A method and system for analyzing the usage status of homesteads

Abstract

The present invention belongs to the technical field of homestead management. A method and system for analyzing the usage status of homesteads are provided. The method includes: photographing a first high-definition image of a first target homestead and a second high-definition image of a similar second target homestead within the analysis area; extracting first building contour features of the first target homestead based on each first high-definition image, and constructing a first three-dimensional structure of the first target homestead accordingly; extracting second building contour features of the second target homestead based on each second high-definition image, inputting the first three-dimensional structure and the second building contour features into a three-dimensional structure restoration model to obtain a predicted second three-dimensional structure; calculating a structure deviation value between the second three-dimensional structure and the first three-dimensional structure, and determining that the first target homestead is in a used state when the structure deviation value is lower than a deviation threshold. The present invention realizes the accurate judgment of the usage status of homesteads through the cooperation of an unmanned aerial vehicle and a ground terminal, without the need to allocate a large number of personnel for on-site visits.

Description

Technical Field

[0001] The present invention relates to the technical field of homestead management, and in particular, to a method and system for analyzing the usage status of homesteads. Background Art

[0002] Analyzing and monitoring the usage status of homesteads is a key content for realizing the effective management and use of homesteads. The determination of the existing usage status of homesteads is basically completed manually, that is, relevant personnel determine whether the homestead is in a used state or an idle state through means such as visits. However, this method is very inefficient and requires a large number of personnel to be deployed, and the economy is poor. The solution of the present invention aims to improve this technical problem. Summary of the Invention

[0003] To this end, the present invention provides a method, system, electronic device, computer storage medium, and computer program product for analyzing the usage status of homesteads to solve the above technical problems.

[0004] The present invention discloses a method for analyzing the usage status of homesteads, and the method includes the following steps:

[0005] Taking a plurality of first high-definition images of a first target homestead in the analysis area by a drone, and a plurality of second high-definition images of a plurality of second target homesteads similar to the first target homestead; wherein, the shooting angles of the first high-definition images are different from each other, and the shooting angles of the second high-definition images are also different from each other;

[0006] The ground terminal extracts the first building contour feature of the first target homestead based on each of the first high-definition images, and constructs the first three-dimensional structure of the first target homestead based on the first building contour feature;

[0007] The ground terminal extracts the second building contour features of the second target homesteads based on each of the second high-definition images, and inputs the first three-dimensional structure and the second building contour features into a three-dimensional structure restoration model based on artificial intelligence to obtain a predicted second three-dimensional structure;

[0008] Calculating the structural deviation value between the second three-dimensional structure and the first three-dimensional structure, and if the structural deviation value is lower than the deviation threshold, determining that the first target homestead is in a used state, otherwise determining that the first target homestead is in an idle state.

[0009] In some embodiments, before taking a plurality of first high-definition images of a first target homestead in the analysis area by the drone, the method further includes:

[0010] The third high-definition image of the analysis area is captured by a drone. The ground terminal determines each homestead area from the third high-definition image and determines the area high-definition image corresponding to each homestead area in the third high-definition image;

[0011] In addition, the ground terminal performs color quantity analysis on the area high-definition images of each homestead area and determines each of the first target homesteads according to the obtained color quantity.

[0012] In some embodiments, the capturing a plurality of first high-definition images of the first target homestead in the analysis area by the drone, and a plurality of second high-definition images of a plurality of second target homesteads adjacent to the first target homestead includes:

[0013] Based on the third high-definition image, the type of buildings in the analysis area is analyzed. The building height is determined according to the type, and a set of shooting angles is determined according to the building height. Specifically, the angular span range corresponding to this set of shooting angles is inversely proportional to the building height;

[0014] The drone is controlled to capture a plurality of the first high-definition images of the first target homestead in the analysis area and a plurality of the second high-definition images of a plurality of second target homesteads adjacent to the first target homestead according to this set of shooting angles.

[0015] In some embodiments, the capturing a plurality of second high-definition images of a plurality of second target homesteads similar to the first target homestead in the analysis area by the drone includes:

[0016] Based on the third high-definition image, the first quantity of other homesteads with a building type similar to that of the first target homestead in the analysis area is analyzed, and the second quantity is determined according to the first quantity;

[0017] The second quantity of the second target homesteads adjacent to the first target homestead is determined in the analysis area.

[0018] In some embodiments, the three-dimensional structure restoration model is constructed based on a large model, and the large model is fine-tuned using small sample training data to obtain the three-dimensional structure restoration model.

[0019] In some embodiments, the calculating the structural deviation value between the second three-dimensional structure and the first three-dimensional structure includes:

[0020] The second three-dimensional structure and the first three-dimensional structure are respectively decomposed into a plurality of second three-dimensional sub-structures and first three-dimensional sub-structures according to building monomers;

[0021] Determine the corresponding second three-dimensional substructure and the first three-dimensional substructure based on the same position principle, and calculate the first structure deviation value of the second three-dimensional substructure and the first three-dimensional substructure;

[0022] Perform a fusion calculation on each of the first structure deviation values to obtain a second structure deviation value.

[0023] The present invention also discloses a homestead use status analysis system, which includes a processing device and a storage device. The processing device retrieves and executes the computer code in the storage device to implement the following method steps:

[0024] Take a number of first high-definition images of a first target homestead in the analysis area through a drone, as well as a number of second high-definition images of a number of second target homesteads similar to the first target homestead; wherein, the shooting angles of each of the first high-definition images are different from each other, and the shooting angles of each of the second high-definition images are also different from each other;

[0025] The ground terminal extracts the first building contour feature of the first target homestead based on each of the first high-definition images, and constructs the first three-dimensional structure of the first target homestead based on the first building contour feature;

[0026] The ground terminal extracts the second building contour feature of each of the second target homesteads based on each of the second high-definition images, and inputs the first three-dimensional structure and the second building contour feature into a three-dimensional structure restoration model based on artificial intelligence to obtain a predicted second three-dimensional structure;

[0027] Calculate the structure deviation value between the second three-dimensional structure and the first three-dimensional structure. If the structure deviation value is lower than the deviation threshold, it is determined that the first target homestead is in a used state; otherwise, it is determined that the first target homestead is in an idle state.

[0028] The present invention also discloses an electronic device, including: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor. The processor executes the computer program to implement the method described in any one of the previous items.

[0029] The present invention also discloses a computer storage medium. The computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement the method described in any one of the previous items.

[0030] The present invention also discloses a computer program product. When the computer program product runs on a terminal, the terminal is caused to execute to implement the method described in any one of the previous items.

[0031] The solution of the present invention uses the cooperation between an unmanned aerial vehicle and a ground terminal to accurately judge the usage status of a homestead, without the need to allocate a large number of manpower for on-site visits, and can significantly improve the analysis efficiency of the usage status of the homestead. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for use in the embodiments. It should be understood that the following drawings only show some embodiments of the present invention, and therefore should not be regarded as limiting the scope. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained based on these drawings.

[0033] Figure 1 FIG. is a schematic diagram of a scenario for detecting the usage status of a homestead using an unmanned aerial vehicle according to an embodiment of the present invention;

[0034] Figure 2 FIG. is a schematic flowchart of a method for analyzing the usage status of a homestead according to an embodiment of the present invention;

[0035] Figure 3 FIG. is a schematic structural diagram of a system for analyzing the usage status of a homestead according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0036] The following specific embodiments illustrate the implementation manners of the present application. Those skilled in the art can easily understand other advantages and effects of the present application from the content disclosed in this specification. Obviously, the described embodiments are part of the embodiments of the present application, rather than all of them. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope protected by the present application.

[0037] In addition, the technical features involved in different implementation manners of the present application described below can be combined with each other as long as they do not conflict with each other.

[0038] As Figure 1 、 Figure 2 shown, an embodiment of the present invention discloses a method for analyzing the usage status of a homestead, and the method includes the following steps:

[0039] Taking a plurality of first high-definition images of a first target homestead in the analysis area and a plurality of second high-definition images of a plurality of second target homesteads similar to the first target homestead by using an unmanned aerial vehicle; wherein, the shooting angles of the first high-definition images are different from each other, and the shooting angles of the second high-definition images are also different from each other;

[0040] The ground terminal extracts the first building contour feature of the first target homestead based on each of the first high-definition images, and constructs the first three-dimensional structure of the first target homestead based on the first building contour feature;

[0041] The ground terminal extracts the second building contour features of the second target homesteads based on each of the second high-definition images, and inputs the first three-dimensional structure and the second building contour features into a three-dimensional structure restoration model based on artificial intelligence to obtain a predicted second three-dimensional structure;

[0042] Calculate the structural deviation value between the second three-dimensional structure and the first three-dimensional structure. If the structural deviation value is lower than the deviation threshold, it is determined that the first target homestead is in use; otherwise, it is determined that the first target homestead is idle.

[0043] Compared with the traditional method of determining by manual visits, the present invention uses a drone to capture high-definition images of the target homesteads that need to be analyzed for their usage status, and analyzes their usage status based on the characteristics of the homestead buildings extracted from the high-definition images. Specifically, the drone captures a number of first high-definition images of the first target homestead in the analysis area, as well as a number of second high-definition images of a number of second target homesteads similar to the first target homestead. The first high-definition images and the second high-definition images are all multiple images captured from different shooting angles, which is conducive to generating an accurate three-dimensional structure of the homestead later. The ground terminal, located on the ground and controlled by relevant personnel, receives the first high-definition images and the second high-definition images, and extracts the first building contour features of the buildings (houses, walls, etc.) within the first target homestead from each of the first high-definition images. Based on this, the first three-dimensional structure of the first target homestead can be constructed, and this first three-dimensional structure reflects the true three-dimensional structure state of each building within the first target homestead. At the same time, the contour features of the buildings of each of the second target homesteads are also extracted from each of the second high-definition images, and all the contour features are integrated into the second building contour features; the first three-dimensional structure and the second building contour features are input into a three-dimensional structure restoration model based on artificial intelligence together, and the three-dimensional structure restoration model predicts the second three-dimensional structure of the buildings within the first target homestead under normal use conditions under the guidance of the second building contour features. This second three-dimensional structure reflects the three-dimensional structure state that the first target homestead should have under normal use and maintenance. Finally, by comparing the structural deviation value between the real first three-dimensional structure and the predicted second three-dimensional structure, the judgment of whether the first target homestead is in use or idle can be realized.

[0044] Thus, the solution of the present invention uses the cooperation between the drone and the ground terminal to accurately judge the usage status of the homestead, without the need to deploy a large number of personnel for on-site visits, which can significantly improve the analysis efficiency of the homestead usage status.

[0045] It should be noted that the present invention does not specifically limit the selection of the drone, the configuration method of the camera, as well as the communication and control methods between the drone and the ground terminal.

[0046] In some embodiments, before taking a plurality of first high-definition images of the first target homestead in the analysis area by the drone, the method further includes:

[0047] Taking a third high-definition image of the analysis area by the drone, the ground terminal determines each homestead area from the third high-definition image, and determines a corresponding area high-definition image for each homestead area in the third high-definition image;

[0048] Moreover, the ground terminal performs a color quantity analysis on the area high-definition images of each homestead area, and determines each of the first target homesteads according to the obtained color quantity.

[0049] In the embodiment of the present invention, when the analyst arrives at the analysis area, the drone is launched. The drone first takes a third high-definition image (such as a panoramic image) of the entire analysis area and transmits it to the ground terminal. The ground terminal first determines each homestead area from it, and determines a corresponding area high-definition image for each homestead area in the third high-definition image. Then, since when the homestead is in use, there will be objects such as drying clothes, grains, parked cars, etc. in it, and the colors of these objects are usually different and relatively bright compared to the building itself. Therefore, the color quantity of each area high-definition image is analyzed, and based on the analyzed color quantity, the first target homestead is determined from each homestead area. Specifically, when the color quantity is lower than the color quantity threshold, it can be preliminarily determined that the homestead is in an idle state and used as the first target homestead for subsequent targeted analysis; otherwise, it can be preliminarily determined that the homestead is in use and not used as the first target homestead.

[0050] In some embodiments, taking a plurality of first high-definition images of the first target homestead in the analysis area by the drone, and a plurality of second high-definition images of a plurality of second target homesteads adjacent to the first target homestead, includes:

[0051] Analyzing the type of buildings in the analysis area based on the third high-definition image, determining the building height according to the type, and determining a set of shooting angles according to the building height. Specifically, the angular span range corresponding to this set of shooting angles is inversely proportional to the building height;

[0052] Control the drone to take a number of the first high-definition images of the first target homestead in the analysis area according to this set of shooting angles, as well as a number of the second high-definition images of a number of second target homesteads adjacent to the first target homestead.

[0053] In the embodiments of the present invention, by taking images of the target homestead at different angles, it is beneficial to accurately analyze and obtain the three-dimensional structure data of the buildings in the target homestead, and further beneficial to accurately construct the first three-dimensional structures of the buildings in the first target homestead subsequently. At the same time, the determination of the shooting angles should also be appropriate, neither too many nor too few.

[0054] In response to the above requirements, the present invention first analyzes the types of buildings in the analysis area based on the third high-definition images taken previously, and then determines the building height according to the building types. For example, if the type of the analyzed building is a bungalow, the building height is determined to be 5 meters; if the type of the analyzed building is a building, the building height is determined to be 10 meters, etc. Specifically, it can be determined through a building type-height comparison table. Furthermore, according to the building height determined previously, a set of shooting angles is determined. Specifically, the angular span range corresponding to this set of shooting angles is inversely proportional to the building height, that is, when the height of the building in this analysis area is higher, it is easier to capture the three-dimensional structure data of the building, and at this time, the difference between the maximum angle value and the minimum angle value of this set of shooting angles is set to be smaller; while when the height of the building in this analysis area is lower, due to the insufficient building height, it is more difficult to capture the clear three-dimensional structure data of the building, and at this time, the difference between the maximum angle value and the minimum angle value of this set of shooting angles is set to be larger, so as to overcome the calculation deviation caused by visual errors (generally lower than the actual height of the building).

[0055] It should be noted that the number of angles in this set of shooting angles can also be adjusted correspondingly at the same time, that is, the number of angles corresponding to this set of shooting angles is inversely proportional to the building height. Such a setting is also beneficial to obtaining a more accurate three-dimensional structure of the building.

[0056] In some embodiments, taking a number of second high-definition images of a number of second target homesteads similar to the first target homestead in the analysis area by the drone includes:

[0057] Analyze the first quantity of other homesteads with building types similar to those of the first target homestead in the analysis area based on the third high-definition images, and determine the second quantity according to this first quantity;

[0058] Determine the second quantity of the second target homesteads adjacent to the first target homestead within the analysis area.

[0059] In an embodiment of the present invention, the second building contour features obtained from the second target homesteads can be used to assist the three-dimensional structure restoration model in "restoring" the first three-dimensional structure to the three-dimensional structure state that the first target homestead should have under normal use and maintenance. To ensure the accuracy of the second three-dimensional structure predicted by the three-dimensional structure restoration model, the second building contour features need to include the contour features of an appropriate number of second target homesteads. For this, the present invention adopts the following solution:

[0060] Continue to use the above-mentioned third high-definition image, from which the first quantity of other homesteads similar to the building type of the first target homestead can be analyzed. For example, if the building of the first target homestead is a bungalow, then other homesteads belonging to bungalows are determined from the third high-definition image, and then the second quantity is determined according to the first quantity of these other homesteads. Subsequently, screen out the second quantity of second target homesteads with the above-mentioned similar characteristics to the first target homestead around the first target homestead.

[0061] Thus, by analyzing the quantity of other homesteads with the same type of buildings within the analysis area, the quantity of the second target homesteads to be screened out for generating the second building contour features is determined, thereby ensuring that the second building contour features include the contour features of an appropriate number of second target homesteads, which can not only ensure the accuracy of the prediction results of the three-dimensional structure restoration model, but also reduce the analysis and calculation load of the three-dimensional structure restoration model and improve the prediction efficiency.

[0062] It should be noted that the second quantity can be a value obtained by multiplying the first quantity by a conversion coefficient, and the conversion coefficient should be a value less than 1.

[0063] In some embodiments, the three-dimensional structure restoration model is constructed based on a large model, and the large model is fine-tuned using small sample training data to obtain the three-dimensional structure restoration model.

[0064] In the embodiments of the present invention, with the progress of technology, large models such as ChatGPT, KIMI, and Tongyi Qianwen have developed rapidly. These large models possess more powerful "intelligent" analysis capabilities and have currently made remarkable progress in many aspects such as text understanding, translation, and writing. The present invention constructs the above three-dimensional structure restoration model based on these large models. Specifically, small-sample training data is pre-obtained. The small-sample training data contains a certain amount of training data, and this training data includes various structural form data of various types of buildings. Using this small-sample training data to fine-tune the existing large model, a three-dimensional structure restoration model applicable to the restoration of the three-dimensional structure of buildings in the present invention can be obtained.

[0065] Among them, the small sample in the present invention is relative to the training samples of the vertical model. The vertical model needs to be built using specific algorithms and trained with a large amount of training data. However, the present invention uses a general large model. The large model does not need to be built, and it has been fully trained during its use. At this time, only a small amount of training data, that is, small samples, is needed to train its prediction and analysis capabilities in a certain aspect.

[0066] In some embodiments, calculating the structural deviation value between the second three-dimensional structure and the first three-dimensional structure includes:

[0067] Decomposing the second three-dimensional structure and the first three-dimensional structure into multiple second three-dimensional sub-structures and first three-dimensional sub-structures respectively according to building monomers;

[0068] Determining the corresponding second three-dimensional sub-structure and the first three-dimensional sub-structure based on the same position principle, and calculating the first structural deviation value between the second three-dimensional sub-structure and the first three-dimensional sub-structure;

[0069] Performing a fusion calculation on each of the first structural deviation values to obtain a second structural deviation value.

[0070] In the embodiments of the present invention, the obtained second three-dimensional structure and first three-dimensional structure contain multiple individual buildings, such as the main house, side house, etc. Therefore, the present invention first splits the individual buildings, that is, obtains multiple second three-dimensional sub-structures and first three-dimensional sub-structures. The second three-dimensional sub-structures and first three-dimensional sub-structures in the same position form a set of control data. Calculate the first structural deviation value between the second three-dimensional sub-structure and the first three-dimensional sub-structure in each set of control data respectively, and then perform a fusion calculation on the first structural deviation values corresponding to all individual buildings, that is, obtain the second structural deviation value that can represent the structural deviation value between the second three-dimensional structure and the first three-dimensional structure.

[0071] The fusion calculation can be a weighted average. Among them, the weight of the main house is higher than that of the wing rooms, and the weight of the wing rooms is higher than that of the roofless houses, etc. The above weight relationship is only for example, and can be specifically determined according to the number of building types included in the homestead, and the weights of various types of buildings are specified in advance.

[0072] As Figure 3 shown, an embodiment of the present invention also discloses a homestead usage status analysis system, which includes a processing device and a storage device. The processing device retrieves and executes the computer code in the storage device to implement the following method steps:

[0073] Taking a number of first high-definition images of a first target homestead in the analysis area through a drone, as well as a number of second high-definition images of a number of second target homesteads similar to the first target homestead; among them, the shooting angles of the first high-definition images are different from each other, and the shooting angles of the second high-definition images are also different from each other;

[0074] The ground terminal extracts the first building contour features of the first target homestead based on the first high-definition images, and constructs the first three-dimensional structure of the first target homestead based on the first building contour features;

[0075] The ground terminal extracts the second building contour features of the second target homesteads based on the second high-definition images, and inputs the first three-dimensional structure and the second building contour features into a three-dimensional structure restoration model based on artificial intelligence to obtain a predicted second three-dimensional structure;

[0076] Calculating the structural deviation value between the second three-dimensional structure and the first three-dimensional structure. If the structural deviation value is lower than the deviation threshold, it is determined that the first target homestead is in a used state, otherwise it is determined that the first target homestead is in an idle state.

[0077] An embodiment of the present invention also discloses an electronic device, including: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor. The processor executes the computer program to implement the method as described in the foregoing embodiment.

[0078] An embodiment of the present invention also discloses a computer storage medium, which stores a computer program, and the computer program is executed by a processor to implement the method as described in the foregoing embodiment.

[0079] An embodiment of the present invention also discloses a computer program product. When the computer program product runs on a terminal, it enables the terminal to execute to implement the method as described in the foregoing embodiment.

[0080] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and combinations of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices produce means for implementing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or means for implementing the functions specified in multiple blocks.

[0081] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory produce a manufactured article including instruction means that implement the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or means for implementing the functions specified in multiple blocks.

[0082] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or means for implementing the functions specified in multiple blocks.

[0083] As described above, it is only the preferred embodiment of the present invention, and is not used to limit the protection scope of the present invention.

Claims

1. A method for analyzing the usage status of homesteads, characterized in that , The method includes the following steps: Taking a number of first high-definition images of a first target homestead in the analysis area by a drone, and a number of second high-definition images of a number of second target homesteads similar to the first target homestead; wherein, the shooting angles of the first high-definition images are different from each other, and the shooting angles of the second high-definition images are also different from each other; The ground terminal extracts the first building contour feature of the first target homestead based on each of the first high-definition images, and constructs the first three-dimensional structure of the first target homestead based on the first building contour feature; The ground terminal extracts the second building contour features of the second target homesteads based on each of the second high-definition images, and inputs the first three-dimensional structure and the second building contour features into a three-dimensional structure restoration model based on artificial intelligence to obtain a predicted second three-dimensional structure; the second three-dimensional structure reflects the three-dimensional structure state that the first target homestead should have under normal use and maintenance; Calculating the structural deviation value between the second three-dimensional structure and the first three-dimensional structure, and if the structural deviation value is lower than the deviation threshold, determining that the first target homestead is in a used state, otherwise determining that the first target homestead is in an idle state; Before taking a number of first high-definition images of the first target homestead in the analysis area by the drone, the method further includes: Taking a third high-definition image of the analysis area by a drone, and the ground terminal determines each homestead area from the third high-definition image, and determines the area high-definition image corresponding to each homestead area in the third high-definition image; And, the ground terminal performs a color quantity analysis on the area high-definition images of each homestead area, and determines each of the first target homesteads according to the obtained color quantity: when the color quantity is lower than the color quantity threshold, it can be preliminarily determined that the homestead is in an idle state, and it is used as the first target homestead; Taking a number of first high-definition images of the first target homestead in the analysis area by a drone, and a number of second high-definition images of a number of second target homesteads adjacent to the first target homestead, includes: Analyzing the type of buildings in the analysis area based on the third high-definition image, determining the building height according to the type, and determining a set of shooting angles according to the building height. Specifically, the angular span range corresponding to this set of shooting angles is inversely proportional to the building height; Controlling the drone to take a number of the first high-definition images of the first target homestead in the analysis area and a number of the second high-definition images of a number of second target homesteads adjacent to the first target homestead according to this set of shooting angles.

2. The method for analyzing the usage status of a homestead according to claim 1, wherein: Taking a number of second high-definition images of a number of second target homesteads similar to the first target homestead in the analysis area by a drone, includes: Analyzing the first quantity of other homesteads similar to the building type of the first target homestead in the analysis area based on the third high-definition image, and determining the second quantity according to the first quantity; Determine the second quantity of the second target homesteads adjacent to the first target homestead within the analysis area.

3. The method for analyzing the usage status of a homestead according to claim 2, wherein: The three-dimensional structure restoration model is constructed based on a large model, and the large model is fine-tuned using small-sample training data to obtain the three-dimensional structure restoration model.

4. The method for analyzing the usage status of a homestead according to claim 2, characterized in that: Calculating the structural deviation value between the second three-dimensional structure and the first three-dimensional structure includes: Decomposing the second three-dimensional structure and the first three-dimensional structure into a plurality of second three-dimensional sub-structures and first three-dimensional sub-structures respectively according to building monomers; Determining the corresponding second three-dimensional sub-structure and the first three-dimensional sub-structure based on the same-position principle, and calculating the first structural deviation value of the second three-dimensional sub-structure and the first three-dimensional sub-structure; Performing a fusion calculation on each of the first structural deviation values to obtain a second structural deviation value.

5. A homestead use status analysis system, characterized in that: The system includes a processing device and a storage device. The processing device retrieves and executes computer code in the storage device to implement the following method steps: Taking a plurality of first high-definition images of the first target homestead within the analysis area by a drone, and a plurality of second high-definition images of a plurality of second target homesteads similar to the first target homestead; wherein, the shooting angles of the first high-definition images are different from each other, and the shooting angles of the second high-definition images are also different from each other; The ground terminal extracts the first building contour feature of the first target homestead based on each of the first high-definition images, and constructs the first three-dimensional structure of the first target homestead based on the first building contour feature; The ground terminal extracts the second building contour feature of each of the second target homesteads based on each of the second high-definition images, and inputs the first three-dimensional structure and the second building contour feature into a three-dimensional structure restoration model based on artificial intelligence to obtain a predicted second three-dimensional structure; Calculating the structural deviation value between the second three-dimensional structure and the first three-dimensional structure. If the structural deviation value is lower than the deviation threshold, it is determined that the first target homestead is in use; otherwise, it is determined that the first target homestead is idle; Before taking a plurality of first high-definition images of the first target homestead within the analysis area by the drone, the method further includes: Taking a third high-definition image of the analysis area by a drone. The ground terminal determines each homestead area from the third high-definition image, and determines a regional high-definition image corresponding to each homestead area in the third high-definition image; And, the ground terminal performs a color quantity analysis on the regional high-definition images of each homestead area, and determines each of the first target homesteads according to the obtained color quantity; Taking a plurality of first high-definition images of the first target homestead within the analysis area by a drone, and a plurality of second high-definition images of a plurality of second target homesteads adjacent to the first target homestead, includes: Based on the analysis of the third high-definition image, the type of the building in the analysis area is obtained. According to the type, the building height is determined. According to the building height, a set of shooting angles is determined. Specifically, the angular span range corresponding to this set of shooting angles is inversely proportional to the building height; Control the drone to take a number of the first high-definition images of the first target homestead in the analysis area, and a number of the second high-definition images of a number of second target homesteads adjacent to the first target homestead according to this set of shooting angles.

6. An electronic device, comprising: At least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein: the processor executes the computer program to implement the method according to any one of claims 1-4.

7. A computer storage medium storing a computer program, characterized in that: The computer program is executed by the processor to implement the method according to any one of claims 1-4.

8. A computer program product comprising a computer program stored on a non-transitory computer-readable medium, characterized in that: When the computer program is executed by the processor, it implements the method according to any one of claims 1-4.

Images