A method and system for optimizing the pose parameters of a UAV during aerial photography

Abstract

This invention relates to the field of UAV navigation and high-precision surveying, and discloses a method and system for optimizing pose parameters during UAV aerial photography. The method includes: determining key target points and surrounding auxiliary points during UAV aerial photography, and collecting the UAV's geometric configuration parameters to be optimized; establishing a coefficient matrix based on the UAV's geometric configuration parameters to be optimized, and calculating the error equations for the pixel coordinates of the key target points and surrounding auxiliary points during UAV aerial photography; constructing a cofactor matrix based on the coefficient matrix, and calculating the positioning error coefficients of the key target points during UAV aerial photography; using the positioning error coefficients as the optimization objective function, determining the minimum value of the optimization objective function through a collaborative search algorithm to obtain the optimal configuration parameters; and calculating the UAV pose parameters corresponding to the optimal configuration parameters. Using this invention, by constructing an evaluation system centered on the accuracy of key target points, the UAV's aerial photography pose parameters are efficiently optimized, maximizing the positioning accuracy of key target points.

Description

Technical Field

[0001] This invention relates to the field of UAV navigation and high-precision surveying technology, specifically to a method and system for optimizing pose parameters during UAV aerial photography. Background Technology

[0002] With the development of UAV technology, UAV aerial photography has become a core means of high-precision photogrammetry and is widely used in scenarios such as monitoring key building locations, locating geological sampling points, and fine mapping of cultural relics. Its core technology is to obtain multi-angle images of the target area by UAVs and calculate the three-dimensional coordinates of ground points to provide accurate location data for engineering tasks.

[0003] In practical applications, users' core needs often focus on high-precision positioning of key target points (such as building crack monitoring points, geological hazard markers, and core feature points of cultural relics), rather than the average accuracy of all ground points within a region. However, current UAV orbital configuration optimization technologies generally suffer from the limitation of "emphasizing regional balance while neglecting single-point precision," making it difficult to meet these needs. Specific shortcomings are as follows:

[0004] The lack of specificity in single-point accuracy and the mismatch with core requirements: Existing solutions mostly aim for "balanced positioning accuracy across multiple ground points in a region". This "mean optimization" can easily lead to the accuracy of a single key target point being "diluted" by non-key points. In order to improve the accuracy of edge points, the geometric strength of image intersection of key points may be sacrificed, ultimately causing the error of core points to exceed the allowable range of engineering, and failing to meet the high-precision requirements of single points.

[0005] Poor scene adaptability and inability to meet diverse needs: Different scenarios have significantly different accuracy requirements for target points. Planar surveying requires prioritizing planar accuracy, while settlement monitoring requires a focus on improving elevation accuracy. However, existing technologies have fixed optimization logic and lack a "customized optimization mechanism," making it impossible to adjust targets based on "planar priority" or "elevation priority." This results in poor adaptability of the same configuration in different scenarios, necessitating repeated design of optimization logic, which is inefficient.

[0006] Lack of efficient configuration search mechanism: Current methods for determining configuration mostly rely on empirical formulas or enumeration methods. Empirical formulas only cover simple scenarios and cannot cope with multi-parameter collaborative optimization. Enumeration methods need to traverse a large number of parameter combinations. When faced with high-dimensional problems of "massive imagery + multiple ground key points", the search efficiency is low and the optimal configuration is easily missed, making it difficult to meet engineering needs.

[0007] In summary, existing technologies cannot provide precise configuration optimization for "key target points," nor do they have a means to efficiently search for the optimal configuration to meet different needs. Therefore, there is an urgent need for a method for determining the geometric configuration of UAV aerial photography that focuses on "maximizing the accuracy of key target points," supports customized optimization for multiple scenarios, and can efficiently search for optimal parameters. Summary of the Invention

[0008] This invention provides a method and system for optimizing pose parameters during UAV aerial photography. By constructing an evaluation system with the accuracy of key target points as the core, the method optimizes the position parameters and attitude angle parameters of the UAV to maximize the positioning accuracy of key target points.

[0009] Therefore, the present invention provides the following technical solution:

[0010] A method for optimizing pose parameters during UAV aerial photography, the method comprising:

[0011] Step 1: Determine the key target points and surrounding auxiliary points during UAV aerial photography, and collect the UAV's geometric configuration parameters to be optimized;

[0012] Step 2: Based on the geometric configuration parameters of the UAV to be optimized, establish a coefficient matrix and calculate the error equations for the pixel coordinates of key target points and surrounding auxiliary points during UAV aerial photography.

[0013] Step 3: Based on the coefficient matrix, construct the cofactor matrix and calculate the positioning error coefficients of key target points during UAV aerial photography;

[0014] Step 4: Using the positioning error coefficients of the key target points as the optimization objective function, determine the minimum value of the optimization objective function through a collaborative search algorithm to obtain the optimal configuration parameters;

[0015] Step 5: Calculate the UAV pose parameters corresponding to the optimal configuration parameters based on the optimal configuration parameters.

[0016] Optionally, in step 1, given key target points and M surrounding auxiliary points k=1,2,…M, key target points These are the key locations that need to be photographed during drone aerial photography, and their coordinates are shown below. , M surrounding auxiliary points Evenly distributed at key target points All around.

[0017] Optionally, in step 1, the UAV is set to orbit around the key target point. The number of images captured is N, and the drone is capturing images of key target points. The geometric configuration parameters to be optimized for the i-th image are: horizontal distance ,position ,high With attitude angle , , , i=1,2,…N; where, horizontal distance Key target points Planar distance from the location of the drone when it captured the i-th image; orientation Key target points The azimuth and altitude of the image taken by the drone when capturing the i-th image. Key target points The height difference between the image taken by the drone and the image taken at the i-th time.

[0018] Optionally, in step 2, based on the drone's shooting of key target points The attitude angle of the drone in the i-th image, and the drone's position when shooting key target points. Key target points in the i-th image and M surrounding auxiliary points Calculate the parameter values ​​in the coefficient matrix based on the coordinate values ​​in the x, y, and z directions, and then use the coefficient matrix and the data from the UAV when capturing key target points. Key target points in the i-th image and M surrounding auxiliary points Error equations are used to calculate the pixel coordinates of key target points and surrounding auxiliary points during UAV aerial photography, based on coordinate corrections and attitude angle corrections in the x, y, and z directions.

[0019] Optionally, in step 3, the cofactor matrix constructed based on the coefficient matrix B is:

[0020] ,

[0021] in, : Cofactor sub-block of the coordinates of key target points and surrounding auxiliary points;

[0022] : The cross-correlation factor sub-block between the coordinate parameters of key target points and surrounding auxiliary points and the image attitude angle parameters;

[0023] : The cross-correlation factor sub-block between image attitude angle parameters and the coordinate parameters of key target points and surrounding auxiliary points;

[0024] The cofactor sub-block of the attitude angle parameters of the key target point and surrounding auxiliary points is the cofactor matrix. The submatrix contains only attitude angles ( , , Cofactor information of the parameters;

[0025] use Represents the j-th point The cofactor matrix, take , , ,in, , , for Chinese correspondence The diagonal elements; j = 0, 1, 2, ..., M;

[0026] Kxyz j : The j-th point P j The three-dimensional positioning error coefficient comprehensively reflects the position of point P. j Overall positioning accuracy in the X, Y, and Z directions;

[0027] Kxy j : The j-th point P j The planar positioning error coefficient reflects the position of point P. j In X Positioning accuracy in the Y-plane;

[0028] Kz j : The j-th point P j The elevation positioning error coefficient reflects the position of point P. j Positioning accuracy in the Z direction, i.e., the vertical direction;

[0029] :point The cofactor of the X-axis coordinate reflects the accuracy of the X-axis coordinate.

[0030] :point The cofactor of the Y-axis coordinate reflects the accuracy of the Y-axis coordinate;

[0031] :point The cofactor of the Z-axis coordinate reflects the accuracy of the Z-axis coordinate.

[0032] Optionally, step 4 includes:

[0033] Step 41: Encode the UAV's geometric configuration. Each geometric configuration corresponds to one set of UAV orbital configuration parameters, with a dimension of 3. N +3 N =6 N Wei, that is N indivual , N indivual , N indivual 3 N An attitude angle; N is the number of images captured by the drone;

[0034] Step 42, set key target points As the first point in the coefficient matrix, define the optimal objective function:

[0035] ,

[0036] Key Target Points The three-dimensional positioning error coefficient comprehensively reflects the key target points. Overall positioning accuracy in the X, Y, and Z directions;

[0037] Key Target Points The planar positioning error coefficient reflects the critical target point In X Positioning accuracy in the Y-plane;

[0038] Key Target Points The elevation positioning error coefficient reflects the key target points Positioning accuracy in the Z direction, i.e., the vertical direction;

[0039] Step 43: Perform iterative optimization with the goal of finding the geometric configuration that minimizes the optimal objective function;

[0040] Step 44: When the iteration reaches 500 times or the population has the optimal configuration parameter set in the previous t iterations... Changes <10 in 30 consecutive iterations 6 Stop the iteration and obtain the optimal configuration parameters.

[0041] Optionally, step 43 includes:

[0042] Step a: Initialize the geometry by generating 80 geometric configurations and initializing the configuration parameter groups and parameter adjustment strategies for the geometric configurations.

[0043] Step b: For each geometric configuration, calculate the positioning error coefficient using the methods in steps 1-3;

[0044] Step c: Update the optimal value and record the historical optimal value for each geometric configuration. Globally optimal with all configurations ;

[0045] Step d: Perform parameter adjustment strategy and configuration parameter group update:

[0046] ,

[0047] ,

[0048] in, and These are the parameter adjustment strategy vector and configuration parameter set vector for the geometric configuration e in the t-th iteration, respectively. Adjust the parameter strategy vector for the next geometric configuration iteration; This is the configuration parameter set vector for the next geometric configuration iteration; The persistence factor determines the degree to which the algorithm parameter adjustment strategy update is affected by the current geometric configuration parameter adjustment strategy; the individual influence factor c1 and the group influence factor c2 determine the degree to which the geometric configuration's own state and the historical best state affect the next parameter adjustment strategy update; r1 and r2 are random numbers in the interval [0,1], used to increase the randomness of the search; Let be the optimal configuration parameter set for geometric configuration e in the first t iterations; The optimal configuration parameter set for the population in the first t iterations is the optimal configuration parameter set.

[0049] Optionally, in step 5, the optimal configuration parameters include: horizontal distance. ,position ,high With attitude angle parameters , , The formula for calculating the three-dimensional coordinates in the UAV pose parameters based on the optimal configuration parameters is as follows:

[0050] ,

[0051] ,

[0052] .

[0053] A pose parameter optimization system for UAV aerial photography, the system comprising:

[0054] Initialize the unit to determine the key target points and surrounding auxiliary points during UAV aerial photography, and collect the UAV's geometric configuration parameters to be optimized;

[0055] The error equation calculation unit establishes a coefficient matrix based on the UAV's geometric configuration parameters to be optimized, and calculates the error equations for the pixel coordinates of key target points and surrounding auxiliary points during UAV aerial photography.

[0056] The positioning error coefficient calculation unit constructs a cofactor matrix based on the coefficient matrix and calculates the positioning error coefficient of key target points during UAV aerial photography.

[0057] The iterative optimization unit uses the positioning error coefficient of the key target point as the optimization objective function, and determines the minimum value of the optimization objective function through a cooperative search algorithm to obtain the optimal configuration parameters;

[0058] The pose parameter calculation unit calculates the UAV pose parameters corresponding to the optimal configuration parameters based on the optimal configuration parameters.

[0059] A computer-readable storage medium having a computer program stored thereon, the computer program being executed by a processor to perform the steps of the pose parameter optimization method for UAV aerial photography.

[0060] The UAV aerial photography pose parameter optimization method and system provided by this invention solves the problems of low positioning accuracy of key target points and insufficient targeted configuration optimization. By constructing an evaluation system centered on the accuracy of key target points, it breaks through the traditional "regional mean optimization" approach and constructs an evaluation model with the three-dimensional positioning error coefficient of key target points as the sole optimization objective, focusing on the high-precision requirements of specific points. It integrates the pose parameter encoding strategy of photogrammetric adjustment, transforming the UAV aerial photography pose parameters (flight distance, flight altitude, and attitude angle for each image) into quantifiable optimization variables, directly linking them to the coefficient matrix and cofactor matrix in the adjustment model, achieving a precise mapping between pose parameters and positioning accuracy. This invention can significantly improve the accuracy of key target points. Through directional optimization, the positioning accuracy of key target points is significantly improved compared to traditional regional optimization methods. Compared with existing technologies, this invention has the following advantages:

[0061] More targeted: Existing technologies are mostly geared towards the "average accuracy" of large-scale terrain mapping, while this invention focuses on the high-precision orientation of key target points, filling the technical gap in precise optimization of specific locations.

[0062] Higher accuracy controllability: The pose parameters and positioning accuracy are directly linked by the cofactor matrix. During the optimization process, the accuracy changes of key target points can be quantified in real time, while existing technologies mostly rely on experience to adjust the configuration.

[0063] More flexible in adapting to different scenarios: It supports switching between 3D, planar and elevation accuracy requirements, and can adapt to customized aerial photography needs in multiple fields. Existing technologies are mostly optimized for single scenarios. Attached Figure Description

[0064] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly described below. Obviously, the drawings described below are merely some embodiments of the present invention, and those skilled in the art can obtain other drawings based on these drawings without creative effort.

[0065] Figure 1 This is a flowchart of a method for optimizing pose parameters during UAV aerial photography in a specific embodiment of the present invention;

[0066] Figure 2This is a schematic diagram of the structure of the UAV pose parameter optimization system in a specific embodiment of the present invention. Detailed Implementation

[0067] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0068] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0069] like Figure 1 As shown, Figure 1 This is a flowchart of a method for optimizing pose parameters during UAV aerial photography according to a specific embodiment of the present invention. The method for optimizing pose parameters during UAV aerial photography includes:

[0070] Step 1: Determine the key target points and surrounding auxiliary points for drone aerial photography, and set the geometric configuration parameters of the drone to be optimized.

[0071] Key target points in drone aerial photography refer to the critical locations that need to be photographed. For example, in monitoring the structural health of buildings, key target points are the monitoring points for building cracks. In identifying potential geological hazards, key target points are the geological hazard markers. In the digital preservation of cultural heritage, key target points are the core feature points of the cultural relics.

[0072] Given key target points and M surrounding auxiliary points (k=1,2,…M), where Coordinates are represented as , ); surrounding auxiliary points It should be evenly distributed in All around.

[0073] The geometric configuration parameters of a drone refer to a series of quantitative indicators used to define the drone's physical shape, dimensions, relative positions of components, and key angles. Assume the drone orbits... The number of images captured is N, and the drone is capturing images of key target points. The geometric configuration parameters to be optimized for the i-th image are (horizontal distance) ,position ,high ) and attitude angle ( , , (i=1,2,…N). Key target points The planar distance from the image i captured by the drone; Key target points The azimuth and altitude of image i when it is captured by the drone. Key target points The height difference between image i and image i taken by the drone (the height of image i taken by the drone minus) high).

[0074] The three-dimensional coordinates of the UAV when capturing the i-th image satisfy:

[0075] ,

[0076] ,

[0077] ,

[0078] in, ( , ) represent the three-dimensional coordinate components of the camera center, For horizontal distance variables, α This refers to the horizontal field of view of the drone camera.

[0079] Step 2: Calculate the coefficient matrix and construct the error equation;

[0080] The error equation is:

[0081] ,

[0082] Where B is the coefficient matrix; X is the parameter correction vector; and L is the constant term vector.

[0083] ,

[0084] ,

[0085] ,

[0086] ,

[0087] No. i Zhang's drone image is the first j Image point coordinates of a ground point ( j =0,1,..., M The error equation for ) is:

[0088] ,

[0089] in = This represents the coordinate correction number of the j-th target point. , , The core optimization target; = , indicating the first i The attitude angle correction of the image. , and These are the correction values ​​for the three variables of the attitude angle; It is a constant term. , Let j be the image point observation value of the j-th target point on the i-th image. , The initial values ​​are substituted into the theoretical values ​​of the image points in the collinear equation. , For image point ( , The residuals of the target points include both critical target points and auxiliary points. We can assume the first point is the critical target point and the subsequent points are auxiliary points. The coordinate corrections and attitude angle corrections are the quantities corresponding to the X matrix in the preceding error equation, X =

[0090] The partial derivative matrix (2×3) of the image coordinates of the j-th ground point on the i-th image with respect to the ground target point coordinates is calculated as follows:

[0091] ,

[0092] Let be the partial derivative matrix of the image coordinates of the j-th ground point on the i-th image with respect to the attitude angle, calculated as follows:

[0093] ,

[0094] ,

[0095] ,

[0096] ,

[0097] Indicates the camera's focal length. , , Represents the x, y, and z coordinates of the j-th target point;

[0098] , , ,..., , , The direction cosine corresponding to the attitude angle is calculated using the following formula:

[0099] ,

[0100] ,

[0101] ,

[0102] ,

[0103] ,

[0104] ,

[0105] ,

[0106] ,

[0107] ,

[0108] , and are the first, second, and third variables of the attitude angle parameters of image i, respectively.

[0109] Step 3, calculate the target point The positioning error coefficient.

[0110] cofactor matrix ,Pick , , , , , for Chinese correspondence The diagonal elements.

[0111] : Cofactor sub-block of the coordinates of key target points and surrounding auxiliary points;

[0112] : The sub-block of the covariance factors between the coordinate parameters of the key target point and the surrounding auxiliary points and the image attitude angle parameters; specifically, it reflects the degree of correlation between the accuracy change of the "coordinate parameters" and the accuracy change of the image attitude angle parameters, with the "coordinate parameters" as the reference.

[0113] : The co-correlation factor sub-block between the image attitude angle parameter and the coordinate parameters of the key target point / surrounding auxiliary points; specifically, it reflects the degree of correlation between the accuracy change of the "image attitude angle parameter" and the accuracy change of the coordinate parameter, with the "image attitude angle parameter" as the reference. and They are transposes of each other, and describe the accuracy correlation between coordinate parameters and attitude angle parameters from different perspectives.

[0114] The cofactor sub-block of the image attitude angle parameters is The submatrix contains only the image pose angle ( , , Cofactor information of the parameters;

[0115] : The j-th point The cofactor matrix is Corresponding point The submatrix of coordinate parameters reflects the accuracy characteristics of the three-dimensional coordinates of that point.

[0116] : The j-th point The three-dimensional positioning error coefficient comprehensively reflects the overall positioning accuracy of the point in the X, Y, and Z directions (the smaller the value, the higher the accuracy).

[0117] : The j-th point The planar positioning error coefficient reflects the position of the point on the X-axis. Positioning accuracy in the Y-plane;

[0118] : The j-th point The elevation positioning error coefficient reflects the positioning accuracy of the point in the Z direction (vertical direction).

[0119] : The cofactor of the X-axis coordinate reflects the accuracy of the X-axis coordinate.

[0120] : The cofactor of the Y-axis coordinate reflects the accuracy of the Y-axis coordinate;

[0121] : The cofactor of the Z-axis coordinate reflects the accuracy of the Z-axis coordinate.

[0122] Step 4: Using the positioning error coefficient as the objective function, determine its minimum value through a collaborative search algorithm to obtain the optimal configuration parameters.

[0123] Optimize the objective function with key objective points The goal is to achieve the highest positioning accuracy. The optimal geometric configuration for the UAV is then searched, and the specific steps are as follows:

[0124] Step 41, Geometric configuration coding.

[0125] Each geometric configuration corresponds to one set of UAV orbital configuration parameters, with a dimension of 3. N +3 N =6 N dimension( N indivual , N indivual , N indivual 3 N (attitude angles)

[0126] Step 42: Define the objective function for optimization.

[0127] The objective function for optimization is defined as follows:

[0128] ,

[0129] Key Target Points The three-dimensional positioning error coefficient (when constructing the B matrix in the error equation, the key target point can be set as the first point) comprehensively reflects the overall positioning accuracy of the point in the X, Y, and Z directions (the smaller the value, the higher the accuracy).

[0130] Key Target Points The planar positioning error coefficient reflects the position of the point on the X-axis. Positioning accuracy in the Y-plane;

[0131] Key Target Points The elevation positioning error coefficient reflects the positioning accuracy of the point in the Z direction (vertical direction).

[0132] In different application scenarios, when pursuing the highest 3D positioning accuracy of the target point, let F= When pursuing the highest planar accuracy of the target point ; Pursuing the highest accuracy of target point elevation ;

[0133] Step 43, iterative optimization.

[0134] a) Initialization: Generate 80 geometric configurations, and initialize the configuration parameter sets and parameter adjustment strategies for each geometric configuration; the configuration parameter sets represent the parameters in the geometric configuration, i.e., the encoding of the geometric configuration in step 41: N sets. N N 3N attitude angles ( , , The parameter adjustment strategy refers to the method of adjusting the parameters in the configuration parameter set during iteration in order to achieve the goal of minimizing the optimal objective function.

[0135] b) Calculation of positioning error coefficient: For each geometric configuration, analyze the configuration parameters → calculate the direction cosines → construct matrix B → solve... →Calculation (like Strange, endowed = );

[0136] c) Optimal value update: Record the historical optimal value for each geometry configuration. Globally optimal with all configurations Each geometric configuration has a set of configuration parameters, from which positioning error coefficients can be calculated. Then, the iteration proceeds with the objective of minimizing these positioning error coefficients (the optimal objective function in step 42 is to select one of the three positioning error coefficients). To achieve this goal, some key quantities need to be recorded during the iteration process. It is the set of configuration parameters that corresponds to the minimum positioning error coefficient during the iteration process of this geometric configuration. It is the set of configuration parameters that has the smallest positioning error coefficient among all geometric configurations during the iteration process.

[0137] d) Parameter adjustment strategy and configuration parameter group update:

[0138] ,

[0139] ,

[0140] in, and These are the parameter adjustment strategy vector and configuration parameter set vector for the geometric configuration e in the t-th iteration, respectively. Adjust the parameter strategy vector for the next geometric configuration iteration; This is the configuration parameter set vector for the next geometric configuration iteration; The persistence factor determines the degree to which the algorithm parameter adjustment strategy update is affected by the current geometric configuration parameter adjustment strategy; the individual influence factor c1 and the group influence factor c2 determine the degree to which the geometric configuration's own state and the historical best state affect the next parameter adjustment strategy update; r1 and r2 are random numbers in the interval [0,1], used to increase the randomness of the search; Let be the optimal configuration parameter set for geometric configuration e in the first t iterations; The optimal configuration parameter set for the population in the first t iterations is the optimal configuration parameter set.

[0141] e) Termination judgment: Iteration to T =500 times or Changes <10 over 30 consecutive iterations 6 T is the iteration number, at which point iteration stops.

[0142] Step 5: Calculate the UAV pose parameters corresponding to the optimal configuration parameters based on the optimal configuration parameters.

[0143] The optimal configuration parameters include: horizontal distance ,position ,high With attitude angle parameters , ,

[0144] The formula for calculating the three-dimensional coordinates of a UAV is:

[0145] ,

[0146] ,

[0147] .

[0148] Accordingly, embodiments of the present invention also provide a pose parameter optimization system for UAV aerial photography, such as... Figure 2 The image shown is a schematic diagram of one possible structure of the system. This UAV aerial photography pose parameter optimization system includes the following modules:

[0149] Initialization unit 201 determines the key target points and surrounding auxiliary points during UAV aerial photography and collects the UAV's geometric configuration parameters to be optimized;

[0150] Error equation calculation unit 202 establishes a coefficient matrix based on the UAV's geometric configuration parameters to be optimized, and calculates the error equations for the pixel coordinates of key target points and surrounding auxiliary points during UAV aerial photography.

[0151] The positioning error coefficient calculation unit 203 constructs a cofactor matrix based on the coefficient matrix and calculates the positioning error coefficient of key target points during UAV aerial photography.

[0152] The iterative optimization unit 204 uses the positioning error coefficient of the key target point as the optimization objective function, and determines the minimum value of the optimization objective function through a cooperative search algorithm to obtain the optimal configuration parameters;

[0153] The pose parameter calculation unit 205 calculates the UAV pose parameters corresponding to the optimal configuration parameters based on the optimal configuration parameters.

[0154] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that the present invention is not limited to the described order of actions, because according to the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to the present invention.

[0155] The present invention also provides a storage medium, which is a computer-readable storage medium storing a computer program thereon, the computer program being executable when it runs. Figure 1 The method shown may include some or all of the steps. The storage medium may include read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, etc. The storage medium may also include non-volatile memory or non-transitory memory, etc.

[0156] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data provider to another website, computer, server, or data provider via wired or wireless means.

[0157] The embodiments of the present invention have been described in detail above. Specific implementation methods have been used to illustrate the present invention. The descriptions of the embodiments above are only for the purpose of helping to understand the methods and systems of the present invention, and are merely some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention, and the content of this specification should not be construed as a limitation of the present invention. Therefore, any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for optimizing pose parameters during UAV aerial photography, characterized in that, The method includes: Step 1: Determine the key target points and surrounding auxiliary points during UAV aerial photography, and collect the UAV's geometric configuration parameters to be optimized; Step 2: Based on the geometric configuration parameters of the UAV to be optimized, establish a coefficient matrix and calculate the error equation of the pixel coordinates of the key target points and surrounding auxiliary points during UAV aerial photography. Step 3: Based on the coefficient matrix, construct the cofactor matrix and calculate the positioning error coefficients of key target points during UAV aerial photography; Step 4: Using the positioning error coefficients of the key target points as the optimization objective function, determine the minimum value of the optimization objective function through a collaborative search algorithm to obtain the optimal configuration parameters; Step 5: Calculate the UAV pose parameters corresponding to the optimal configuration parameters based on the optimal configuration parameters.

2. The method for optimizing pose parameters during UAV aerial photography according to claim 1, characterized in that, In step 1, given the key target points and M surrounding auxiliary points k=1,2,…M, key target points These are the key locations that need to be photographed during drone aerial photography, and their coordinates are shown below. , M surrounding auxiliary points Evenly distributed at key target points All around.

3. The method for optimizing pose parameters during UAV aerial photography according to claim 2, characterized in that, In step 1, the drone is set to orbit the key target point. The number of images captured is N, and the drone is capturing images of key target points. The geometric configuration parameters to be optimized for the i-th image are: horizontal distance ,position ,high With attitude angle , , , i=1,2,…N; where, horizontal distance Key target points Planar distance from the location of the drone when it captured the i-th image; orientation Key target points The azimuth and altitude of the image taken by the drone when capturing the i-th image. Key target points The height difference between the image taken by the drone and the image taken at the i-th time.

4. The method for optimizing pose parameters during UAV aerial photography according to claim 3, characterized in that, In step 2, based on the drone's shooting of key target points The attitude angle of the drone in the i-th image, and the drone's position when shooting key target points. Key target points in the i-th image and M surrounding auxiliary points Calculate the parameter values ​​in the coefficient matrix based on the coordinate values ​​in the x, y, and z directions, and then use the coefficient matrix and the data from the UAV when capturing key target points. Key target points in the i-th image and M surrounding auxiliary points Error equations are used to calculate the pixel coordinates of key target points and surrounding auxiliary points during UAV aerial photography, based on coordinate corrections and attitude angle corrections in the x, y, and z directions.

5. The method for optimizing pose parameters during UAV aerial photography according to claim 4, characterized in that, In step 3, the cofactor matrix constructed based on the coefficient matrix B is as follows: ， in, : Cofactor sub-block of the coordinates of key target points and surrounding auxiliary points; : Sub-block of the covariance factors between the coordinate parameters of key target points and surrounding auxiliary points and the image attitude angle parameters; : The cross-correlation factor sub-block between image attitude angle parameters and the coordinate parameters of key target points and surrounding auxiliary points; The cofactor sub-block of the attitude angle parameters of the key target point and surrounding auxiliary points is the cofactor matrix. The submatrix contains only attitude angles ( , , Cofactor information of the parameters; use Represents the j-th point The cofactor matrix, take , , ,in, , , for Chinese correspondence The diagonal elements; j = 0, 1, 2, ..., M; Kxyz j : the three-dimensional positioning error coefficient of the jth point P j , which comprehensively reflects the overall positioning accuracy of the point P j in the X, Y, and Z directions; Kxy j : The j-th point P j The planar positioning error coefficient reflects the position of point P. j In X Positioning accuracy in the Y-plane; Kz j : The j-th point P j The elevation positioning error coefficient reflects the position of point P. j Positioning accuracy in the Z direction, i.e., the vertical direction; :point The cofactor of the X-axis coordinate reflects the accuracy of the X-axis coordinate. :point The cofactor of the Y-axis coordinate reflects the accuracy of the Y-axis coordinate; :point The cofactor of the Z-axis coordinate reflects the accuracy of the Z-axis coordinate.

6. The method for optimizing pose parameters during UAV aerial photography according to claim 5, characterized in that, Step 4 includes: Step 41: Encode the UAV's geometric configuration. Each geometric configuration corresponds to one set of UAV orbital configuration parameters, with a dimension of 3. N +3 N =6 N Wei, that is N indivual , N indivual , N indivual 3 N One attitude angle; Step 42, set key target points As the first point in the coefficient matrix, define the optimal objective function: ， Key Target Points The three-dimensional positioning error coefficient comprehensively reflects the key target points. The overall positioning accuracy of the point in the X, Y, and Z directions; Key Target Points The planar positioning error coefficient reflects the critical target point In X Positioning accuracy in the Y-plane; Key Target Points The elevation positioning error coefficient reflects the key target points Positioning accuracy in the Z direction, i.e., the vertical direction; Step 43: Perform iterative optimization with the goal of finding the geometric configuration that minimizes the optimal objective function; Step 44: When the iteration reaches 500 times or the population has the optimal configuration parameter set in the previous t iterations... Changes <10 over 30 consecutive iterations 6 Stop the iteration to obtain the optimal configuration parameters.

7. The method for optimizing pose parameters during UAV aerial photography according to claim 6, characterized in that, Step 43 includes: Step a: Initialize the configuration, generate 80 geometric configurations, and initialize the configuration parameter group and parameter adjustment strategy of the geometric configurations; Step b: For each geometric configuration, calculate the positioning error coefficient using the methods in steps 1-3; Step c: Update the optimal value and record the historical optimal value for each geometric configuration. Globally optimal with all configurations ; Step d: Perform parameter adjustment strategy and configuration parameter group update: ， ， in, and These are the parameter adjustment strategy vector and configuration parameter set vector for the geometric configuration e in the t-th iteration, respectively. Adjust the parameter strategy vector for the next geometric configuration iteration; This is the configuration parameter set vector for the next geometric configuration iteration; The persistence factor determines the degree to which the algorithm parameter adjustment strategy update is affected by the current geometric configuration parameter adjustment strategy; the individual influence factor c1 and the group influence factor c2 determine the degree to which the geometric configuration's own state and the historical best state affect the next parameter adjustment strategy update; r1 and r2 are random numbers in the interval [0,1], used to increase the randomness of the search; Let be the optimal configuration parameter set for geometric configuration e in the first t iterations; The optimal configuration parameter set for the population in the first t iterations is the optimal configuration parameter set.

8. The method for optimizing pose parameters during UAV aerial photography according to claim 7, characterized in that, In step 5, the optimal configuration parameters include: horizontal distance. ,position ,high With attitude angle parameters , , The formula for calculating the three-dimensional coordinates in the UAV pose parameters based on the optimal configuration parameters is as follows: ， ， 。 9. A pose parameter optimization system for UAV aerial photography, characterized in that, The system includes: The initialization unit determines the key target points and surrounding auxiliary points during UAV aerial photography and collects the UAV's geometric configuration parameters to be optimized. The error equation calculation unit establishes a coefficient matrix based on the UAV's geometric configuration parameters to be optimized, and calculates the error equations for the pixel coordinates of key target points and surrounding auxiliary points during UAV aerial photography. The positioning error coefficient calculation unit constructs a cofactor matrix based on the coefficient matrix and calculates the positioning error coefficient of key target points during UAV aerial photography. The iterative optimization unit uses the positioning error coefficient of the key target point as the optimization objective function, and determines the minimum value of the optimization objective function through a cooperative search algorithm to obtain the optimal configuration parameters; The pose parameter calculation unit calculates the UAV pose parameters corresponding to the optimal configuration parameters based on the optimal configuration parameters.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is run by the processor, it executes the steps of the pose parameter optimization method for UAV aerial photography according to any one of claims 1 to 8.

Images