Aerial photogrammetry data processing method and system and storage medium

Abstract

The invention relates to the field of camera measurement, in particular to an aerial photogrammetry data processing method and system and a storage medium. The method comprises the steps that camera measurement data collected by the aviation equipment in one flight are obtained, and the camera measurement data comprise photos collected by the aviation equipment from multiple positions and multiple angles and POS parameters of the aviation equipment in all areas; wherein the POS parameters comprise the flight height, the longitude and latitude and the attitude of a camera on the aviation equipment, and the plurality of areas form areas through which the flight path of the aviation equipment passes; and interpreting the photos, and removing the photos with abnormal interpretation results. The method and the device have the effect of selecting the data with relatively high accuracy from the aerial photogrammetry data.

Description

Technical Field

[0001] The present application relates to the field of photogrammetry, and in particular to a method, system and storage medium for processing aerial photogrammetry data. Background Art

[0002] Aerial photogrammetry data refers to surface image data obtained through aerial photography. These data are usually used to make maps, topographic maps, 3D models and other geographic information system (GIS) related applications. The full name of aerial triangulation is aerial triangulation in aerial photogrammetry. It refers to the use of images obtained in aerial photogrammetry to restore the position and posture of the camera at the time of photography (i.e. the external parameters of the camera) through certain mathematical models and algorithms, and then restore the three-dimensional coordinates of each point on the ground.

[0003] In the prior art, after obtaining aerial photogrammetry data, the image quality is poor due to the complex ground and shooting environment, which leads to large errors in the subsequent aerial triangulation solution. Therefore, how to select data with higher accuracy from the aerial photogrammetry data has become an urgent problem to be solved. Summary of the invention

[0004] In order to select data with higher accuracy from aerial photogrammetry data, the present application provides a method, system and storage medium for processing aerial photogrammetry data.

[0005] In a first aspect, the present application provides a method for processing aerial photogrammetry data, which adopts the following technical solution:

[0006] A method for processing aerial photogrammetry data comprises the following steps: obtaining the photogrammetry data collected by an aerial device in a flight, the photogrammetry data comprising: photos collected by the aerial device from multiple positions and multiple angles in each region and POS parameters of the aerial device; wherein the POS parameters comprise: flight altitude, longitude and latitude, and the posture of the camera on the aerial device, and the region through which the flight route of the aerial device passes through multiple regions; interpreting the photos and removing photos with abnormal interpretation results; wherein the interpretation method comprises: extracting two frames of photos at a preset time interval, and taking the overlapping region of the two frames of photos as the interpretation region; calculating the abnormal value on the flight path of the aerial device in the interpretation region, constructing a first objective function according to the abnormal value, wherein the first solution set of the first objective function is the interpretation results of each of the multiple frames of photos, and the interpretation results are normal or abnormal, and solving the first objective function according to a group optimization intelligent algorithm; and removing photos with abnormal interpretation results in the first solution set corresponding to the minimum value of the first objective function.

[0007] Optionally, two frames of photos with a preset time interval are extracted, and the overlapping area in the two frames of photos is used as the judgment area, including: obtaining two frames of photos, using an image registration algorithm to spatially align the overlapping area in the two frames of photos for image registration, and the overlapping area of ​​the two frames of photos after registration is the judgment area.

[0008] Optionally, the method for calculating the outlier is: calculating the absolute value of the difference in pixel values ​​between two pixel points at the same position in two frames of photos, traversing all pixel points in the interpretation area, and calculating the average of the absolute values; calculating the information entropy of the pixel points in the next frame of photos; wherein the proportion of the pixel value corresponding to the pixel point in the interpretation area is used as the probability in the information entropy calculation; and the product of the mean and the normalized information entropy is used as the outlier.

[0009] Optionally, the method for calculating the outlier is: calculating the absolute value of the difference in pixel values ​​between two pixel points at the same position in the two frames of photos, traversing all the pixel points in the reading area, and calculating the average of the absolute values; calculating the information entropy of the pixel points in the two frames of photos; wherein the proportion of the pixel values ​​corresponding to the pixel points in the reading area is used as the probability in the information entropy calculation; taking the minimum value of the information entropy corresponding to each pixel point in the two frames of photos for normalization; and taking the product of the mean and the normalized information entropy as the outlier.

[0010] Optionally, a first objective function is constructed according to the outlier value, and the expression of the first objective function is: For the i-th frame photo and the n+i-th frame photo, n is the preset time interval, D n+i represents the first objective function, min represents the minimum function, sum represents the cumulative sum, C 1，x represents the first outlier in the first solution set, C N，x represents the Nth abnormal value in the first solution set, N represents the number of solutions preset in the first solution set, that is, there are N groups of photos, each group of photos includes two frames of photos with a time interval of n; the x value is set to 1 when the judgment result is normal, and the x value is 0 when the judgment result is abnormal, E represents the number of judgment results with an x ​​value of 1, and w represents the preset hyperparameter.

[0011] Optionally, when the number of eliminated photos is greater than a preset elimination threshold, the camera measurement data collected by the aerial equipment in the second flight is obtained, the area of ​​the second flight is the interpretation area in the abnormal photos, the camera parameters are adjusted before the second flight, and the photos collected in the second flight are interpreted according to the interpretation method of the photos in the first flight. If the interpretation result is abnormal, the area is marked.

[0012] Optionally, the method for adjusting camera parameters before the second flight is: determining the starting position of the second flight; adjusting the camera parameters at the starting position to obtain optimal parameters, performing a second flight along the flight path of the first flight, and not adjusting the camera parameters during the second flight.

[0013] Optionally, adjusting the parameters of the camera at the starting position to obtain the optimal parameters includes the steps of: constructing a second objective function of the LM algorithm, taking a second solution set corresponding to the minimum value of the second objective function as the optimal parameters; the expression of the second objective function is: F f =min(C i，f , …, C i-n，f )+D N，f , where F f represents the second objective function, min represents the minimum function, C 1，f represents the outlier value of the i-th frame, C i-n， represents the outlier value of the in frame, D N，f It is the feedback evaluation result, which represents the first objective function value calculated N times after the first photo acquisition with the camera parameters solved by the current iteration.

[0014] In the second aspect, the present application provides a system for processing aerial photogrammetry data, which adopts the following technical solution: a system for processing aerial photogrammetry data, comprising: a processor and a memory, the memory storing computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned method for processing aerial photogrammetry data is implemented.

[0015] In a third aspect, the present application provides a computer-readable storage medium, which adopts the following technical solution: a computer-readable storage medium, in which a computer program is stored, and when the computer program is loaded and executed by a processor, the above-mentioned method for processing aerial photogrammetry data is adopted.

[0016] This application has the following technical effects:

[0017] 1. Perform photo interpretation on aerial photogrammetry data taken at different locations and angles in the same area, and delete photos with abnormal interpretation results. By eliminating abnormal photos, the data for subsequent analysis can be ensured to be more accurate and reliable. Eliminating abnormal data can also reduce the amount of data that needs to be processed in subsequent processing steps, thereby saving computing resources and time.

[0018] 2. According to the situation of abnormal photo removal, determine whether a second flight collection is needed. If necessary, the improved LM algorithm is used to optimize the camera parameters, and the camera parameters are adjusted before flying again to reduce the problem of inaccurate data caused by improper camera parameter settings. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] By reading the detailed description below with reference to the accompanying drawings, the above and other purposes, features and advantages of the exemplary embodiments of the present application will become easily understood. In the accompanying drawings, several embodiments of the present application are shown in an exemplary and non-restrictive manner, and the same or corresponding numbers represent the same or corresponding parts.

[0020] Figure 1 It is a method flow chart of a method for processing aerial photogrammetry data in an embodiment of the present application.

[0021] Figure 2 It is a flow chart of a method for photo interpretation in a method for processing aerial photogrammetry data in an embodiment of the present application. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0023] It should be understood that when the terms "first", "second", etc. are used in the claims, specification and drawings of the present application, they are only used to distinguish different objects, rather than to describe a specific order. The terms "include" and "comprise" used in the specification and claims of the present application indicate the presence of the described features, wholes, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or their collections.

[0024] The present application embodiment discloses a method for processing aerial photogrammetry data, which can interpret aerial photogrammetry data taken at different positions and angles in the same area and eliminate aerial photogrammetry data with abnormal interpretation. Figure 1 , including steps S1 to S3, which are specifically as follows:

[0025] S1: Acquire the video measurement data collected by the aerial equipment in a flight, the video measurement data including: photos collected by the aerial equipment from multiple positions and angles in each area and POS parameters of the aerial equipment.

[0026] In one embodiment, the aerial photogrammetry data is stored in the aerial equipment without real-time data transmission, so as to reduce the interruption of data transmission due to high flight altitude or poor base station signal and reduce the probability of information loss.

[0027] When aerial equipment is collecting aerial photogrammetry data, it is necessary for the aerial equipment to interpret the aerial photogrammetry data to prevent local aerial photogrammetry data from being too poor.

[0028] In one embodiment, the aerial photogrammetry data includes: oblique photography data and POS data. The oblique photography data are photos of the same area taken from multiple angles by multiple cameras. The POS data include: the flight altitude and longitude and latitude of the aerial equipment, the attitude of the aerial equipment camera and control point data. The control points are calibration points built on the ground for calibrating the longitude and latitude and altitude information of the aerial equipment.

[0029] When interpreting the oblique photography data in the aerial photogrammetry data, since the images (photos) collected in different areas are different, it is necessary to use photos of different positions and angles of the same area to interpret the oblique photography data. Multiple areas constitute the area through which the flight route of the aerial equipment passes, and multiple local paths constitute the flight route. This application takes one area as an example for analysis, and other areas will not be repeated.

[0030] S2: Interpret the photos and remove those with abnormal results. Figure 2 The method for photo interpretation includes steps S20 to S23, which are as follows:

[0031] S20: extracting two frames of photos at a preset time interval, and taking the overlapping area of ​​the two frames of photos as the judgment area.

[0032] Two frames of photos are obtained, and the overlapping areas in the two frames of photos are spatially aligned using an image registration algorithm to perform image registration. The overlapping area of ​​the two frames of photos after registration is the interpretation area.

[0033] Since it is impossible to ensure that the overlapping areas in the oblique photography data collected twice are exactly the same, the image registration algorithm is used to register the overlapping areas in the two collected photos. The image registration algorithm can align the spatial position of two or more images containing the same scene, so that the same features in multiple images are in the same position as much as possible. The image matching algorithm is a prior art and will not be described in detail here.

[0034] Set the time interval to n. When the aerial equipment maintains a uniform flight speed, the aerial equipment usually flies faster, at a higher altitude, and the imaging frame rate is often not less than 30 frames, resulting in a large overlap in the continuously collected oblique photography data. When overexposure or darkness occurs, the photos collected continuously will have the same overexposure or darkness problem, that is, the changes between the continuously collected photos are not obvious. Therefore, this application does not use continuous adjacent images to interpret different oblique photography data in the same area, and selects oblique photography data collected at a time interval of n times for photo judgment, that is, compares the currently collected photos with the photos collected n times later to analyze the changes in different oblique photography data in the same area. The value of n can be adjusted by the implementer according to the specific implementation scenario. This solution takes n=10. Then obtain the i-th frame and the n+i-th frame photo.

[0035] In one embodiment, the area of ​​the overlapping region in the two frames of photos of the present application is required to be no less than 30% of the area of ​​the photos. The larger n is, the smaller the overlapping area is. The value of 30% can be adjusted by the implementer according to the specific implementation scenario. The area of ​​the overlapping region can be adjusted by adjusting the value of n.

[0036] S21: Calculate the abnormal values ​​on the flight path of the aviation equipment in the judgment area, and construct a first objective function according to the abnormal values. The first solution set of the first objective function is the judgment results of each of the multiple frames of photos, and the judgment results are normal or abnormal.

[0037] In one embodiment, the calculation method of the outlier is: calculate the absolute value of the difference between the pixel values ​​of two pixels at the same position in the two frames of photos, traverse all the pixels in the judgment area, and calculate the average of the absolute values, which is recorded as A. i，n+i ; Calculate the information entropy of the pixels in the next frame of the photo, denoted as B n+i ; Among them, the proportion of the pixel value corresponding to the pixel point in the judgment area is used as the probability in the information entropy calculation; the product of the mean and the normalized information entropy is used as the abnormal value. The calculation method of information entropy is an existing technology and will not be described here.

[0038] The mathematical expression of the outlier is: C i，n+i =A i，n+i ×exp(-B n+i ), where C i，n+i The abnormal values ​​of the judgment area on the i-th frame and the n+i-th frame photos are represented by negative correlation mapping using the exp(-x) function. The calculation method is well known and will not be repeated here.

[0039] In one embodiment, in the above mathematical expression of the abnormal value, B n+i It can also be: the minimum value of the information entropy corresponding to each pixel in two frames of photos (the i-th frame and the n+i-th frame of photos).

[0040] C i，n+i The larger the (C value), the more it should be eliminated during photo interpretation. However, when eliminating, it is necessary to determine whether to eliminate the i-th frame of photo or the i+n-th frame of photo. If the elimination value is too large, more oblique photography data will be lost during elimination, which will affect the quality of ground modeling restoration.

[0041] Then, this application constructs the first objective function D n+i , to determine the abnormal value corresponding to the local path of the current aviation equipment in the continuous flight, and the method for constructing the first objective function is as shown in step S22.

[0042] S22: Construct a first objective function according to the abnormal value, and solve the first objective function according to the group optimization intelligent algorithm. The expression of the first objective function is:

[0043] For the i-th frame photo and the n+i-th frame photo, n is the preset time interval, D n+i represents the first objective function, min represents the minimum function, sum represents the cumulative sum, C 1，x represents the first outlier in the first solution set, C N，x It represents the Nth outlier in the first solution set, and N represents the number of preset solutions in the first solution set, that is, there are N groups of photos, and each group of photos includes two frames of photos with a time interval of n.

[0044] Exemplarily, N=20 means that oblique photography data collected 20 times in succession is used to evaluate the corresponding outliers on the local path for the purpose of eliminating and interpreting the photos. N=20 is a hyperparameter and has nothing to do with the above n=10. N can be adjusted by the implementer according to the specific implementation scenario.

[0045] w represents a preset hyperparameter, which is used to balance the weights of the two parts before and after the plus sign in the first objective function formula, w=2.6, and can be adjusted by the implementer according to the specific implementation scenario.

[0046] Set the x value to 1 when the judgment result is normal, and the x value to 0 when the judgment result is abnormal. E represents the number of judgment results with the x value of 1. If x=1, the photo is retained, and if x=0, the photo is eliminated. In order to reduce the situation where continuous elimination causes excessive information loss, set To prevent excessive local information loss.

[0047] In the above objective function D n+i When solving the minimum value, the oblique photography interval restriction of x=1 is performed so that the interval between any two oblique photography data of x=1 is not greater than n, so as to prevent the continuous interval from being too large, and further reduce the probability of excessive continuous loss.

[0048] Accumulate the abnormal values ​​corresponding to the first solution set to get sum(C 1，x , ..., C N，x＝20 ), sum(C 1，x , ..., C N，x＝20 ), it means that after removing some photos, the corresponding outlier value of the current aerial equipment on the local path of continuous flight is lower.

[0049] In the process of iterative solution of the first objective function, for the outlier C 1，x , ..., C N，x＝20 Perform iterative calculations. In the first iteration, if x=1 for the 1st frame photo, x=1 for the 9th frame photo, x=0 for the 10th frame photo, and x=0 for the 11th frame photo, since x=0 for the 10th and 11th frames photo, the 10th and 11th frames need to be removed. In the second iteration, the 11th frame photo has been removed, so for the 1st frame photo, select the photo with x=1 before the 11th frame photo and closest to the 11th frame photo, that is, the 9th frame photo and the 1st frame photo to calculate the outlier value.

[0050] In order to reduce the amount of calculation, the threshold C is set r , when there is an abnormal value of the collected oblique photography data greater than C r When , the calculation of the first objective function is performed after the subsequent N times of tilt photography data collection. r =1.2, which can be adjusted by the implementer according to the specific implementation scenario.

[0051] Exemplarily, the swarm optimization intelligent algorithm uses a particle swarm algorithm. Other swarm optimization intelligent algorithms that are applicable to this scenario may also be used. The particle swarm algorithm is a prior art and will not be described in detail here.

[0052] S23: Eliminate photos with abnormal results from the first solution set corresponding to the minimum value of the first objective function.

[0053] S3: When the number of discarded photos is greater than the preset rejection threshold, obtain the camera measurement data collected by the aerial equipment in the second flight. The area of ​​the second flight is the interpretation area in the abnormal photos. Adjust the camera parameters before the second flight, and interpret the photos collected in the second flight according to the interpretation method of the photos in the first flight. If the interpretation result is abnormal, mark the area.

[0054] Step S3 includes the following steps:

[0055] Set the rejection threshold D r , if the minimum value is greater than the rejection threshold D r, it is considered that too many photos have been removed, which may cause serious information loss, so a second flight is needed to collect more data. Otherwise, no second flight is performed.

[0056] Determine the starting position of the second flight. If a second flight is performed, at the end of the first flight, the aerial device returns to the starting position of the second flight along the flight path of the first flight at the end point of the first flight. The starting position of the second flight is the starting point of the route segment in the first eliminated photo. The camera parameters are adjusted at the starting position to obtain the optimal parameters, and the second flight is performed along the flight path of the first flight. The camera parameters are not adjusted during the second flight.

[0057] When collecting data during the second flight, if the camera parameters used in the first flight are used, the same oblique photography data collection results may be obtained. Therefore, adjusting the camera parameters at the starting point to optimize the camera parameters before performing a second flight collection can eliminate the phenomenon of being too dark or overexposed due to camera parameters.

[0058] In the prior art, the existing LM (Levenberg-Marquardt) algorithm can be used to solve the camera parameters. However, when the LM algorithm solves the camera parameters, its objective function is to solve the minimum reprojection error. However, the existing minimum reprojection error is calculated based on the calibration results and the shooting results. However, in aerial photography, there is no calibration object, which is not suitable for the application scenario of this application. Therefore, this application chooses to construct the second objective function of the LM algorithm, and takes the second solution set corresponding to the minimum value of the second objective function as the optimal parameter.

[0059] The expression of the second objective function is: F f =min(C i，f , ..., C i-n，f )+D N，f , where F f represents the second objective function, min represents the minimum function, C 1，f represents the outlier value of the i-th frame, C i-n，f Indicates the outlier value of frame in.

[0060] min(C 1，f , ..., C n＝10，f ) is smaller, indicating that the camera parameters corresponding to the current second solution set are better. In one embodiment, the camera parameters refer to the internal and external parameters of the camera, and the internal parameters include: focal length, principal point, radial distortion coefficient, and tangential distortion coefficient. The external parameters include: rotation matrix and translation vector.

[0061] Among the intrinsic parameters, the focal length determines the magnification of the image. In the image coordinate system, the focal length is usually expressed in pixels. The principal point is a point on the image plane, usually the center of the image, but it may not be at the center due to lens distortion. The radial distortion coefficient expresses the radial distortion, that is, the stretching or compression of the image edge. The tangential distortion coefficient expresses the tangential distortion, that is, the tilt of the image edge.

[0062] Among the external parameters, the rotation matrix expresses the rotation of the camera coordinate system relative to the world coordinate system, and the translation vector expresses the translation of the camera coordinate system relative to the world coordinate system.

[0063] D N，f is the feedback evaluation result, which represents the first objective function value (D value) calculated N times after the first photo acquisition with the current iteratively solved camera parameters. If the quadratic objective function only considers min(C i，f , ..., C i-n，f ), the camera parameters do not implement the update of the D value, and it is easy to fall into a local optimal solution, making it difficult to obtain better camera parameters.

[0064] In one embodiment, the small and large round iterations are set, and the small round iteration refers to the iteration within G=8 times since the last large round iteration, using min(C i，f , ..., C i-n，f ) Partially to F f Solve it, and at G+1, use the iterated camera parameters to collect subsequent oblique photography data to calculate D N，f , G+1 is a large round, and after the large round, G=8 small rounds of iteration are continued until the iteration stops. The iteration stopping condition is: the camera parameters corresponding to the optimal solutions of consecutive large and small rounds remain unchanged or the number of iterations reaches 50. The number of iterations can be adjusted by the implementer according to the specific implementation scenario.

[0065] If the adjusted camera parameter values ​​are the same as those of the first flight, the data anomaly may not be caused by camera parameter problems. Keep the camera parameters used in the first flight and conduct a second flight. Interpret the photos collected in the second flight according to the interpretation method of the photos in the first flight.

[0066] If the number of photos eliminated by the second reading is the same as that eliminated by the first reading, or the number of photos eliminated by the second reading is different from that eliminated by the first reading but the number of photos eliminated by the second reading is still greater than the elimination threshold, the photos and POS data corresponding to the area are marked. Otherwise, the reading result with a smaller number of eliminated photos is selected and retained, and the photos and corresponding POS data retained after being eliminated in the reading result are used as the better aerial photogrammetry data.

[0067] An embodiment of the present application also discloses a system for processing aerial photogrammetry data, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein when the processor executes the computer program, the method for processing aerial photogrammetry data in the above embodiment is adopted.

[0068] The system may adopt computer devices such as desktop computers, laptop computers or cloud servers, and the system includes but is not limited to processors and memories. For example, the system may also include input and output devices, network access devices and buses.

[0069] Among them, the processor can adopt a central processing unit (CPU). Of course, according to actual usage, other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. can also be adopted. The general-purpose processor can adopt a microprocessor or any conventional processor, etc., and this application does not impose any restrictions on this.

[0070] Among them, the memory can be an internal storage unit of the system, such as the hard disk or memory of the system, or it can be an external storage device of the system, such as a plug-in hard disk, smart memory card (SMC), secure digital card (SD) or flash memory card (FC) equipped on the system, and the memory can also be a combination of the internal storage unit and the external storage device of the system. The memory is used to store computer programs and other programs and data required by the system. The memory can also be used to temporarily store data that has been output or is to be output, and this application does not impose any restrictions on this.

[0071] Among them, through this system, the processing method of aerial photogrammetry data in the above embodiment is stored in the memory of the system, and is loaded and executed on the processor of the system, which is convenient for use.

[0072] The embodiment of the present application further discloses a computer-readable storage medium, and the computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, the method for processing aerial photogrammetry data in the above embodiment is adopted.

[0073] Among them, the computer program can be stored in a computer-readable medium, the computer program includes computer program code, the computer program code can be in the form of source code, object code, executable file or certain middleware, etc. The computer-readable medium includes any entity or device that can carry the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the computer-readable medium includes but is not limited to the above-mentioned components.

[0074] Among them, through this computer-readable storage medium, the processing method of aerial photogrammetry data in the above embodiment is stored in a computer-readable storage medium, and is loaded and executed on a processor to facilitate the storage and application of the above method.

[0075] The above are all preferred embodiments of the present application, and the protection scope of the present application is not limited thereto. Therefore, any equivalent changes made according to the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims

1. A method for processing aerial photogrammetry data, characterized in that: Includes steps: Acquire the photogrammetry data collected by the aerial equipment during a flight. The photogrammetry data includes: photos collected by the aerial equipment from multiple positions and angles in each area and the POS parameters of the aerial equipment; the POS parameters include: flight altitude, longitude and latitude, and the attitude of the camera on the aerial equipment. Multiple areas constitute the area through which the flight route of the aerial equipment passes; The photos are interpreted and the photos with abnormal interpretation results are eliminated; wherein the interpretation method is: extract two frames of photos with a preset time interval, and use the overlapping area of ​​the two frames of photos as the interpretation area; calculate the abnormal values ​​on the flight path of the aviation equipment in the interpretation area, and construct a first objective function according to the abnormal values. The first solution set of the first objective function is the interpretation results of each of the multiple frames of photos. The interpretation results are normal or abnormal. The first objective function is solved according to the group optimization intelligent algorithm; and the photos with abnormal interpretation results in the first solution set corresponding to the minimum value of the first objective function are eliminated.

2. The method for processing aerial photogrammetry data according to claim 1, characterized in that: Extract two frames of photos at a preset time interval, and use the overlapping area of ​​the two frames of photos as the judgment area, including: obtaining two frames of photos, using an image registration algorithm to spatially align the overlapping areas of the two frames of photos for image registration, and the overlapping area of ​​the two frames of photos after registration is the judgment area.

3. The method for processing aerial photogrammetry data according to claim 1, characterized in that: The calculation method of the outlier value is: Calculate the absolute value of the difference between the pixel values ​​of two pixels at the same position in the two frames of photos, traverse all the pixels in the judgment area, and calculate the average of the absolute values; Calculate the information entropy of the pixel points in the next frame of the photo; wherein the proportion of the pixel value corresponding to the pixel point in the reading area is used as the probability in the information entropy calculation; The product of the mean and the normalized information entropy is taken as the outlier.

4. The method for processing aerial photogrammetry data according to claim 1, characterized in that: The calculation method of the outlier value is: Calculate the absolute value of the difference between the pixel values ​​of two pixels at the same position in the two frames of photos, traverse all the pixels in the judgment area, and calculate the average of the absolute values; Calculate the information entropy of the pixels in the two frames of photos; the proportion of the pixel value corresponding to the pixel in the reading area is used as the probability in the information entropy calculation; Take the minimum value of the information entropy corresponding to each pixel in the two frames of photos and normalize them; The product of the mean and the normalized information entropy is taken as the outlier.

5. The method for processing aerial photogrammetry data according to claim 3 or 4, characterized in that: The first objective function is constructed according to the outliers. The expression of the first objective function is: For the i-th frame photo and the n+i-th frame photo, n is the preset time interval, D n+i represents the first objective function, min represents the minimum function, sum represents the cumulative sum, C 1，x represents the first outlier in the first solution set, C N，x represents the Nth abnormal value in the first solution set, N represents the number of solutions preset in the first solution set, that is, there are N groups of photos, each group of photos includes two frames of photos with a time interval of n; the x value is set to 1 when the judgment result is normal, and the x value is 0 when the judgment result is abnormal, E represents the number of judgment results with an x ​​value of 1, and w represents the preset hyperparameter.

6. The method for processing aerial photogrammetry data according to claim 1, characterized in that: The method also includes the following steps: when the number of discarded photos is greater than a preset discarding threshold, obtaining the camera measurement data collected by the aerial equipment in the second flight, the area of ​​the second flight is the interpretation area in the abnormal photos, adjusting the camera parameters before the second flight, interpreting the photos collected in the second flight according to the interpretation method of the photos in the first flight, and marking the area if the interpretation result is abnormal.

7. The method for processing aerial photogrammetry data according to claim 6, characterized in that: The method for adjusting the camera parameters before the second flight is: determining the starting position of the second flight; adjusting the camera parameters at the starting position to obtain the optimal parameters, performing the second flight along the flight path of the first flight, and not adjusting the camera parameters during the second flight.

8. The method for processing aerial photogrammetry data according to claim 7, characterized in that: Adjusting the parameters of the camera at the starting position to obtain the optimal parameters includes the steps of: constructing a second objective function of the LM algorithm, and taking a second solution set corresponding to the minimum value of the second objective function as the optimal parameters; The expression of the second objective function is: F f =min(C i，f , ..., C i-n，f )+D N，f , where F f represents the second objective function, min represents the minimum function, C 1，f represents the outlier value of the i-th frame, C i-n，f represents the outlier value of the in frame, D N，f It is the feedback evaluation result, which represents the first objective function value calculated N times after the first photo acquisition with the camera parameters solved by the current iteration.

9. A system for processing aerial photogrammetry data, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the method for processing aerial photogrammetry data according to any one of claims 1 to 8 is implemented.

10. A computer-readable storage medium having a computer program stored therein, characterized in that: When the computer program is loaded and executed by the processor, the method for processing aerial photogrammetry data according to any one of claims 1 to 8 is adopted.

Images