Real estate surveying and mapping area management method and system based on remote sensing image

Abstract

The invention discloses a real estate surveying and mapping area management method and system based on remote sensing images, and particularly relates to the technical field of real estate surveying and mapping, and the method comprises the steps: photographing a real estate surveying and mapping area through an unmanned aerial vehicle, and obtaining the image data of the real estate of the unmanned aerial vehicle under different flight paths and different attitudes; designing a fitness function of an unmanned aerial vehicle flight path genetic algorithm, optimizing the unmanned aerial vehicle flight path, performing optimization in combination with flight information and remote sensing image information of the unmanned aerial vehicle in different flight paths, and obtaining quality constraint coefficients of different real estate in the unmanned aerial vehicle flight path through error information and appearance information of each real estate; and setting a quality constraint coefficient threshold, and determining the optimal flight path of the unmanned aerial vehicle in the real estate surveying and mapping area based on a double-layer optimization strategy in combination with the fitness function and the quality constraint coefficient of the genetic algorithm, so that the invention provides an efficient and accurate unmanned aerial vehicle flight path planning scheme which is beneficial to the surveying and mapping precision and efficiency.

Description

Technical Field

[0001] The present invention relates to the field of real estate surveying and mapping technology, and more specifically, to a real estate surveying and mapping area management method and system based on remote sensing images. Background Art

[0002] Real estate surveying and mapping by drones can obtain high-precision ground image data, which has important applications in many fields such as urban planning, land management, and real estate evaluation. In order to ensure the quality of real estate surveying and mapping data, flight path planning is particularly critical. Optimizing the flight path of drones can not only improve surveying and mapping efficiency, but also ensure the clarity and integrity of the image. At present, most UAV flight path optimization methods mainly focus on factors such as flight time, heading angle and flight stability. However, these methods often ignore the influence of factors such as image quality, error information and shape fitting. When surveying a large number of real estate with as little adjustment of the UAV posture as possible, it is not easy to comprehensively consider the surveying efficiency and the accuracy of the surveying data.

[0003] In order to solve the above defects, a technical solution is now provided. Summary of the invention

[0004] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present invention provide a real estate surveying and mapping area management method and system based on remote sensing images to solve the problems raised in the above-mentioned background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions: The real estate surveying and mapping area management method based on remote sensing images specifically includes the following steps: S1: Use drones to photograph the real estate surveying area and obtain image data of the real estate under different flight paths and postures; S2: Design the fitness function of the UAV flight path genetic algorithm, optimize the UAV flight path, and combine the flight information and remote sensing image information of the UAV in different flight paths for optimization; S3: Obtain the mass constraint coefficients of different real estates in the UAV flight path through the error information and shape information of each real estate, and set the mass constraint coefficient threshold; S4: Combining the fitness function and mass constraint coefficient of the genetic algorithm, the optimal flight path of the UAV in the real estate mapping area is determined based on a two-level optimization strategy.

[0006] In a preferred embodiment, the flight information includes: The flight information of the UAV on different flight paths is represented by flight time and area coverage coefficient; Among them, the flight time is determined by the historical records of the UAV in different flight paths, which represents the total time the UAV completes the mapping in the flight path; The logic for obtaining the area coverage coefficient is as follows: obtaining the remote sensing image collected by the drone according to the flight path of the drone, dividing the real estate area and the non-real estate area in the remote sensing image collected by the drone, obtaining the number of pixels of the real estate area and the non-real estate area in the remote sensing image, and marking the number of pixels of the real estate area and the non-real estate area in the remote sensing image as follows: and ; According to the resolution of the remote sensing image, the areas of the real estate area and the non-real estate area in the remote sensing image are determined, and the areas of the real estate area and the non-real estate area in the remote sensing image are marked as follows: and ,in, , , is the area represented by each pixel; Determine the total area of ​​each property in the property area based on cadastral information, and mark the total area of ​​each property in the property area as: ; Calculate the coverage deviation of the flight path, the calculation formula is: ;in, is the coverage deviation of the flight path; Calculate the false coverage rate of the flight path, the calculation formula is: ;in, is the false coverage rate of the flight path; Calculate the area coverage coefficient using the following formula: ;in, is the area coverage factor.

[0007] In a preferred embodiment, the remote sensing image information includes: Remote sensing image information is represented by the clarity anomaly variation coefficient; The acquisition logic of the definition abnormal variation coefficient is as follows: according to the shooting process of the UAV in the flight path, a time interval T is set, and images shot by the UAV are collected and analyzed every time interval T in the shooting process in the flight path, and the Laplace operator is applied to the images shot by the UAV every time interval T to calculate the image definition, and the definition of the images shot by the UAV every time interval T is marked as: , where n=1, 2, 3, ..., N, N is a positive integer, and n is the number of the image collected by the drone at every time interval T during the shooting process in the flight path; Calculate the mean and standard deviation of the clarity of the images taken by the drone in the flight path, and mark the mean and standard deviation of the clarity of the images taken by the drone in the flight path as: and ;in, , ; Calculate the coefficient of variation of the clarity of the images taken by the drone in the flight path. The calculation formula is: ;in, The coefficient of variation of the clarity of the images taken by the drone in the flight path; Calculate the clarity abnormal variation coefficient, the calculation formula is: ;in, is the clarity abnormal variation coefficient.

[0008] In a preferred embodiment, the fitness function of the UAV flight path genetic algorithm is designed, including: Through comprehensive analysis of the flight information and remote sensing image information of the UAV on different flight paths, the flight time, area coverage coefficient and clarity abnormal change coefficient are weighted and calculated to generate a path evaluation model and a path evaluation coefficient. The calculation formula of the path evaluation coefficient is: ;in, is the path evaluation coefficient of the UAV flight path, SJ is the flight time of the UAV flight path, They are the proportional coefficients of flight time, area coverage coefficient, and clarity abnormal variation coefficient, They are both greater than 0; The path evaluation coefficient is used as the fitness function of the genetic algorithm.

[0009] In a preferred embodiment, obtaining mass constraint coefficients of different real estates in the flight path of the drone includes: The quality of each UAV flight path is evaluated by maximizing the fitness function. For each flight path, the quality constraint coefficient of each real estate of each flight path is checked to ensure that each real estate meets the requirements of the predetermined quality constraint coefficient. Through the removal mechanism, the flight path is removed from the population to avoid entering the next generation. The quality constraint coefficient requirement is set by setting the quality constraint coefficient threshold, comparing the quality constraint coefficient of each real estate area with the quality constraint coefficient threshold. If there is real estate with a value greater than the quality constraint coefficient threshold in the flight path, the removal mechanism is triggered. If there is no real estate with a value greater than the quality constraint coefficient threshold in the flight path, the best performing individual is selected from the population as the parent according to the value of the fitness function. The parent individuals will perform crossover and mutation operations according to their fitness, and generate offspring by exchanging part of the path information of the parent individuals. The crossover operation generates new individuals and finds a better solution in different flight paths. When crossing, it is ensured that the generated offspring path meets the requirements of the quality constraint coefficient. Recalculate the fitness of the offspring paths after crossover and mutation, and check whether the quality constraint coefficient of the real estate in each flight path is greater than the quality constraint coefficient threshold. Select the individuals with the highest fitness to enter the next generation population, reach the preset algebraic upper limit or the fitness function converges, and output the optimal flight path as the final solution.

[0010] In a preferred embodiment, the mass constraint coefficient includes: The error information and shape information of each real estate in the remote sensing image are collected, and the error information of each real estate is represented by the special point deviation coefficient, and the shape information of each real estate is represented by the shape fitting difference coefficient. The quality constraint coefficient is generated by weighted calculation of the special point deviation coefficient and the shape fitting difference coefficient. The calculation formula of the quality constraint coefficient is: ;in, is the mass constraint coefficient, is the special point deviation coefficient, is the shape fitting difference coefficient, , is the proportional coefficient of the special point deviation coefficient and the shape fitting difference coefficient, Both are greater than 0.

[0011] In a preferred embodiment, the acquisition logic of the special point deviation coefficient is: Obtain the special points of the real estate, determine the coordinate positions of the special points of the real estate in the remote sensing image, and mark the coordinate positions of the special points of the real estate in the remote sensing image as: , where m = 1, 2, 3, ..., M, M is a positive integer, and m is the number of the special point of the real estate; Get the actual geographic coordinates of the real estate special point, and mark the actual geographic coordinates of the real estate special point as: ; Calculate the distance error of special points of real estate, the calculation formula is: ;in, is the distance error of each specific point of the real estate; Set the distance error threshold and mark the distance error threshold as: , compare the distance error of each special point of the real estate with the distance error threshold, obtain the special points whose distance errors are greater than the distance error threshold, and re-mark the distance errors of the special points greater than the distance error threshold as: , where j=1, 2, 3, ..., J, J is a positive integer, and j is the number of the special point that is greater than the distance error threshold; Calculate the special point deviation coefficient, the calculation formula is: .

[0012] In a preferred embodiment, the logic for obtaining the shape fitting difference coefficient is: Extract the contour of the real estate from the remote sensing image, obtain the boundary point set of the remote sensing image contour, obtain the boundary point set of the actual cadastral contour, and express the boundary point set of the remote sensing image contour and the boundary point set of the actual cadastral contour with complex numbers, where the complex form of the boundary point set of the remote sensing image contour is: , the plural form of the boundary points of the actual cadastral shape is: ,k=1, 2, 3, …, K, K is a positive integer, k is the number of the sampling point of the real estate boundary, and i is an imaginary unit; Perform discrete Fourier transform on the complex form, calculate the Fourier descriptor, and mark the Fourier descriptor of the remote sensing image shape as: ,in, is the Fourier descriptor of the remote sensing image appearance, that is, the Fourier transform result corresponding to frequency c in the frequency domain, is a complex index, indicating the rotation factor corresponding to each frequency component, c=1, 2, 3, ..., K, c is the index of the corresponding frequency in the frequency domain, and the actual cadastral shape Fourier descriptor is marked as: ,in, is the Fourier descriptor of the actual cadastral shape; According to the difference of Fourier components, the shape fitting difference coefficient is calculated, and the calculation formula is: ; Where h is the number of low-frequency components selected.

[0013] In a preferred embodiment, a real estate surveying and mapping area management system based on remote sensing images includes a data acquisition module, a path evaluation module, a quality constraint module, and a genetic algorithm module, and the modules are signal-connected; Data acquisition module, used to capture remote sensing images at multiple viewing angles and postures, ensuring coverage of different areas and providing multi-angle image information for subsequent path optimization and quality assessment; The path evaluation module is used to design the genetic algorithm fitness function of the UAV flight path and evaluate the flight path by combining flight information and remote sensing image information; The quality constraint module is used to obtain the quality constraint coefficient of each real estate under the flight path and determine whether the path meets the set quality threshold; The genetic algorithm module is used to optimize the flight path using the genetic algorithm, combining the fitness function of the path evaluation module and the constraints of the quality constraint module to ensure that the final path meets the optimal standard.

[0014] Technical effects and advantages of the present invention: The present invention uses a drone to shoot real estate surveying and mapping areas to obtain real estate image data under different flight paths and postures, provide basic data for subsequent path optimization and quality evaluation, design a fitness function, and combine the flight information and remote sensing image data of the drone flight path to optimize the flight path. The genetic algorithm optimizes the path through continuous evolution to improve the global performance of the path. By analyzing the error information and shape information of each real estate, the quality constraint coefficients of different real estate in different flight paths are calculated, and a quality constraint coefficient threshold is set to ensure that the flight path can meet the accuracy requirements. In combination with the fitness function and the quality constraint coefficient of the genetic algorithm, a double-layer optimization strategy is used to determine the optimal flight path. In the optimization process, both the global goal is considered and the quality constraint of each real estate area is ensured to be met. The present invention provides an efficient and accurate drone flight path planning scheme, which is conducive to the accuracy and efficiency of surveying and mapping. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to facilitate understanding by those skilled in the art, the present invention is further described below in conjunction with the accompanying drawings; Figure 1 It is a flow chart of the real estate surveying and mapping area management method based on remote sensing images of the present invention; Figure 2 It is a structural schematic diagram of the real estate surveying and mapping area management system based on remote sensing images of the present invention. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] Embodiment 1, as Figure 1 A flow chart of a method for real estate surveying and mapping area management based on remote sensing images of the present invention is provided, which specifically includes the following steps: S1: Use drones to photograph the real estate surveying area and obtain image data of the real estate under different flight paths and postures; S2: Design the fitness function of the UAV flight path genetic algorithm, optimize the UAV flight path, and combine the flight information and remote sensing image information of the UAV in different flight paths for optimization; S3: Obtain the mass constraint coefficients of different real estates in the UAV flight path through the error information and shape information of each real estate, and set the mass constraint coefficient threshold; S4: Combining the fitness function and mass constraint coefficient of the genetic algorithm, the optimal flight path of the UAV in the real estate mapping area is determined based on a two-level optimization strategy.

[0018] The remote sensing images of real estate surveying and mapping areas are image data of real estate obtained through remote sensing technology. Remote sensing images of different real estate can help complete the spatial positioning and feature extraction of real estate such as buildings, land boundaries, and facilities. Through remote sensing images, large areas can be monitored and mapped efficiently and quickly without the need for traditional on-site surveys.

[0019] Acquiring remote sensing images of real estate through drones means flying to the target area and collecting image data of the ground or buildings by carrying remote sensing equipment, including high-definition cameras, lidar, infrared sensors, etc. The advantage of drone remote sensing images is that drones can fly at a lower altitude and provide higher resolution images than satellite remote sensing, which can clearly capture details such as the boundaries of buildings, the height of houses, and small features of land. Drones can flexibly choose flight paths and altitudes based on demand and quickly cover target areas. They are particularly suitable for areas that are difficult to reach or for scenarios where a large amount of data needs to be collected in a short period of time.

[0020] Before flying, you first need to survey the target area and design the flight path of the drone. The design of the flight path needs to consider factors such as coverage, flight altitude, shooting angle, overlap rate, etc. to ensure that enough remote sensing images are obtained for subsequent analysis. The drone flies along the planned route and collects image data in real time. Different sensors are carried as needed. During data collection, the drone usually needs to remain stable during flight to ensure the quality and clarity of the image. Remote sensing image data collected by drones usually need to be post-processed, including image denoising, geometric correction, image stitching, etc. The stitched image data can generate a complete image map covering a wide area; Multiple images are registered according to a certain geographic coordinate system and stitched into a complete remote sensing image to ensure the geometric accuracy of the image and correspond each pixel in the image to the coordinates in the real world; Image analysis technology is used to extract building outlines, house heights, roads, green spaces and other information from remote sensing images. Commonly used techniques include image segmentation, edge detection, target recognition and other methods.

[0021] The genetic algorithm is used to determine the flight path of the drone when shooting remote sensing images of the real estate surveying area. The genetic algorithm is used to optimize the existing problems when the drone shoots remote sensing images, including: The attitude of the drone (such as pitch angle, roll angle, heading angle, etc.) needs to be constantly adjusted according to the requirements of the real estate to ensure the accuracy of the shooting angle and coverage area. During the flight, frequent attitude adjustments may cause the flight stability to decrease, or even cause temporary instability or tremor, affecting the clarity and accuracy of the image; The shooting angle and resolution of the drone in different postures will be different. Different postures may cause the angle of the image to change, thus affecting the geometric accuracy of the image and increasing the difficulty of subsequent processing. If the posture adjustment of the drone is not accurate, it may also lead to insufficient overlap of the shooting area, affecting the data coverage; During the flight of a drone, terrain changes, different flight altitudes, and shooting angles will affect the coverage of the image. Especially in complex terrain (such as mountainous areas and densely built urban areas), improper selection of flight altitude and flight path may lead to insufficient image coverage in some areas, affecting the accuracy and completeness of the data; The flight speed directly affects the resolution and clarity of the captured images. When the flight is too fast, the details of the captured images will be blurred; when the flight is too slow, the flight time may be too long, affecting the efficiency of the mission. Improper path planning may cause the drone to fail to cover all target areas during flight, or require multiple repeated flights, increasing time and energy consumption. During the shooting process, due to factors such as errors in the drone's attitude adjustment and inaccurate flight paths, the captured images may be misaligned or deformed, and need to be stitched and registered. If the overlap between images is insufficient or the flight path planning is unreasonable, the stitching process may be very complicated and prone to errors; During a long flight, due to GPS positioning errors, sensor errors and attitude adjustment errors, the image may gradually deviate from the actual position. Especially in large-scale shooting, the errors may gradually accumulate and affect the final mapping results.

[0022] It should be noted that different real estate has different requirements for flight paths and attitudes, which is mainly related to the following factors: Different types of real estate, such as high-rise buildings, independent houses, commercial buildings, etc., have different shapes and spatial distributions. For high-rise or large buildings, a larger flight range and higher flight altitude may be required to ensure that panoramic images can be obtained, while for low or scattered buildings, the flight path may need to be more precise to ensure sufficient coverage and image clarity.

[0023] In the process of UAV mapping of real estate areas, the flight path of the UAV is optimized through genetic algorithm, which includes generating an initial population, each individual in the population represents a possible solution, and each individual is represented as a flight path, which contains information such as the order of flight points, flight path, attitude adjustment, etc. In genetic algorithms, individuals usually need to be encoded in a way that is suitable for algorithm processing. For the flight path problem, individuals can be constructed using parameters such as the point order of the path, flight altitude, flight angle, etc. For example, a flight path can be represented as a sequence of multiple flight points, and each flight point can also contain information such as flight attitude and flight time.

[0024] The flight information and remote sensing image information of the UAV on different flight paths are determined through the historical records of UAV mapping of real estate areas. The flight information of the UAV on different flight paths is represented by the flight time and area coverage coefficient, and the remote sensing image information is represented by the clarity abnormality change coefficient.

[0025] Among them, the flight time is determined by the historical records of the UAV in different flight paths, which represents the total time the UAV completes the mapping in the flight path.

[0026] The logic for obtaining the area coverage coefficient is as follows: obtaining the remote sensing image collected by the drone according to the flight path of the drone, dividing the real estate area and the non-real estate area in the remote sensing image collected by the drone, obtaining the number of pixels of the real estate area and the non-real estate area in the remote sensing image, and marking the number of pixels of the real estate area and the non-real estate area in the remote sensing image as follows: and ; It should be noted that when drones are taking remote sensing images, they usually obtain multiple images through the flight path. These images usually overlap. In order to cover the entire area of ​​the target area and obtain continuous ground information, image stitching (or image mosaicking) technology is usually used to synthesize these separate remote sensing images into a complete image.

[0027] According to the resolution of the remote sensing image, the areas of the real estate area and the non-real estate area in the remote sensing image are determined, and the areas of the real estate area and the non-real estate area in the remote sensing image are marked as follows: and ,in, , , is the area represented by each pixel; Determine the total area of ​​each property in the property area based on cadastral information, and mark the total area of ​​each property in the property area as: ; It should be noted that cadastral information usually includes data such as the boundaries, location, area, and owner of each real estate unit, and is an official record that describes the physical properties of real estate.

[0028] Calculate the coverage deviation of the flight path, the calculation formula is: ;in, is the coverage deviation of the flight path; Calculate the false coverage rate of the flight path, the calculation formula is: ;in, is the false coverage rate of the flight path; Calculate the area coverage coefficient using the following formula: ;in, is the area coverage factor.

[0029] It can be seen from the formula that the larger the area coverage coefficient is, the more real estate is covered by the remote sensing image drawn by the flight path, and the more it can reflect the overall situation of the real estate area. In addition, the area of ​​non-real estate areas in the remote sensing image may be smaller, which means that the pixels in the remote sensing image are more concentrated in the target real estate area, reducing interference from irrelevant areas, making the mapping of real estate areas more accurate, and reducing the amount of image data that needs to be processed, thereby reducing the computing resources and time required for post-processing and analysis.

[0030] The acquisition logic of the definition abnormal variation coefficient is as follows: according to the shooting process of the UAV in the flight path, a time interval T is set, and images shot by the UAV are collected and analyzed every time interval T in the shooting process in the flight path, and the Laplace operator is applied to the images shot by the UAV every time interval T to calculate the image definition, and the definition of the images shot by the UAV every time interval T is marked as: , where n=1, 2, 3, ..., N, N is a positive integer, and n is the number of the image collected by the drone at every time interval T during the shooting process in the flight path; It should be noted that the time interval T is set by professional staff. The interval time indicates the frequency of acquisition. The minimum acquisition frequency can be one frame, which means that the clarity of the image will be calculated immediately after each acquisition. However, if the time interval T is too short, the complexity of the calculation will increase. Frequent calculation of the clarity of each frame of the image may bring unnecessary computational burden.

[0031] Calculate the mean and standard deviation of the clarity of the images taken by the drone in the flight path, and mark the mean and standard deviation of the clarity of the images taken by the drone in the flight path as: and ;in, , ; It should be noted that the change in the clarity of the UAV in the flight path reflects the change in the posture of the UAV in the flight path. During the flight, the change in posture of the human and the machine (for example, tilt, rotation, pitch, etc.) will directly affect the quality of the remote sensing image captured, especially the clarity of the image.

[0032] Calculate the coefficient of variation of the clarity of the images taken by the drone in the flight path. The calculation formula is: ;in, The coefficient of variation of the clarity of the images taken by the drone in the flight path; Calculate the clarity abnormal variation coefficient, the calculation formula is: ;in, is the clarity abnormal variation coefficient.

[0033] It can be seen from the formula that the larger the clarity abnormal variation coefficient is, the greater the attitude fluctuation or flight stability problem of the UAV in the flight path. Due to the need to shoot different real estate, the UAV needs to adjust its flight attitude when flying over different real estate, thereby constantly changing its flight attitude during the flight.

[0034] When the attitude of the drone changes, especially when the angle changes or the pitch angle is adjusted during flight, the field of view and shooting angle of the camera will change. If the drone has an angle deviation or an unstable attitude when shooting, it may cause image distortion, making part of the image blurred or distorted, affecting the clarity of the image. When the UAV's flight attitude is unstable, such as when the heading and pitch angle fluctuate too much, the camera's viewing angle will change irregularly. This instability may cause the image to be out of focus, especially when the attitude changes frequently in a short period of time, the clarity of the image will be affected. Therefore, by monitoring the fluctuation of image clarity, the stability of the flight attitude can be indirectly reflected. For example, if the clarity of the image drops sharply in certain time intervals, it may indicate that the UAV's attitude has changed significantly during this period, or that the UAV is unstable, resulting in a decrease in image quality.

[0035] Through comprehensive analysis of the flight information and remote sensing image information of the UAV on different flight paths, the flight time, area coverage coefficient and clarity abnormal change coefficient are weighted and calculated to generate a path evaluation model and a path evaluation coefficient. The calculation formula of the path evaluation coefficient is: ;in, is the path evaluation coefficient of the UAV flight path, SJ is the flight time of the UAV flight path, They are the proportional coefficients of flight time, area coverage coefficient, and clarity abnormal variation coefficient, Both are greater than 0.

[0036] It can be seen from the formula that the larger the area coverage coefficient and the smaller the flight time and clarity abnormal variation coefficient, the larger the path evaluation coefficient, which means that the drone spends less time in the flight path, photographs more real estate areas, and has a more stable clarity. Conversely, the smaller the area coverage coefficient and the larger the flight time and clarity abnormal variation coefficient, the smaller the path evaluation coefficient, which means that the drone spends more time in the flight path, photographs fewer real estate areas, and has a more unstable clarity.

[0037] The path evaluation coefficient is used as the fitness function of the genetic algorithm. The quality of each UAV flight path is evaluated by maximizing the fitness function. For each flight path, the quality constraint coefficient of each real estate of each flight path is checked to ensure that each real estate meets the requirements of the predetermined quality constraint coefficient. Through the removal mechanism, the flight path is removed from the population to avoid entering the next generation. The quality constraint coefficient requirement is set by setting the quality constraint coefficient threshold, comparing the quality constraint coefficient of each real estate area with the quality constraint coefficient threshold. If there is real estate with a value greater than the quality constraint coefficient threshold in the flight path, the removal mechanism is triggered. If there is no real estate with a value greater than the quality constraint coefficient threshold in the flight path, the best performing individual is selected from the population as the parent according to the value of the fitness function. The usual selection methods include roulette selection, tournament selection, etc. The parent individuals will perform crossover and mutation operations according to their fitness, and generate offspring by exchanging part of the path information of the parent individuals. The crossover operation generates new individuals and may find a better solution in different flight paths. When crossing, it is guaranteed that the generated offspring path meets the requirements of the quality constraint coefficient. Recalculate the fitness of the offspring paths after crossover and mutation, and check whether the quality constraint coefficient of the real estate in each flight path is greater than the quality constraint coefficient threshold. Select the individuals (flight paths) with the highest fitness to enter the next generation population, reach the preset algebraic upper limit or the fitness function converges, and output the optimal flight path as the final solution.

[0038] Among them, by collecting remote sensing images of the drone in the flight path, the error information and shape information of each real estate in the remote sensing image are collected, the error information of each real estate is represented by the special point deviation coefficient, and the shape information of each real estate is represented by the shape fitting difference coefficient. The quality constraint coefficient is generated by weighted calculation of the special point deviation coefficient and the shape fitting difference coefficient. The calculation formula of the quality constraint coefficient is: ;in, is the mass constraint coefficient, is the special point deviation coefficient, is the shape fitting difference coefficient, , is the proportional coefficient of the special point deviation coefficient and the shape fitting difference coefficient, , Both are greater than 0.

[0039] The acquisition logic of the special point deviation coefficient is: obtain the special point of the real estate, determine the coordinate position of the special point of the real estate in the remote sensing image, and mark the coordinate position of the special point of the real estate in the remote sensing image as: , where m = 1, 2, 3, ..., M, M is a positive integer, and m is the number of the special point of the real estate; It should be noted that special points can provide precise geometric reference points when matching remote sensing images with actual geographic information. These marked special points are usually the corners, inflection points or other location points with significant features of real estate. They play a vital role in image alignment, geographic registration and spatial analysis. By determining the coordinates of the special points in the remote sensing image and comparing them with the coordinates of the actual special points, it is ensured that the remote sensing image can accurately reflect the actual geographic location.

[0040] Get the actual geographic coordinates of the real estate special point, and mark the actual geographic coordinates of the real estate special point as: ; Calculate the distance error of special points of real estate, the calculation formula is: ;in, is the distance error of each specific point of the real estate; Set the distance error threshold and mark the distance error threshold as: , compare the distance error of each special point of the real estate with the distance error threshold, obtain the special points whose distance errors are greater than the distance error threshold, and re-mark the distance errors of the special points greater than the distance error threshold as: , where j=1, 2, 3, ..., J, J is a positive integer, and j is the number of the special point that is greater than the distance error threshold; Calculate the special point deviation coefficient, the calculation formula is: ;in, is the special point deviation coefficient.

[0041] It can be seen from the formula that the larger the special point deviation coefficient is, the greater the difference between the special points marked in the remote sensing image and the actual special points in the cadastral information is, indicating that there are large errors in the surveying and mapping of the real estate, including errors in the image processing algorithm's recognition of the special points, or interference from environmental noise (such as shadows, lighting changes, etc.) during marking.

[0042] The logic for obtaining the shape fitting difference coefficient is as follows: extracting the shape outline of the real estate from the remote sensing image, obtaining the boundary point set of the remote sensing image shape, obtaining the boundary point set of the actual cadastral shape, and expressing the boundary point set of the remote sensing image shape and the boundary point set of the actual cadastral shape with complex numbers, wherein the complex form of the boundary point set of the remote sensing image shape is: , the plural form of the boundary points of the actual cadastral shape is: ,k=1, 2, 3, …, K, K is a positive integer, k is the number of the sampling point of the real estate boundary, and i is an imaginary unit; Perform discrete Fourier transform on the complex form, calculate the Fourier descriptor, and mark the Fourier descriptor of the remote sensing image shape as: ,in, is the Fourier descriptor of the remote sensing image appearance, that is, the Fourier transform result corresponding to frequency c in the frequency domain, is a complex index, indicating the rotation factor corresponding to each frequency component, c=1, 2, 3, ..., K, c is the index of the corresponding frequency in the frequency domain, and the actual cadastral shape Fourier descriptor is marked as: ,in, is the Fourier descriptor of the actual cadastral shape; According to the difference of Fourier components, the shape fitting difference coefficient is calculated, and the calculation formula is: ;in, is the shape fitting difference coefficient, h is the number of selected low-frequency components, usually h< <K。

[0043] It can be seen from the formula that the larger the shape fitting difference coefficient is, the greater the difference in the corresponding frequency components between the contour of the remote sensing image and the contour of the actual cadastral shape is, which means that there are significant differences in the shape, structure or details of the two contours.

[0044] It should be noted that the greater the difference in Fourier components, the greater the shape fitting difference coefficient, which means that the general shapes or main structures of the two contours are significantly different. The low-frequency component captures the general outline of the shape. Therefore, when the difference in low-frequency components is large, the overall shape difference of the two shapes is also large. The difference in Fourier components can remove detail noise and focus on the global features of the shape. For complex or irregular shapes, the Fourier descriptor can provide a more concise representation.

[0045] The two-layer optimization strategy usually consists of an outer optimization problem and an inner optimization problem. The outer optimization problem focuses on the global goal, while the inner optimization problem focuses on the local constraints. The use of quality constraint coefficients can ensure that in the optimization process, not only the global goal is paid attention to, but also the satisfaction of local constraints is considered. This strategy can ensure that the optimization results are globally optimal while meeting specific quality requirements. The quality constraint coefficient can limit unacceptable results in the optimization process and ensure the quality of the solution. For example, in real estate area surveying, the accuracy requirement cannot be ignored. The quality constraint coefficient can ensure that the measurement accuracy of each real estate area meets the set standards. By introducing this coefficient in the two-layer optimization, it can avoid generating invalid or substandard flight paths and avoid infeasible solutions caused by ignoring local quality requirements.

[0046] The present invention uses a drone to shoot real estate surveying and mapping areas to obtain real estate image data under different flight paths and postures, provide basic data for subsequent path optimization and quality evaluation, design a fitness function, and combine the flight information and remote sensing image data of the drone flight path to optimize the flight path. The genetic algorithm optimizes the path through continuous evolution to improve the global performance of the path. By analyzing the error information and shape information of each real estate, the quality constraint coefficients of different real estate in different flight paths are calculated, and a quality constraint coefficient threshold is set to ensure that the flight path can meet the accuracy requirements. In combination with the fitness function and the quality constraint coefficient of the genetic algorithm, a double-layer optimization strategy is used to determine the optimal flight path. In the optimization process, both the global goal is considered and the quality constraint of each real estate area is ensured to be met. The present invention provides an efficient and accurate drone flight path planning scheme, which is conducive to the accuracy and efficiency of surveying and mapping.

[0047] Embodiment 2, as Figure 2The structural diagram of the real estate surveying and mapping area management system based on remote sensing images of the present invention is given, which specifically includes a data acquisition module, a path evaluation module, a quality constraint module, and a genetic algorithm module, and the signal connections between the modules; Data acquisition module, used to capture remote sensing images at multiple viewing angles and postures, ensuring coverage of different areas and providing multi-angle image information for subsequent path optimization and quality assessment; The path evaluation module is used to design the genetic algorithm fitness function of the UAV flight path and evaluate the flight path by combining flight information and remote sensing image information; The quality constraint module is used to obtain the quality constraint coefficient of each real estate under the flight path and determine whether the path meets the set quality threshold; The genetic algorithm module is used to optimize the flight path using the genetic algorithm, combining the fitness function of the path evaluation module and the constraints of the quality constraint module to ensure that the final path meets the optimal standard.

[0048] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0049] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination. It should be understood that in various embodiments of the present application, the sequence numbers of the above processes do not mean the order of execution, and the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0050] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A real estate surveying and mapping area management method based on remote sensing images, characterized in that: The specific steps include: S1: Use drones to photograph the real estate surveying area and obtain image data of the real estate under different flight paths and postures; S2: Design the fitness function of the UAV flight path genetic algorithm, optimize the UAV flight path, and combine the flight information and remote sensing image information of the UAV in different flight paths for optimization; S3: Obtain the mass constraint coefficients of different real estates in the UAV flight path through the error information and shape information of each real estate, and set the mass constraint coefficient threshold; S4: Combining the fitness function and mass constraint coefficient of the genetic algorithm, the optimal flight path of the UAV in the real estate mapping area is determined based on a two-level optimization strategy.

2. The real estate surveying and mapping area management method based on remote sensing images according to claim 1 is characterized in that: Flight information, including: The flight information of the UAV on different flight paths is represented by flight time and area coverage coefficient; Among them, the flight time is determined by the historical records of the UAV in different flight paths, which represents the total time the UAV completes the mapping in the flight path; The logic for obtaining the area coverage coefficient is as follows: obtaining the remote sensing image collected by the drone according to the flight path of the drone, dividing the real estate area and the non-real estate area in the remote sensing image collected by the drone, obtaining the number of pixels of the real estate area and the non-real estate area in the remote sensing image, and marking the number of pixels of the real estate area and the non-real estate area in the remote sensing image as follows: and ; According to the resolution of the remote sensing image, the areas of the real estate area and the non-real estate area in the remote sensing image are determined, and the areas of the real estate area and the non-real estate area in the remote sensing image are marked as follows: and ,in, , , is the area represented by each pixel; Determine the total area of ​​each property in the property area based on cadastral information, and mark the total area of ​​each property in the property area as: ; Calculate the coverage deviation of the flight path, the calculation formula is: ;in, is the coverage deviation of the flight path; Calculate the false coverage rate of the flight path, the calculation formula is: ;in, is the false coverage rate of the flight path; Calculate the area coverage coefficient using the following formula: ;in, is the area coverage factor.

3. The real estate surveying and mapping area management method based on remote sensing images according to claim 2 is characterized in that: Remote sensing image information, including: Remote sensing image information is represented by the clarity anomaly variation coefficient; The acquisition logic of the definition abnormal variation coefficient is as follows: according to the shooting process of the UAV in the flight path, a time interval T is set, and images shot by the UAV are collected and analyzed every time interval T in the shooting process in the flight path, and the Laplace operator is applied to the images shot by the UAV every time interval T to calculate the image definition, and the definition of the images shot by the UAV every time interval T is marked as: , where n=1, 2, 3, ..., N, N is a positive integer, and n is the number of the image collected by the drone at every time interval T during the shooting process in the flight path; Calculate the mean and standard deviation of the clarity of the images taken by the drone in the flight path, and mark the mean and standard deviation of the clarity of the images taken by the drone in the flight path as: and ;in, , ; Calculate the coefficient of variation of the clarity of the images taken by the drone in the flight path. The calculation formula is: ;in, The coefficient of variation of the clarity of the images taken by the drone in the flight path; Calculate the clarity abnormal variation coefficient, the calculation formula is: ;in, is the clarity abnormal variation coefficient.

4. The real estate surveying and mapping area management method based on remote sensing images according to claim 3 is characterized in that: Design the fitness function of the UAV flight path genetic algorithm, including: Through comprehensive analysis of the flight information and remote sensing image information of the UAV on different flight paths, the flight time, area coverage coefficient and clarity abnormal change coefficient are weighted and calculated to generate a path evaluation model and a path evaluation coefficient. The calculation formula of the path evaluation coefficient is: ;in, is the path evaluation coefficient of the UAV flight path, SJ is the flight time of the UAV flight path, They are the proportional coefficients of flight time, area coverage coefficient, and clarity abnormal variation coefficient, They are both greater than 0; The path evaluation coefficient is used as the fitness function of the genetic algorithm.

5. The real estate surveying and mapping area management method based on remote sensing images according to claim 4 is characterized in that: Obtain mass constraint coefficients for different real estate in the UAV flight path, including: The quality of each UAV flight path is evaluated by maximizing the fitness function. For each flight path, the quality constraint coefficient of each real estate of each flight path is checked to ensure that each real estate meets the requirements of the predetermined quality constraint coefficient. Through the removal mechanism, the flight path is removed from the population to avoid entering the next generation. The quality constraint coefficient requirement is set by setting the quality constraint coefficient threshold, comparing the quality constraint coefficient of each real estate area with the quality constraint coefficient threshold. If there is real estate with a value greater than the quality constraint coefficient threshold in the flight path, the removal mechanism is triggered. If there is no real estate with a value greater than the quality constraint coefficient threshold in the flight path, the best performing individual is selected from the population as the parent according to the value of the fitness function. The parent individuals will perform crossover and mutation operations according to their fitness, and generate offspring by exchanging part of the path information of the parent individuals. The crossover operation generates new individuals and finds a better solution in different flight paths. When crossing, it is ensured that the generated offspring path meets the requirements of the quality constraint coefficient. Recalculate the fitness of the offspring paths after crossover and mutation, and check whether the quality constraint coefficient of the real estate in each flight path is greater than the quality constraint coefficient threshold. Select the individuals with the highest fitness to enter the next generation population, reach the preset algebraic upper limit or the fitness function converges, and output the optimal flight path as the final solution.

6. The method for real estate surveying and mapping area management based on remote sensing images according to claim 5 is characterized in that: Quality constraint factors, including: The error information and shape information of each real estate in the remote sensing image are collected, and the error information of each real estate is represented by the special point deviation coefficient, and the shape information of each real estate is represented by the shape fitting difference coefficient. The quality constraint coefficient is generated by weighted calculation of the special point deviation coefficient and the shape fitting difference coefficient. The calculation formula of the quality constraint coefficient is: ;in, is the mass constraint coefficient, is the special point deviation coefficient, is the shape fitting difference coefficient, , is the proportional coefficient of the special point deviation coefficient and the shape fitting difference coefficient, , Both are greater than 0.

7. The real estate surveying and mapping area management method based on remote sensing images according to claim 6 is characterized in that: The acquisition logic of the special point deviation coefficient is: Obtain the special points of the real estate, determine the coordinate positions of the special points of the real estate in the remote sensing image, and mark the coordinate positions of the special points of the real estate in the remote sensing image as: , where m = 1, 2, 3, ..., M, M is a positive integer, and m is the number of the special point of the real estate; Get the actual geographic coordinates of the real estate special point, and mark the actual geographic coordinates of the real estate special point as: ; Calculate the distance error of special points of real estate, the calculation formula is: ;in, is the distance error of each specific point of the real estate; Set the distance error threshold and mark the distance error threshold as: , compare the distance error of each special point of the real estate with the distance error threshold, obtain the special points whose distance errors are greater than the distance error threshold, and re-mark the distance errors of the special points greater than the distance error threshold as: , where j=1, 2, 3, ..., J, J is a positive integer, and j is the number of the special point that is greater than the distance error threshold; Calculate the special point deviation coefficient, the calculation formula is: .

8. The method for real estate surveying and mapping area management based on remote sensing images according to claim 7 is characterized in that: The acquisition logic of the shape fitting difference coefficient is: Extract the contour of the real estate from the remote sensing image, obtain the boundary point set of the remote sensing image contour, obtain the boundary point set of the actual cadastral contour, and express the boundary point set of the remote sensing image contour and the boundary point set of the actual cadastral contour with complex numbers, where the complex form of the boundary point set of the remote sensing image contour is: , the plural form of the boundary points of the actual cadastral shape is: ,k=1, 2, 3, …, K, K is a positive integer, k is the number of the sampling point of the real estate boundary, and i is an imaginary unit; Perform discrete Fourier transform on the complex form, calculate the Fourier descriptor, and mark the Fourier descriptor of the remote sensing image shape as: ,in, is the Fourier descriptor of the remote sensing image appearance, that is, the Fourier transform result corresponding to frequency c in the frequency domain, is a complex index, indicating the rotation factor corresponding to each frequency component, c=1, 2, 3, ..., K, c is the index of the corresponding frequency in the frequency domain, and the actual cadastral shape Fourier descriptor is marked as: ,in, is the Fourier descriptor of the actual cadastral shape; According to the difference of Fourier components, the shape fitting difference coefficient is calculated, and the calculation formula is: ; Where h is the number of low-frequency components selected.

9. A real estate surveying and mapping area management system based on remote sensing images, used to implement the real estate surveying and mapping area management method based on remote sensing images as described in any one of claims 1 to 8, characterized in that: It includes a data acquisition module, a path evaluation module, a quality constraint module, and a genetic algorithm module, and signal connections between the modules; Data acquisition module, used to capture remote sensing images at multiple viewing angles and postures, ensuring coverage of different areas and providing multi-angle image information for subsequent path optimization and quality assessment; The path evaluation module is used to design the genetic algorithm fitness function of the UAV flight path and evaluate the flight path by combining flight information and remote sensing image information; The quality constraint module is used to obtain the quality constraint coefficient of each real estate under the flight path and determine whether the path meets the set quality threshold; The genetic algorithm module is used to optimize the flight path using the genetic algorithm, combining the fitness function of the path evaluation module and the constraints of the quality constraint module to ensure that the final path meets the optimal standard.

Images