature point sets are determined by the random sample consistency algorithm. That is, the image location of the image target or the same object that can be used as a distinguishing feature in the two sets of image samples is determined. Then, based on the two sets of most similar matching feature points, the transformation matrix is ​​determined by comparison calculation.

[0061] By determining the transformation matrix using two sets of image samples and digitizing the positional relationship between the two sets of image samples, the efficiency and accuracy of subsequent adjustments to the moving camera can be effectively improved.

[0062] Finally, the moving camera is adjusted according to the transformation matrix. Even after obtaining the adjusted moving camera, it's still uncertain whether it has been adjusted to the target position coordinates and gimbal angle. Therefore, the adjustment effect needs to be verified. In step S104, as... Figure 3 As shown, Figure 3 A flowchart illustrating an embodiment of adjusting a moving camera provided by the present invention includes:

[0063] Step S141: Based on the transformation matrix, adjust the moving camera to the third position coordinate and the third gimbal angle, and take pictures of the target to obtain the third image sample set of the target. Then, based on the third feature point model, obtain the third feature point set of the third image sample set.

[0064] Step S142: Based on whether the third feature point set has reached the preset position, determine whether the moving camera has reached the target position and the target gimbal angle. If yes, determine that the third position coordinates and the third gimbal angle are the target position and the target gimbal angle. If not, treat the third feature point set as the second feature point set, update the transformation matrix to the second transformation matrix, and repeatedly adjust the moving camera according to the second transformation matrix until the final feature point set reaches the preset position. Then, determine the target position and the target gimbal angle accordingly based on the final feature point set.

[0065] In this embodiment, firstly, the position coordinates and gimbal angle of the moving camera are adjusted according to the obtained transformation matrix, and a third image sample set is captured based on the adjusted moving camera. Then, based on whether the position of the target image in the third image sample set has reached the preset position, it is determined in reverse whether the moving camera has reached the target position and the target gimbal angle. If yes, the adjustment of the moving camera is stopped, and the position coordinates and gimbal angle of the third image sample set are the final result. If no, a new transformation matrix, namely the second transformation matrix, is generated, and the moving camera is readjusted until the position of the target image in the obtained image sample set reaches the target requirement.

[0066] Generally, the preset position of the target image is set as follows: the target appears within the camera's field of view and occupies more than half of the entire image area. In other words, the image is considered to have reached the preset position only when the target appears within the camera's field of view and occupies more than half of the entire image area. The corresponding position coordinates of the moving camera and the gimbal angle are then the target position coordinates and the target gimbal angle.

[0067] In other embodiments, a point in the target image may be set to reach the exact center of the shooting field of view, or a predetermined point may be set as its preset position.

[0068] In one specific embodiment, since in most cases the operation is to align the moving camera, the relative displacement adjustment is small. Therefore, the second position coordinates and the third position coordinates can be the same as the first position coordinates. That is, in the process of adjusting the shooting position and angle of the moving camera, only the angle of the gimbal is adjusted.

[0069] The above method involves first acquiring two sets of image samples using a moving camera at different coordinates and gimbal angles. The difference between these two sets of images, along with their corresponding coordinates and gimbal angles, is calculated and represented as a transformation matrix. Then, the moving camera is adjusted based on this transformation matrix to obtain a new set of image samples. Finally, the acquired image samples are compared with the target requirements, and the moving camera is continuously adjusted to ultimately obtain an image that meets the target requirements using the moving camera.

[0070] This application adjusts the moving camera based on a transformation matrix. The data in the transformation matrix and the position and angle of the moving camera are in the form of data, presenting a corresponding relationship. Therefore, it can quickly approach the target position. On the other hand, by comparing the relationship between the obtained image sample set and the target requirements, it is determined whether the moving camera has reached the target position. The judgment basis is reliable and can ensure that the moving camera meets the target requirements in subsequent use.

[0071] To address the aforementioned problems, the present invention also provides a mobile camera positioning correction device based on long-distance shooting, such as... Figure 4 As shown, Figure 4 This is a schematic diagram of an embodiment of the mobile camera positioning correction device based on long-distance shooting provided by the present invention. The mobile camera positioning correction device 400 based on long-distance shooting includes:

[0072] The first feature point set acquisition module 401 is used to acquire the first position coordinates and the first gimbal angle of the moving camera, capture the target, obtain the first image sample set, and acquire the first feature point set of the first image sample set according to the first feature point model.

[0073] The second feature point set acquisition module 402 is used to adjust the moving camera to the second position coordinates and the second gimbal angle, capture the target, obtain the second image sample set, and obtain the second feature point set of the second image sample set according to the second feature point model;

[0074] The transformation matrix acquisition module 403 is used to determine the transformation matrix based on the first feature point set and the second feature point set;

[0075] The positioning module 404 is used to adjust the moving camera to the target position and the target gimbal angle based on the transformation matrix, the first position coordinates, the first gimbal angle, the second position coordinates, and the second gimbal angle.

[0076] The present invention also provides an electronic device, such as... Figure 5 As shown, Figure 5 This is a structural block diagram of an embodiment of the electronic device provided by the present invention. The electronic device 500 can be a computing device such as a mobile terminal, desktop computer, laptop, handheld computer, and server. The electronic device 500 includes a processor 501 and a memory 502, wherein the memory 502 stores a mobile camera positioning correction program 503 based on long-distance shooting.

[0077] In some embodiments, memory 502 may be an internal storage unit of a computer device, such as a hard disk or memory. In other embodiments, memory 502 may be an external storage device of a computer device, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc. Furthermore, memory 502 may include both internal and external storage units of the computer device. Memory 502 is used to store application software and various types of data installed on the computer device, such as program code for installing the computer device. Memory 502 may also be used to temporarily store data that has been output or will be output. In one embodiment, a mobile camera positioning correction program 503 based on long-distance shooting can be executed by processor 501 to implement the mobile camera positioning correction method based on long-distance shooting according to various embodiments of the present invention.

[0078] In some embodiments, processor 501 may be a central processing unit (CPU), microprocessor or other data processing chip, used to run program code stored in memory 502 or process data, such as executing a mobile camera positioning correction program based on long-distance shooting.

[0079] This embodiment also provides a computer-readable storage medium storing a mobile camera positioning correction program based on long-distance shooting. When the computer processor executes the program, it implements the mobile camera positioning correction method based on long-distance shooting as described in any of the above technical solutions.

[0080] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0081] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for positioning and correcting a moving camera based on long-distance shooting, characterized in that, include: The first position coordinates and the first gimbal angle of the mobile camera are obtained, the target is photographed, a first image sample set is obtained, and the first feature point set of the first image sample set is obtained according to the first feature point model. Adjust the moving camera to the second position coordinates and the second gimbal angle, capture the target, obtain the second image sample set, and obtain the second feature point set of the second image sample set according to the second feature point model; Determine the transformation matrix based on the first feature point set and the second feature point set; Based on the transformation matrix, the first position coordinates, the first gimbal angle, the second position coordinates, and the second gimbal angle, adjust the moving camera to the target position and the target gimbal angle; The first feature point model and the second feature point model have the same structure, both including convolutional layers, feature learning layers and classification layers, and both are deep learning models; The step of determining the transformation matrix based on the first feature point set and the second feature point set includes: Based on the first set of feature points and the second set of feature points, two sets of most similar matching feature points are determined by a random sample consistency algorithm. The two sets of most similar matching feature points represent the image targets in the first image sample set and the second image sample set, or the image locations of the same object as distinguishing features. The transformation matrix is ​​determined based on the two sets of most similar matching feature points.

2. The method for positioning and correcting a mobile camera based on long-distance shooting according to claim 1, characterized in that, The process of obtaining the first position coordinates and first gimbal angle of the mobile camera, capturing the target, and obtaining the first image sample set includes: Based on the first position coordinates of the mobile camera and the first gimbal angle, the target is photographed sequentially from the maximum magnification to the minimum magnification to obtain the first image sample set.

3. The method for positioning and correcting a mobile camera based on long-distance shooting according to claim 1, characterized in that, The step of obtaining the first feature point set / second feature point set of the first image sample set / second image sample set according to the first feature point model / second feature point model includes: The first image sample set / second image sample set is input into the convolutional layer to perform a convolution operation, thereby obtaining the first image feature vector / second image feature vector. The first image feature vector / the second image feature vector are input into the feature learning layer to obtain the first image feature matrix / the second image feature matrix; The first image feature matrix / second image feature matrix is ​​input into the classification layer to determine the first feature point set / second feature point set.

4. The method for positioning and correcting a mobile camera based on long-distance shooting according to claim 1, characterized in that, The step of adjusting the moving camera to the target position and the target gimbal angle based on the transformation matrix, the first position coordinates, the first gimbal angle, the second position coordinates, and the second gimbal angle includes: According to the transformation matrix, the moving camera is adjusted to the third position coordinates and the third gimbal angle, and the target is photographed to obtain the third image sample set of the target. The third feature point set of the third image sample set is obtained according to the third feature point model. Based on whether the third feature point set has reached the preset position, it is determined whether the moving camera has reached the target position and the target gimbal angle. If yes, the third position coordinates and the third gimbal angle are determined to be the target position and the target gimbal angle. If no, the third feature point set is regarded as the second feature point set, the transformation matrix is ​​updated to the second transformation matrix, and the moving camera is repeatedly adjusted according to the second transformation matrix until the final feature point set reaches the preset position. The target position and the target gimbal angle are then determined accordingly based on the final feature point set.

5. The method for positioning and correcting a moving camera based on long-distance shooting according to claim 4, characterized in that, The second and third position coordinates can be the same as the first position coordinates.

6. A mobile camera positioning and correction device based on long-distance shooting, characterized in that, include: The first feature point set acquisition module is used to acquire the first position coordinates and the first gimbal angle of the mobile camera, capture the target, obtain the first image sample set, and acquire the first feature point set of the first image sample set according to the first feature point model. The second feature point set acquisition module is used to adjust the moving camera to the second position coordinates and the second gimbal angle, capture the target, obtain the second image sample set, and obtain the second feature point set of the second image sample set according to the second feature point model; The transformation matrix acquisition module is used to determine the transformation matrix based on the first feature point set and the second feature point set; The positioning module is used to adjust the moving camera to the target position and the target gimbal angle according to the transformation matrix, the first position coordinates, the first gimbal angle, the second position coordinates, and the second gimbal angle; The first feature point model and the second feature point model have the same structure, both including convolutional layers, feature learning layers and classification layers, and both are deep learning models; The step of determining the transformation matrix based on the first feature point set and the second feature point set includes: Based on the first set of feature points and the second set of feature points, two sets of most similar matching feature points are determined by a random sample consistency algorithm. The two sets of most similar matching feature points represent the image targets in the first image sample set and the second image sample set, or the image locations of the same object as distinguishing features. The transformation matrix is ​​determined based on the two sets of most similar matching feature points.

7. An electronic device, characterized in that, It includes a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, it implements the mobile camera positioning correction method based on long-distance shooting as described in any one of claims 1-5.

8. A storage medium, characterized in that, The storage medium stores computer program instructions, which, when executed by a computer, cause the computer to perform the mobile camera positioning correction method based on long-distance shooting as described in any one of claims 1 to 5.

Images