A UAV Aerial Photography Path Planning Method for Dynamic Modeling

Abstract

The present invention relates to the field of architectural surveying and mapping, and provides a method for drone aerial photography path planning for dynamic modeling, including steps of target data acquisition, modeling structure analysis, mapping point benefit calculation, aerial photography path planning, and path interference compensation. The present invention introduces interference compensation and flight route adjustment strategies, and through the investigation and avoidance of obstacles interfering with the drone, achieves the beneficial effect of improving the stability of the drone modeling aerial photography system. Through precise structure analysis, scientific benefit calculation, and optimized path planning, the scientificity, efficiency, and adaptability of the drone aerial photography path planning are comprehensively improved, providing high-quality data support for dynamic modeling.

Description

Technical Field

[0001] The present invention relates to the field of architectural surveying and mapping, and specifically to an unmanned aerial vehicle (UAV) aerial photography path planning method for dynamic modeling. Background Art

[0002] Traditional UAV aerial photography path planning methods often rely on static and preset flight paths, lacking sufficient consideration of the dynamic characteristics of the modeling target and environmental changes. For example, some methods may simply set one or more fixed flight paths for image acquisition based on the approximate position and scope of the modeling target. This approach is difficult to adapt to the complex shape and structural characteristics of the modeling target, and may result in the omission or repeated acquisition of key information, thereby affecting the accuracy and efficiency of modeling.

[0003] In addition, existing UAV aerial photography path planning technologies lack flexibility and pertinence when facing modeling targets with different structures. For modeling targets with conventional three-dimensional structures, such as regular-shaped building complexes, traditional planning methods may not be able to effectively optimize the aerial photography path to make full use of the symmetry and structural repeatability of these targets to reduce unnecessary flights and surveying work. For special-shaped three-dimensional structures, such as art buildings with complex curves and polyline turns, existing methods may be difficult to accurately capture these shape changes, resulting in poor modeling effects. Moreover, existing technologies also lack comprehensive consideration of environmental factors. During aerial photography, the UAV may encounter various obstacles, such as trees, power poles, etc., which may pose threats to the flight path and safety of the UAV. Traditional path planning methods often lack real-time detection and avoidance strategies for obstacles on the aerial photography route, which may lead to collisions during the flight of the UAV or frequent adjustment of the flight path, thereby reducing the aerial photography efficiency and modeling quality. In addition, existing UAV aerial photography path planning methods also have deficiencies in comprehensive benefit evaluation. During aerial photography, different surveying points may have different sampling information and surveying costs, such as pixel clarity, texture richness, feature salience, as well as the UAV flight distance, flight tortuosity, and environmental information. Traditional planning methods often do not fully consider the impact of these factors on the comprehensive benefit, so it is difficult to optimize the aerial photography path and reduce the surveying cost while ensuring the modeling quality.

[0004] Therefore, to solve the problems existing in the above-mentioned prior art, the present invention proposes an unmanned aerial vehicle (UAV) aerial photography path planning method for dynamic modeling to improve the aerial photography efficiency, modeling quality, and system stability. Summary of the Invention

[0005] Aiming at the deficiencies of the prior art, the purpose of the present invention is to provide an unmanned aerial vehicle (UAV) aerial photography path planning method for dynamic modeling.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A method for UAV aerial photography path planning for dynamic modeling, comprising the following steps:

[0008] A target data acquisition step of acquiring an image of a modeling target collected by a camera carried by a UAV as a model analysis image, and analyzing the size, shape, and structural features of the modeling target from the model analysis image as basic model information;

[0009] A modeling structure analysis step of determining the structural type of the modeling target through a structure analysis model according to the basic model information, and analyzing the basic surveying and mapping points of the modeling target according to the structural type in combination with the basic model information;

[0010] A surveying and mapping point benefit calculation step of analyzing the sampling information and surveying and mapping cost of the basic surveying and mapping points from the model analysis image, and calculating the comprehensive benefit value of the basic surveying and mapping points through a comprehensive benefit model according to the sampling information and surveying and mapping cost of the basic surveying and mapping points;

[0011] An aerial photography path planning step of dividing the basic surveying and mapping points into several key surveying and mapping areas by a surveying and mapping point planning strategy according to the structural features, prioritizing the basic surveying and mapping points in each key surveying and mapping area according to the comprehensive benefit value, and selecting the basic surveying and mapping points according to the priority ranking result to complete the UAV aerial photography path planning.

[0012] As a further improvement of the present invention, the structure analysis model analyzes the structural type of the modeling target according to the shape and structural features of the modeling target, and the structural type includes a conventional three-dimensional structure and a special-shaped three-dimensional structure; when the structure analysis model determines that the similarity between the shape of the modeling target and a preset geometric parameter is greater than a preset threshold, and the number of symmetric structures and structural repeating units in the structural features of the modeling target is greater than a preset threshold, the output structural type is a conventional three-dimensional structure, otherwise the output structural type is a special-shaped three-dimensional structure; when the structural type is a conventional three-dimensional structure, the basic surveying and mapping points are selected according to the shape and size of the modeling target on the geometric center plane, connection points, and key feature points of each geometric shape of the modeling target. When the structural type is a special-shaped three-dimensional structure, the basic surveying and mapping points are set at special nodes according to the shape information of the modeling target, including densely selecting basic surveying and mapping points at positions where the curve curvature changes violently, and setting basic surveying and mapping points at the turning points of the broken line according to the analysis of the model analysis image.

[0013] As a further improvement of the present invention, the mapping revenue calculation strategy includes extracting the pixel clarity, texture richness, and feature saliency of the area where the basic mapping points are located in the model analysis image according to the structure type as sampling information, and comprehensively obtaining the mapping cost by analyzing the flight distance of the unmanned aerial vehicle, the flight tortuosity of the unmanned aerial vehicle at key points, and the environmental information according to the size and shape of the modeling target. The sampling information of each mapping point and the mapping cost are respectively calculated through a comprehensive revenue model to obtain the comprehensive revenue value of each basic mapping point.

[0014] As a further improvement of the present invention, the comprehensive revenue model is configured with:

[0015] ;

[0016] wherein, R i is the comprehensive revenue value of the i-th basic mapping point, C i is the pixel clarity of the i-th basic mapping point, T i is the texture richness of the i-th basic mapping point, S i is the feature saliency of the i-th basic mapping point, indicating the degree of prominence of the features of this mapping point; D i is the flight distance of the unmanned aerial vehicle to the position of the i-th basic mapping point, W i is the flight tortuosity of the unmanned aerial vehicle at the position of the i-th basic mapping point, indicating the flight difficulty of the unmanned aerial vehicle at this point; E i is the environmental impact factor at the position of the i-th basic mapping point.

[0017] As a further improvement of the present invention, the aerial photography path planning step further includes dividing key mapping areas according to the spatial coordinates of the basic mapping points through distribution density and the structural characteristics of the modeling target, constructing a comprehensive revenue matrix for the basic mapping points within each key mapping area based on the comprehensive revenue value, confirming the weights of each basic mapping point by analyzing the comprehensive revenue matrix, and sorting the basic mapping points from high to low according to the weights.

[0018] As a further improvement of the present invention, the path interference compensation step further includes extracting a theoretical target image corresponding to the position extracted from the basic mapping points of the sampling image from the theoretical target model, respectively extracting the image features of the sampling image and the theoretical target image through edge detection and pixel coordinate calibration. The image features include the contour perimeter, area, shape descriptor, texture feature, and key point position of the modeling target in the sampling image. The spatial distribution value is obtained by statistically matching the spatial distribution of the deviation of the key point positions in the two images. The deviation coefficient is calculated through an interference deviation model based on the difference in the image features between the sampling image and the theoretical target image, and a primary deviation threshold, a secondary deviation threshold, and a spatial distribution threshold are set. When the deviation coefficient is higher than the primary deviation threshold and lower than the secondary deviation threshold, and the spatial distribution value is lower than the spatial distribution threshold, it is output that there is a microscopic deviation at this basic mapping point. When the deviation coefficient is higher than the secondary deviation threshold and the spatial distribution value is higher than the spatial distribution threshold, it is output that there is a global deviation at this basic mapping point.

[0019] As a further improvement of the present invention, the interference deviation model is configured as:

[0020] ;

[0021] Among them, D is the deviation coefficient, which is used to judge the deviation degree between the sampling image and the theoretical target image; n represents the number of pairs of pictures at the modeling target in the image, usually 1. When continuous sampling is required for this basic mapping point, n is the number of pairs of corresponding sampling images and theoretical target images; and <000; and and and H H represents the Hausdorff distance between the shape descriptors, indicating the distance metric between the shape descriptors; and and respectively represent the position vectors of the k-th key point of the i-th sampled image and the theoretical target image, p is the number of key points, and respectively represent the norms of the sets of all key point position vectors of the i-th sampled image and the theoretical target image, representing the sum of the distances from all key points to the centroid. α, β, γ, δ, and ε respectively represent the weight factors of each parameter, where the value of ε is greater than α, β, γ, δ, and α + β + γ + δ + ε = 1.

[0022] As a further improvement of the present invention, the micro path adjustment strategy includes calculating the offset of the basic surveying point according to the contour perimeter, area, shape descriptor, texture feature, and key point position deviation value of the modeling target in the sampled image and the theoretical target image, adjusting the position coordinates of the basic surveying point according to the offset of the basic surveying point, and re-acquiring the sampled image of the modeling target at the updated position of the basic surveying point. The calculation of the offset of the basic surveying point is configured as:

[0023] ;

[0024] Among them, is the comprehensive correction vector, representing the offset of the basic surveying point coordinates, (v PAx , v PAy ) is the first correction vector based on the deviation of the contour perimeter and area, where , , k P is the contour perimeter deviation correction coefficient, k A is the area deviation correction coefficient, is the contour perimeter deviation, is the area deviation, P0 is the reference contour perimeter of the modeling target based on the theoretical target image, A0 is the reference area of the modeling target based on the theoretical target image, θ represents the correction direction angle calculated according to the deviation of the contour perimeter and area, (v Jx , v Jy ) is the second correction vector based on the deviation of the shape descriptor, , , where, ΔH i is the i-th element in the difference vector matrix, ω i is the weight coefficient of the i-th element in the x direction, ω ＇ i is the weight coefficient of the i-th element in the y direction, (v Xx , v Xy ) is the third correction vector based on the deviation of the texture feature, , , where, k X is the texture deviation correction coefficient, is the correction direction angle determined according to the texture deviation distribution, and (Δx, Δy) is the key point position deviation;

[0025] As a further improvement of the present invention, the global path adjustment strategy includes calculating a benefit adjustment factor for the basic surveying and mapping points at the corresponding positions of the sampled images according to the deviations of the contour perimeter, area, shape descriptor, texture feature, and key point position. The benefit adjustment factor includes a texture richness adjustment factor, a feature saliency adjustment factor, and a flight distance adjustment factor, and recalculating the comprehensive benefit adjustment value of the basic surveying and mapping point through a comprehensive benefit recalculation model according to the comprehensive benefit adjustment factor. Replacing the comprehensive benefit value of the basic surveying and mapping point with the recalculated comprehensive benefit adjustment value as the final comprehensive benefit value, re-prioritizing the basic surveying and mapping points in each key surveying and mapping area according to the comprehensive benefit values of the basic surveying and mapping points, and re-selecting the basic surveying and mapping points according to the priority ranking result for the second aerial photography path planning.

[0026] As a further improvement of the present invention, the comprehensive benefit recalculation model is configured with:

[0027] ;

[0028] wherein, is the comprehensive benefit adjustment value, g1 is the texture richness adjustment factor, used to represent the influence of the deviation value on the texture richness in the comprehensive benefit value, , 、 are the adjustment coefficients of the contour perimeter and area respectively, g2 is the feature saliency adjustment factor , 、 are adjustment factors, used to represent the influence of the deviation value on the feature saliency in the comprehensive benefit value, g3 is the flight distance adjustment factor ,used to represent the influence of the deviation value on the flight distance in the comprehensive benefit value, g4 is the flight tortuosity adjustment factor, used to represent the influence of the deviation value on the flight tortuosity in the comprehensive benefit value.

[0029] The beneficial effects of the present invention are as follows: The present invention divides key surveying and mapping areas according to the spatial coordinates, distribution density of basic surveying and mapping points, and the structural characteristics of the modeling target. Within each area, a comprehensive benefit matrix is constructed based on the comprehensive benefit value, and the weights of each basic surveying and mapping point are determined and sorted through methods such as principal component analysis. This method makes the aerial photography path planning more scientific and reasonable, preferentially selects surveying and mapping points with high value, improves the aerial photography efficiency, and reduces unnecessary flight paths. Moreover, the present invention introduces a disturbance compensation and flight route adjustment strategy, and achieves the beneficial effect of improving the stability of the UAV modeling aerial photography system by detecting and avoiding obstacles that interfere with the UAV. Through precise structural analysis, scientific benefit calculation, and optimized path planning, the scientificity, efficiency, and adaptability of the UAV aerial photography path planning are comprehensively improved, providing high-quality data support for dynamic modeling. Description of the Drawings

[0030] Figure 1 is a flowchart of a method for UAV aerial photography path planning for dynamic modeling according to the present invention.

[0031] Figure 2 is a flowchart of the surveying and mapping benefit calculation strategy of the present invention.

[0032] Figure 3 is a flowchart of the path interference compensation step of the present invention. Detailed Embodiment

[0033] The present invention will be further described in detail below with reference to the drawings and embodiments. The same components are denoted by the same reference numerals. It should be noted that the terms "front", "rear", "left", "right", "upper" and "lower" used in the following description refer to the directions in the drawings, and the terms "bottom surface" and "top surface", "inner" and "outer" respectively refer to the directions towards or away from the geometric center of a specific component.

[0034] What are the disadvantages of the current prior art? Therefore, the present invention designs this method for UAV aerial photography path planning for dynamic modeling, as Figure 1 shown, including:

[0035] Target data acquisition step: Obtain the images of the modeling target collected by the camera carried by the drone as model analysis images, and analyze the size, shape, and structural features of the modeling target from the model analysis images as basic model information; Collect images of the modeling target from multiple directions through the high-precision camera carried by the drone, and the collected images cover all angles of the modeling target to ensure that the acquisition of model information can more comprehensively display the size, shape, and structural features of the modeling target. Preprocess the collected images, and use a filtering algorithm such as Gaussian filtering to remove sensor noise and environmental noise; Gaussian filtering performs a convolution operation on each pixel in the image by setting appropriate kernel size and standard deviation, smooths out the noise points, and at the same time retains the edge and detail information of the image. The target data acquisition step further includes, after obtaining the model analysis images, analyzing the actual size of the modeling target from the model analysis images through a vision algorithm to obtain the size of the modeling target, and using a contour extraction algorithm to obtain the contour information of the modeling target from the model analysis images, and quantitatively describing the contour information through a shape descriptor to obtain the shape of the modeling target; Identify and locate the predefined structural features from the model analysis images by constructing a deep learning object detection algorithm. After image preprocessing, extract the size data of the modeling target from the preprocessed model analysis images through computer vision algorithms, and estimate the actual size of the modeling target through the reference objects with known ratios in the images; Use a contour extraction algorithm to obtain the contour information of the modeling target, and then use a shape descriptor to quantitatively describe the shape. The shape descriptor is preset in the database and is a basic shape unit for judging the contour information of the modeling target.

[0036] Modeling structure analysis step: Determine the structural type of the modeling target through a structural analysis model according to the basic model information, and analyze the basic surveying and mapping points of the modeling target in combination with the basic model information according to the structural type; Construct a structural analysis model based on deep learning, and use a convolutional neural network to train a large amount of image data of different types of structures, so that the model can more accurately identify the structural type of the modeling target, and analyze the basic surveying and mapping points according to the determined structural type and basic information using different judgment criteria.

[0037] The structural analysis model analyzes the structural type of the modeling target according to the shape and structural features of the modeling target. The structural types include regular three-dimensional structures and special-shaped three-dimensional structures; When the structural analysis model determines that the similarity between the shape of the modeling target and the preset geometric parameters is greater than the preset threshold, and the number of symmetric structures and structural repeating units in the structural features of the modeling target is greater than the preset threshold, the output structural type is a regular three-dimensional structure, otherwise the output structural type is a special-shaped three-dimensional structure. Among them, let the set of actual shape parameters of the modeling target be , for example, for a cuboid shape, Let it be the length, Let it be the width, Let it be the height. The preset set of geometric shape parameters is .

[0038] At this time, the shape similarity S is configured as: , the closer S is to 1, the more it indicates that the shape of the modeling target

[0039] is similar to the preset geometric shape. By analyzing the structural characteristics of the modeling target, use the spatial geometric transformation algorithm to detect whether there is a symmetric part in the target structure, and record the number of symmetric structures. Identify the structural repeating units in the modeling target through the pattern recognition algorithm: extract the feature vectors of its structure for the modeling target, and then cluster the similar structural feature vectors into one category through the clustering algorithm. At this time, each category is a structural repeating unit, and the classification standard is adjusted by setting the similarity threshold.

[0040] After completing the classification of structural features, select the basic surveying and mapping points of the modeling target according to the type of structural features.

[0041] When the structural type is a conventional three-dimensional structure, select the basic surveying and mapping points according to the shape and size of the modeling target on the geometric center plane, connection points and key feature points of each geometric shape of the modeling target. For a conventional three-dimensional structure, according to the shape characteristics of the modeling target, a conventional three-dimensional structure is mostly composed of common geometric shapes. Taking the cuboid-shaped building in the model analysis image as an example, determine the length, width and height according to its size information, and set the basic surveying and mapping points at the centers of each face, the midpoints of the edges and the vertices. This is because these parts can effectively reflect the overall shape and size of the cuboid, and in subsequent modeling, the framework of the model can be accurately constructed using these points. This strategy takes into account the shape and size factors of the modeling target.

[0042] Utilize the symmetry of the modeling target to select the corresponding points on the symmetric plane as the basic surveying and mapping points. For example, for a cylindrical building, the structural characteristics of the corresponding points on both sides of its axis of symmetry are the same. By selecting several points on one side, the corresponding points on the other side can be determined according to the symmetry relationship. This not only reduces the surveying and mapping workload but also ensures the symmetry of the model. For a conventional three-dimensional structure composed of repeating units, analyze the distribution law of the repeating units in the model analysis image. Select representative points from each repeating unit. For example, for each workshop unit of an industrial plant, select points with obvious features such as the entrance and the center. These points can represent the structural characteristics of the repeating unit, so selecting these points as the basic surveying and mapping points can improve the modeling efficiency.

[0043] When the structural type is an irregular three-dimensional structure, basic surveying and mapping points are set at special nodes according to the shape information of the modeling target, including densely selecting basic surveying and mapping points at positions where the curve curvature changes drastically, and setting basic surveying and mapping points at the turning points of the broken lines according to the analysis of the model analysis image. For an irregular three-dimensional structure, since its shape is irregular, according to the complex shape of the target in the model analysis image, basic surveying and mapping points are densely set at positions where the curve curvature changes greatly, at the turning points of the broken lines, and at the surface mutations. For example, for an art building with a strange shape, at the curved and turning parts of its surface, the number of surveying and mapping points is increased to accurately capture the shape changes. This strategy is based on the shape and structural characteristics of the modeling target to ensure that sufficient information can be obtained to accurately depict the complex shape. According to the structural characteristics of the modeling target, for unique structures in the irregular three-dimensional structure, such as special modeling parts of large sculptures and cantilever structures in buildings, surveying and mapping points are set comprehensively and meticulously. By analyzing the model analysis image, the boundaries and internal structures of these parts are clarified, and the surveying and mapping points are evenly distributed to completely obtain their structural information, taking into account the unique structural characteristics of the irregular structure and its importance in modeling. Considering that there may be occluded and hidden parts in the irregular three-dimensional structure, using the size and shape information of the modeling target and combining the results of observing from different angles in the model analysis image, the positions and structures of the occluded and hidden parts are estimated. Surveying and mapping points are set at the junctions between the visible parts and the possible occluded and hidden parts, and at the key hidden parts judged according to experience, so as to restore the complete structure to the greatest extent during subsequent modeling.

[0044] Steps for calculating the benefit of surveying and mapping points: Obtain the sampling information and surveying and mapping cost of the basic surveying and mapping points according to the analysis of the model analysis image, and calculate the comprehensive benefit value of the basic surveying and mapping points through a comprehensive benefit model based on the sampling information and surveying and mapping cost of the basic surveying and mapping points.

[0045] The surveying and mapping benefit calculation strategy includes, as Figure 2 shown, extracting the pixel clarity, texture richness, and feature saliency of the region where the basic surveying and mapping points are located in the model analysis image as sampling information according to the structural type, and comprehensively obtaining the surveying and mapping cost based on the size and shape analysis of the modeling target, the flight curvature of the unmanned aerial vehicle at key points, and the environmental information. The comprehensive benefit value of each basic surveying and mapping point is calculated through the comprehensive benefit model for the sampling information and the surveying and mapping cost of each surveying and mapping point respectively.

[0046] When the structure type is a conventional solid structure, given their regular shape and relatively stable characteristics, the sampling information calculation focuses on analyzing the integrity and clarity of the images of the base mapping points at different shooting angles. Pixel information for the area where the base mapping point is located is extracted from the model analysis image. A higher sampling information score is assigned if the pixel distribution in this area is uniform, the edges are clear, and the contrast with the surrounding area is moderate. For example, for a base mapping point on a rectangular building, a higher sampling information score is assigned if the shooting angle fully captures the features of the surface where the point is located, and the image is free of blur and distortion. Furthermore, considering the size of the modeled object, the scores of base mapping points that are farther away but fully included in the model analysis image are adjusted appropriately based on their image proportion. If the base mapping point's image proportion is too small, the accuracy of its information will be affected, and the sampling information score will be reduced accordingly. Since the position and shape of conventional solid structures are relatively fixed, the mapping cost primarily considers the flight distance and difficulty required for the drone to reach the base mapping point. The flight distance is calculated based on the spatial coordinates of the base mapping point and the drone's initial position. If the base mapping point is located near a straight line in the drone's flight direction and is clear of obstacles, the flight difficulty is low and the mapping cost is reduced accordingly. Conversely, if reaching the base mapping point requires the drone to make significant turns, traverse complex environments, or avoid no-fly zones, the mapping cost will increase. For example, if the base mapping point is located on a conventional building in an open area, the flight cost is low; if it is located on a conventional building in an area surrounded by tall buildings and subject to numerous flight restrictions, the flight cost will increase.

[0047] When the structure type is an irregular three-dimensional structure, due to the complex shape and rich details of the irregular three-dimensional structure, when calculating the sampling information, in addition to considering the integrity and clarity of the image, it is also necessary to focus on the amount of information of the unique structural features contained in the basic surveying and mapping points. Extract the texture, curvature and other feature information around the basic surveying and mapping points from the model analysis image. If the point is located at the key feature parts of the irregular structure, such as the extreme points of complex curved surfaces, the central positions of unique decorations, etc., and these features are clearly distinguishable in the image, a higher sampling information score is given. At the same time, considering the information complementarity between different basic surveying and mapping points, if a certain basic surveying and mapping point can provide key information different from other points and helps to construct the model more comprehensively, its sampling information score is increased. For example, in an irregular building with a unique sculpture shape, the texture and shape changes on the surface of the sculpture are rich. The sampling information score of the basic surveying and mapping points located in the area with obvious texture changes will be increased because they contain unique texture information. The irregularity of the irregular three-dimensional structure increases the flight difficulty and risk of the drone, and more factors need to be comprehensively considered in the calculation of the surveying and mapping cost. In addition to the flight distance, it is also necessary to consider the energy and time consumed by the complex attitude adjustment of the drone when approaching the basic surveying and mapping points in order to obtain the best shooting angle. For example, when the basic surveying and mapping point is located in the concave part of the irregular building or under the protruding overhang, the drone needs to adjust its attitude to avoid collision and find a suitable shooting angle, which will significantly increase the surveying and mapping cost. In addition, due to the occlusion problem caused by the irregular structure, the drone needs to adjust its position multiple times to obtain complete information, and each adjustment increases the surveying and mapping cost. At the same time, combining the previously collected model analysis images and structural feature data, if it is found that the environmental interference in a certain area is large, such as strong wind, light reflection, etc., it will also increase the surveying and mapping cost of the basic surveying and mapping points in this area.

[0048] The comprehensive benefit model is configured with:

[0049] ;

[0050] Among them, is the comprehensive benefit adjustment value, g1 is the texture richness adjustment factor, which is used to represent the influence of the deviation value on the texture richness in the comprehensive benefit value, , 、 are the adjustment coefficients of the contour perimeter and area respectively, g2 is the feature salience adjustment factor , 、 are the adjustment factors, which are used to represent the influence of the deviation value on the feature salience in the comprehensive benefit value, g3 is the flight distance adjustment factor , which is used to represent the influence of the deviation value on the flight distance in the comprehensive benefit value, g4 is the flight tortuosity adjustment factor, which is used to represent the influence of the deviation value on the flight tortuosity in the comprehensive benefit value.

[0051] C i is the pixel clarity of the area where the i-th basic surveying and mapping point is located, and is the quantization value of the clarity of the image of this surveying and mapping point calculated by the image processing algorithm; T i is the texture richness of the area where the i-th basic surveying and mapping point is located, and is the quantization value of the complexity of the texture of the image of this surveying and mapping point obtained by the texture analysis algorithm; S i is the feature saliency of the area where the i-th basic surveying and mapping point is located, and is the quantization value of the prominence of the features of this surveying and mapping point obtained based on the feature extraction algorithm; D i is the flight distance of the drone to the i-th basic surveying and mapping point, and is the actual flight distance calculated according to the coordinates of the drone and the surveying and mapping point; W i is the flight tortuosity of the drone at the key position of the i-th basic surveying and mapping point, and is the quantization value of the tortuosity obtained by analyzing information such as the curvature of the drone's flight path; E i is the environmental information influence factor of the i-th basic surveying and mapping point, and is the quantization value obtained by comprehensively considering the influence of environmental factors such as wind speed and light on surveying and mapping; ω1, ω2, ω3 are the weight coefficients of pixel clarity, texture richness, and feature saliency respectively, which are preset constants according to the actual modeling situation and are used to adjust the importance of each sampling information factor in the comprehensive benefit. ω4, ω5, ω6 are the weight coefficients of flight distance, flight tortuosity, and environmental information respectively, which are used to adjust the importance of each surveying cost factor in the comprehensive benefit, is a preset constant used to ensure that the denominator is not zero.

[0052] is a function for processing pixel clarity, where is the maximum value of the pixel clarity of all surveying and mapping points, is the mean value of the pixel clarity, is the standard deviation of the pixel clarity;

[0053] is a function for processing texture richness, where is the maximum value of the texture richness of all surveying and mapping points, is the mean value of the texture richness, is the standard deviation of the texture richness;

[0054] is a function for processing feature saliency, where is the maximum value of the feature saliency of all surveying and mapping points, is the mean value of the feature saliency, is the standard deviation of the feature saliency;

[0055] is a function for processing flight distance, where is the maximum value of the flight distances of all survey points, is the average value of the flight distances, is the standard deviation of the flight distances;

[0056] is a function for processing the flight tortuosity, where is the maximum value of the flight tortuosities of all survey points, is the average value of the flight tortuosities, is the standard deviation of the flight tortuosities;

[0057] is a function for processing the flight distance, where is the maximum value of the flight distances of all survey points, is the average value of the flight distances, is the standard deviation of the flight distances.

[0058] Steps for aerial photography path planning: According to the described structural features, the basic survey points are divided into regions and composed into several key survey regions through the survey point planning strategy. The basic survey points in each key survey region are sorted by priority according to the comprehensive benefit value, and the basic survey points are selected according to the priority sorting result to complete the unmanned aerial vehicle (UAV) aerial photography path planning. The steps for aerial photography path planning further include dividing the key survey regions according to the spatial coordinates of the basic survey points through the distribution density and the structural features of the modeling target, constructing a comprehensive benefit matrix based on the comprehensive benefit value for the basic survey points within each key survey region, confirming the weights of each basic survey point by analyzing the comprehensive benefit matrix, and sorting the basic survey points from high to low according to the weights.

[0059] For the basic survey points within each key survey region, a comprehensive benefit matrix M is constructed. The rows of the matrix represent the basic survey points, and the columns represent the factors affecting the comprehensive benefit, such as pixel clarity C, texture richness T, feature salience S, flight distance D, flight tortuosity W, and environmental information E, etc. That is, M ij represents the quantization value of the i-th basic survey point on the j-th influencing factor.

[0060] The principal component analysis method is used to analyze the comprehensive benefit matrix M. PCA transforms the original multiple correlated variables into a set of new uncorrelated variables through linear transformation, that is, the principal components. First, calculate the covariance matrix of the comprehensive benefit matrix M, and then solve the eigenvalues λ1 to λm (m is the number of influencing factors) and the corresponding eigenvectors to . The magnitude of the eigenvalue reflects the contribution degree of the principal component to the data variance. The eigenvalues are normalized to obtain the weights τ1 to τ of each principal component m, namely , and these weights represent the relative importance of different influencing factors in the comprehensive income.

[0061] For each basic surveying and mapping point, calculate the comprehensive score according to its value in the comprehensive income matrix and the weights of each factor, and sort the basic surveying and mapping points in each key surveying and mapping area from high to low according to the comprehensive score. The sorting result will be used as an important basis for the UAV aerial photography path planning, and the basic surveying and mapping points with high comprehensive scores will be preferentially selected for shooting to improve the aerial photography efficiency and modeling quality.

[0062] Specifically, the path interference compensation step further includes, for example Figure 3As shown, the theoretical target image corresponding to the position extracted according to the basic surveying and mapping points of the sampling image is obtained from the theoretical target model. The image features of the sampling image and the theoretical target image are extracted through edge detection and pixel coordinate calibration respectively. The image features include the contour perimeter, area, shape descriptor, texture feature, and key point position of the modeling target in the sampling image. The edge contours of the modeling target in the theoretical target image and the sampling image are extracted through the Canny algorithm to obtain the contour perimeter. The difference in the contour perimeter between the sampling image and the theoretical target image can be used to judge the difference between the shape of the modeling target in the image obtained during the actual shooting process and the theoretical modeling. Moreover, through pixel coordinate calibration, precise coordinate calibration is performed for each pixel in the image to determine the position information of each element in the image, so as to be able to obtain the area and shape descriptor of the modeling target in the sampling image and the theoretical target image. The area difference can also be used to judge whether there is deformation of the modeling target in the image, and the deviation of its shape characteristics can be more intuitively reflected according to the shape descriptor. The texture feature is extracted through the gray-level co-occurrence matrix method, and the detailed characteristics of the modeling target are detailedly reflected through the elements of the gray-level co-occurrence matrix. The key point positions that have a significant influence on the modeling in the image are extracted through the SIFT algorithm. After obtaining the various features of the two images, the spatial distribution value of the deviation of the key point positions in the two images is statistically matched in space. The spatial distribution value can intuitively show the distribution state of the deviation of the key point positions in the entire modeling target space. The deviation coefficient is calculated through the interference deviation model according to the difference in the image features of the sampling image and the theoretical target image, and the first-level deviation threshold, the second-level deviation threshold, and the spatial distribution threshold are set. The deviation threshold represents the deviation value of the image features between the two images, and it can be reflected in the form of quantifying the deviation, which is convenient for judging the specific deviation degree of the image. When the deviation coefficient is higher than the first-level deviation threshold and lower than the second-level deviation threshold, and the spatial distribution value is lower than the spatial distribution threshold, it indicates that there is a certain degree of difference between the sampling image and the theoretical target image, but this difference is mainly concentrated in the local area and has a relatively small impact on the overall modeling. Therefore, it is output that there is a microscopic deviation at this basic surveying and mapping point. At this time, it is output that there is a microscopic deviation at this basic surveying and mapping point. At this time, the subsequent aerial photography path can be finely adjusted through the microscopic path adjustment strategy to ensure the accuracy of the aerial photography data.When the deviation coefficient is higher than the secondary deviation threshold and the spatial distribution value is higher than the spatial distribution threshold, it indicates that there is a large difference between the sampled image and the theoretical target image, and this difference is widely distributed in space, which may have a serious impact on the overall modeling. Therefore, it is output that there is a global deviation at this basic surveying and mapping point. In this case, it is necessary to activate the global path adjustment strategy to re-plan and adjust the entire aerial photography path to ensure that aerial photography data meeting the modeling requirements is obtained. Through the above path interference compensation steps, the present invention can accurately judge the types of deviations occurring during the UAV aerial photography process in real time, and adopt corresponding adjustment strategies according to different deviation types, effectively improving the accuracy of the UAV aerial photography path planning and the reliability of dynamic modeling, and providing more efficient and accurate technical support for applications in related fields.

[0063] Specifically, the interference deviation model is configured as follows:

[0064] ;

[0065] Among them, D is the deviation coefficient, which is used to judge the deviation degree between the sampled image and the theoretical target image. The main function of the interference deviation model is to calculate a deviation coefficient D, which can comprehensively and synthetically reflect the difference degree between the sampled image and the theoretical target image. During the UAV aerial photography process, due to the interference of various factors, such as environmental light changes and unstable UAV attitudes, the sampled image may deviate from the theoretical target image. The deviation coefficient D calculated by the interference deviation model can quantify this deviation, thereby helping us accurately judge whether micro-path adjustment or global-path adjustment is needed to ensure the accuracy of aerial photography data and the reliability of modeling; n represents the number of pairs of pictures at the modeling target in the image, usually 1. When continuous sampling is required at this basic surveying and mapping point, n is the number of pairs of corresponding sampled images and theoretical target images; and are the contour perimeters of the i-th sampled image and the theoretical target image respectively. In order to eliminate the influence of the difference in the numerical size of the contour perimeters between different images, max( , ) is used for normalization processing by the maximum value, so that the contour perimeters of different images can be compared on a unified scale; and are the area components of the i-th sampled image and the theoretical target image in the j direction respectively. The area component represents the length after the modeling target is mapped in the j direction. By decomposing the area of the modeling target in multiple directions, multiple area components are obtained. m represents the number of area components, and respectively represent the norms of the area vectors of the $i$-th sampled image and the theoretical target image, which are comprehensive measures of multiple area components. For a modeling target with an irregular shape, its area metric value can be calculated through the area components, so as to more comprehensively reflect the area characteristics of the modeling target;

[0066] and respectively represent the shape descriptors of the $i$-th sampled image and the theoretical target image. The shape descriptor is used to describe the shape characteristics of the modeling target, and $d$ H represents the Hausdorff distance between the shape descriptors, which is a distance metric between the shape descriptors and a way to measure the distance between two points, used to measure the difference between the shape descriptors. ; and respectively represent the texture feature values of the $i$-th sampled image and the theoretical target image, is the domain of the texture feature. When calculating the difference of the texture feature, it is limited within the domain for comparison to ensure the rationality and effectiveness of the calculation.

[0067] and respectively represent the position vectors of the $k$-th key point of the $i$-th sampled image and the theoretical target image, and $p$ is the number of key points, and respectively represent the norms of the sets of all key point position vectors of the $i$-th sampled image and the theoretical target image, which represent the sum of the distances from all key points to the centroid. By using the key point position differences, the deviation between the images can be judged more accurately. $\alpha$, $\beta$, $\gamma$, $\delta$ and $\varepsilon$ respectively represent the weight factors of each parameter, where the value of $\varepsilon$ is greater than $\alpha$, $\beta$, $\gamma$, $\delta$, and $\alpha+\beta+\gamma+\delta+\varepsilon = 1$.

[0068] During the UAV aerial photography process, due to the interference of various factors, such as slight air current influence, sensor errors, etc., there may be certain differences between the sampled image and the theoretical target image. However, this kind of difference is mainly concentrated locally and has little impact on the overall modeling, that is, micro deviations occur. The main function of the micro path adjustment strategy is to calculate the offset of the basic surveying and mapping points according to this deviation information and adjust their position coordinates, so that the subsequently acquired sampled images are closer to the theoretical target image and reduce the impact of deviations on modeling. This strategy can quickly and effectively correct local deviations without large-scale path re-planning, improving the aerial photography efficiency and the accuracy of modeling. Specifically, the micro path adjustment strategy includes calculating the offset of the basic surveying and mapping points according to the contour perimeter, area, shape descriptor, texture features and key point position deviation values of the modeling target in the sampled image and the theoretical target image, adjusting the position coordinates of the basic surveying and mapping points according to the offset of the basic surveying and mapping points, and re-acquiring the sampled image of the modeling target at the updated position of the basic surveying and mapping points. The contour perimeter of the modeling target in the sampled image and the theoretical target image is extracted respectively through the edge detection algorithm, and then their difference is calculated as the contour perimeter deviation value. The area of the modeling target in the sampled image and the theoretical target image is obtained by using the pixel coordinate calibration and area calculation method, and the area deviation value is calculated. The shape descriptors of the sampled image and the theoretical target image are obtained through the shape description method of Hu moments, and the difference between them is calculated through the Hausdorff distance metric to obtain the shape descriptor deviation value. The calculation of the offset of the basic surveying and mapping points is configured as:

[0069] ;

[0070] wherein, is the comprehensive correction vector, representing the offset of the basic surveying and mapping point coordinates, (v PAx 、v PAy ) is the first correction vector based on the deviation of the contour perimeter and area, where , , k P is the contour perimeter deviation correction coefficient, k A is the area deviation correction coefficient, is the contour perimeter deviation, is the area deviation, P0 is the reference contour perimeter of the modeling target based on the theoretical target image, A0 is the reference area of the modeling target based on the theoretical target image, θ represents the correction direction angle calculated according to the deviation of the contour perimeter and area, (v Jx 、v Jy ) is the second correction vector based on the deviation of the shape descriptor, , , wherein, ΔH i is the i-th element in the difference vector matrix, ωi is the weight coefficient of the i-th element in the x direction, ω ＇ i is the weight coefficient of the i-th element in the y direction, (v Xx 、v Xy ) is the third correction vector for the deviation based on the texture feature, , , where k X is the texture deviation correction coefficient, α is the correction direction angle determined according to the texture deviation distribution, and (Δx, Δy) is the key point position deviation;

[0071] The main function of the global path adjustment strategy is to re-evaluate and adjust the comprehensive benefits of basic surveying and mapping points when global deviations occur, so as to re-plan the aerial photography path, improve the quality of aerial photography data and the accuracy of modeling. During the actual aerial photography process, due to the influence of various complex factors, such as environmental changes and large deviations in the flight attitude of the UAV, there may be a large difference between the sampled image and the theoretical target image, affecting the modeling effect. Through the global path adjustment strategy, the priority of basic surveying and mapping points can be dynamically adjusted according to the deviation situation, enabling the UAV to collect data more targeted, reducing ineffective shooting, and improving the aerial photography efficiency and modeling quality. Specifically, the global path adjustment strategy includes calculating the benefit adjustment factor of the basic surveying and mapping point at the corresponding position of the sampled image according to the deviations of the contour perimeter, area, shape descriptor, texture feature and key point position. The benefit adjustment factor includes the texture richness adjustment factor, the feature salience adjustment factor and the flight distance adjustment factor. These deviation values reflect the degree of difference between the sampled image and the theoretical target image in various aspects and are important bases for calculating the benefit adjustment factor. The texture richness adjustment factor is used to measure the impact of the difference in the texture features of the modeling target in the sampled image and the theoretical target image on the benefit of the basic surveying and mapping point. If the texture feature of the sampled image is too simple or complex compared with the theoretical target image, it may affect the accuracy of modeling, so it is necessary to adjust the benefit of the basic surveying and mapping point. The feature salience adjustment factor considers the impact of the differences in the shape, contour and other features of the modeling target in the sampled image and the theoretical target image on the benefit of the basic surveying and mapping point. The flight distance adjustment factor considers the impact of the flight distance of the UAV from the current position to the basic surveying and mapping point on the benefit, and recalculates the comprehensive benefit adjustment value of the basic surveying and mapping point through the comprehensive benefit recalculation model according to the comprehensive benefit adjustment factor. Replace the comprehensive benefit value of the basic surveying and mapping point with the recalculated comprehensive benefit adjustment value as the final comprehensive benefit value, re-prioritize it according to the comprehensive benefit value of the basic surveying and mapping points in each key surveying area, and re-select the basic surveying and mapping points according to the priority ranking result for the second aerial photography path planning. The global path adjustment strategy has the characteristics of self-adaptation and flexibility, can dynamically adjust the aerial photography path according to the actual global deviation situation, and improve the quality of aerial photography data. This strategy is applicable to various complex dynamic modeling scenarios, especially in situations where the environment changes greatly and the UAV flight attitude is prone to large deviations, such as when modeling in mountainous areas and areas with high-rise buildings in cities, it can effectively cope with global deviations and ensure the smooth progress of the modeling work.

[0072] Specifically, the comprehensive benefit recalculation model is configured with:

[0073] ;

[0074] The core function of the comprehensive revenue recalculation model is to recalculate the comprehensive revenue value of basic surveying and mapping points in the global path adjustment strategy based on the deviation situations between the sampled image and the theoretical target image in multiple aspects. The application of this model enables the aerial photography path planning to more accurately adapt to the actual shooting situation, improving the quality of aerial photography data and the accuracy of modeling. During the actual aerial photography process, due to the interference of various complex factors, global deviations may occur. At this time, the original comprehensive revenue value can no longer accurately reflect the actual value of the basic surveying and mapping points. By recalculating the comprehensive revenue adjustment value through this model, a more scientific basis can be provided for subsequent path replanning. Among them, is the comprehensive revenue adjustment value, g1 is the texture richness adjustment factor, which is used to represent the influence of the deviation value on the texture richness in the comprehensive revenue value, , 、 are the adjustment coefficients of the contour perimeter and area respectively. In some modeling scenarios with high requirements for texture details, and may be increased to more sensitively reflect the influence of the deviation of texture features on the comprehensive revenue. If the contour perimeter and area of the sampled image are significantly different from those of the theoretical target image, the value of g1 will increase accordingly, thereby reducing the comprehensive revenue of this basic surveying and mapping point in terms of texture richness. g2 is the feature saliency adjustment factor , 、 are adjustment factors, which are used to represent the influence of the deviation value on the feature saliency in the comprehensive revenue value. For modeling targets with regular shapes, may increase the weight of the shape descriptor deviation; while for targets with large contour variations, may increase the weight of the contour perimeter deviation. Through the calculation of g2, the value of the basic surveying and mapping point in terms of feature saliency can be more accurately evaluated, thereby reasonably adjusting the comprehensive revenue.

[0075] g3 is the flight distance adjustment factor , which is used to represent the influence of the deviation value on the flight distance in the comprehensive revenue value. During the aerial photography process, the farther the flight distance, the more energy the unmanned aerial vehicle consumes and the longer the shooting time required, which will reduce the actual revenue of this basic surveying and mapping point. Therefore, g3 is used to quantify this influence, enabling the selection of basic surveying and mapping points with shorter flight distances to be prioritized when replanning the aerial photography path, improving the aerial photography efficiency. g4 is the flight tortuosity adjustment factor, which is used to represent the influence of the deviation value on the flight tortuosity in the comprehensive revenue value. A tortuous flight path will increase the flight difficulty and time cost of the unmanned aerial vehicle, and also affect the shooting stability and data quality. By considering the flight tortuosity, basic surveying and mapping points with overly tortuous flight paths can be avoided when recalculating the comprehensive revenue, thereby optimizing the aerial photography path.

[0076] The basic features, principles and advantages of the present invention have been shown and described above. It should be noted that the present invention is not limited by the above embodiments, which are only partial embodiments. Without departing from the spirit and scope of the present invention, several improvements and supplements made are regarded as the protection scope of the present invention.

Claims

1. A method for UAV aerial photography path planning for dynamic modeling, characterized in that, The steps are as follows: Target data acquisition step: acquiring a remote sensing image of the modeling target as the model analysis image, and analyzing the size, shape, and structural features of the modeling target from the model analysis image as the basic model information; Modeling structure analysis step: determining the structural type of the modeling target through a structure analysis model based on the basic model information, where the structural type includes a conventional three-dimensional structure and a special-shaped three-dimensional structure, constructing a theoretical target model, and analyzing the basic surveying and mapping points of the modeling target based on the structural type combined with the basic model information; Surveying and mapping point benefit calculation step: obtaining sampling information and surveying and mapping costs through a surveying and mapping benefit calculation strategy based on the basic surveying and mapping points, and calculating the comprehensive benefit value of the basic surveying and mapping points through a comprehensive benefit model based on the sampling information and surveying and mapping costs of the basic surveying and mapping points. The comprehensive benefit model is configured with: ； Among them, R i is the comprehensive income value of the i-th basic surveying and mapping point, C i is the pixel clarity of the i-th basic surveying and mapping point, T i is the texture richness of the i-th basic surveying and mapping point, S i is the feature saliency of the i-th basic surveying and mapping point, indicating the degree of prominence of the features of this surveying and mapping point; D i is the flight distance from the drone to the position of the i-th basic surveying and mapping point, W i is the flight tortuosity of the drone at the position of the i-th basic surveying and mapping point, indicating the flight difficulty of the drone at this point; E i is the environmental impact factor at the position of the i-th basic surveying and mapping point; Aerial photography path planning step: dividing the basic surveying and mapping points into regions and forming several key surveying and mapping regions according to the structural features through a surveying and mapping point planning strategy, prioritizing the basic surveying and mapping points in each key surveying and mapping region according to the comprehensive benefit value, and selecting the basic surveying and mapping points according to the priority ranking result to complete the unmanned aerial vehicle (UAV) aerial photography path planning; Path interference compensation step: when the UAV reaches the basic surveying and mapping point in the aerial photography path, acquiring the sampling image of the modeling target in real time, comparing the features of the sampling image and the theoretical target model, calculating the deviation coefficient through an interference deviation model according to the feature comparison result, outputting a micro deviation or a global deviation according to the comparison result between the deviation coefficient and the deviation threshold. When the output is a micro deviation, entering the micro path adjustment strategy; when the output is a global deviation, entering the global path adjustment strategy.

2. The method for unmanned aerial vehicle aerial photography path planning for dynamic modeling according to claim 1, wherein The structure analysis model analyzes the structural type of the modeling target according to the shape and structural features of the modeling target. When the structure analysis model determines that the similarity between the shape of the modeling target and the preset geometric parameters is greater than the preset threshold, and the number of symmetric structures and structural repeating units in the structural features of the modeling target is greater than the preset threshold, it outputs the structural type as a conventional three-dimensional structure; otherwise, it outputs the structural type as a special-shaped three-dimensional structure. When the structural type is a conventional three-dimensional structure, the basic surveying and mapping points are selected from the geometric center plane, connection points, and key feature points of each geometric shape of the modeling target according to the shape and size of the modeling target; when the structural type is a special-shaped three-dimensional structure, the basic surveying and mapping points are set at special nodes according to the shape information of the modeling target, including densely selecting basic surveying and mapping points at positions where the curve curvature changes violently, and setting basic surveying and mapping points at the turning points of the broken line according to the analysis of the model analysis image.

3. A method for UAV aerial photography path planning for dynamic modeling according to claim 1, characterized in that, The mapping revenue calculation strategy includes extracting the pixel clarity, texture richness, and feature saliency of the area where the basic mapping points are located in the model analysis image according to the structure type as sampling information, and comprehensively obtaining the mapping cost by analyzing the UAV flight distance, the flight tortuosity of the UAV at key points, and the environmental information based on the size and shape of the modeling target. The comprehensive revenue value of each basic mapping point is calculated through the comprehensive revenue model for the sampling information of each mapping point and the mapping cost respectively.

4. A method for unmanned aerial vehicle aerial photography path planning for dynamic modeling according to claim 1, characterized in that, The aerial photography path planning step further includes dividing the key mapping areas according to the spatial coordinates of the basic mapping points through the distribution density and the structural characteristics of the modeling target, constructing a comprehensive revenue matrix for the basic mapping points within each key mapping area based on the comprehensive revenue value, confirming the weights of each basic mapping point by analyzing the comprehensive revenue matrix, and sorting the basic mapping points from high to low according to the weights.

5. A method for unmanned aerial vehicle aerial photography path planning for dynamic modeling according to claim 1, characterized in that, The path interference compensation step further includes extracting the theoretical target image corresponding to the position extracted from the basic mapping points of the sampling image for the theoretical target model, respectively extracting the image features of the sampling image and the theoretical target image through edge detection and pixel coordinate calibration. The image features include the contour perimeter, area, shape descriptor, texture feature, and key point position of the modeling target in the sampling image. The spatial distribution value is obtained by statistically analyzing the spatial distribution of the deviation of the key point positions in the two images. The deviation coefficient is calculated through the interference deviation model based on the difference in the image features of the sampling image and the theoretical target image. The first deviation threshold, the second deviation threshold, and the spatial distribution threshold are set. When the deviation coefficient is higher than the first deviation threshold and lower than the second deviation threshold, and the spatial distribution value is lower than the spatial distribution threshold, it is output that there is a micro deviation at this basic mapping point. When the deviation coefficient is higher than the second deviation threshold and the spatial distribution value is higher than the spatial distribution threshold, it is output that there is a global deviation at this basic mapping point.

6. A method for UAV aerial photography path planning for dynamic modeling according to claim 1, characterized in that, The interference deviation model is configured with: ； Wherein, D is the deviation coefficient, which is used to judge the deviation degree between the sampled image and the theoretical target image; n represents the number of pairs of pictures at the modeling target in the image, usually 1. When continuous sampling is required for this basic surveying and mapping point, n is the number of pairs of the corresponding sampled image and the theoretical target image; and are respectively the contour perimeters of the i-th sampled image and the theoretical target image, and max( , ) represents normalization by the maximum value; and are respectively the area components of the i-th sampled image and the theoretical target image in the j direction. The area component represents the length of the mapped modeling target in the j direction, and m represents the number of area components. and respectively represent the norms of the area vectors of the i-th sampled image and the theoretical target image; and respectively represent the shape descriptors of the i-th sampled image and the theoretical target image, and d H represents the Hausdorff distance between the shape descriptors, indicating the distance metric between the shape descriptors; and respectively represent the texture feature values of the i-th sampled image and the theoretical target image, is the domain of definition of the texture feature; and respectively represent the position vectors of the k-th key point of the i-th sampled image and the theoretical target image, and p is the number of key points. and respectively represent the norms of the sets of all key point position vectors of the i-th sampled image and the theoretical target image, representing the sum of the distances from all key points to the centroid. α, β, γ, δ, and ε respectively represent the weight factors of each parameter, where the value of ε is greater than α, β, γ, δ, and α + β + γ + δ + ε = 1.

7. A method for UAV aerial photography path planning for dynamic modeling according to claim 1, characterized in that, The micro path adjustment strategy includes calculating the offset of the basic mapping point according to the contour perimeter, area, shape descriptor, texture feature, and key point position deviation values of the modeling target in the sampling image and the theoretical target image, adjusting the position coordinates of the basic mapping point according to the offset of the basic mapping point, and re - obtaining the sampling image of the modeling target at the updated position of the basic mapping point. The calculation of the offset of the basic mapping point is configured with: ； Among them, is the comprehensive correction vector, representing the offset of the basic surveying and mapping point coordinates. (v PAx , v PAy ) is the first correction vector based on the deviation of the contour perimeter and area, where , , k P is the contour perimeter deviation correction coefficient, k A is the area deviation correction coefficient. is the contour perimeter deviation, is the area deviation, P0 is the reference contour perimeter of the modeled object in the theoretical target image, A0 is the reference area of the modeled object in the theoretical target image, θ represents the correction direction angle calculated based on the deviation of the contour perimeter and area, (v Jx , v Jy ) is the second correction vector based on the deviation of the shape descriptor. , , where ΔH i is the i-th element in the difference vector matrix, ω i is the weight coefficient of the i-th element in the x direction, ω ＇ i is the weight coefficient of the i-th element in the y direction, (v Xx , v Xy ) is the third correction vector based on the deviation of the texture feature. , , where k X is the texture deviation correction coefficient, α is the correction direction angle determined according to the texture deviation distribution, and (Δx, Δy) is the key point position deviation.

8. A method for UAV aerial photography path planning for dynamic modeling according to claim 7, characterized in that The global path adjustment strategy includes calculating a benefit adjustment factor for the basic surveying and mapping points at the corresponding positions of the sampled images based on the deviations of the contour perimeter, area, shape descriptors, texture features, and key point positions. The benefit adjustment factor includes a texture richness adjustment factor, a feature saliency adjustment factor, and a flight distance adjustment factor. Then, based on the comprehensive benefit adjustment factor, the comprehensive benefit adjustment value of the basic surveying and mapping points is recalculated through a comprehensive benefit recalculation model. The recalculated comprehensive benefit adjustment value is used to replace the comprehensive benefit value of the basic surveying and mapping points as the final comprehensive benefit value. The basic surveying and mapping points in each key surveying and mapping area are re-prioritized according to their comprehensive benefit values, and the basic surveying and mapping points are re-selected according to the priority ranking results for the second aerial photography path planning.

9. A method for UAV aerial photography path planning for dynamic modeling according to claim 8, characterized in that, The comprehensive benefit recalculation model is configured with: ； Among them, is the comprehensive income adjustment value, C i is the pixel clarity of the i-th basic surveying and mapping point, T i is the texture richness of the i-th basic surveying and mapping point, S i is the feature salience of the i-th basic surveying and mapping point, indicating the degree of prominence of the features of this surveying and mapping point; D i is the flight distance from the drone to the position of the i-th basic surveying and mapping point, W i is the flight tortuosity of the drone at the position of the i-th basic surveying and mapping point, indicating the flight difficulty of the drone at this point; E i is the environmental impact factor at the position of the i-th basic surveying and mapping point; g1 is the texture richness adjustment factor, used to represent the impact of the deviation value on the texture richness in the comprehensive income value, , 、 are the adjustment coefficients of the contour perimeter and area respectively, and g2 is the feature salience adjustment factor , 、 are adjustment factors, used to represent the impact of the deviation value on the feature salience in the comprehensive income value, and g3 is the flight distance adjustment factor , used to represent the impact of the deviation value on the flight distance in the comprehensive income value, and g4 is the flight tortuosity adjustment factor, used to represent the impact of the deviation value on the flight tortuosity in the comprehensive income value.

Images