Target object division method based on three-dimensional model

Abstract

The invention discloses a target object division method based on a three-dimensional model. The method comprises the following steps: acquiring a target image corresponding to a three-dimensional model; wherein the target image is a two-dimensional image comprising at least one target object; determining vertex information of the target object; according to the vertex information of the target object, obtaining position information corresponding to a vertex of the target object in a three-dimensional model; determining the height of the vertex of the target object according to the position information corresponding to the vertex of the target object, and mapping the position information and the height corresponding to the vertex of the target object into a three-dimensional model; and based on the mapped position information and height corresponding to the vertex of the target object, dividing the three-dimensional model to obtain a division result. According to the technical scheme, the longitude and latitude height data of the target object is automatically divided in the three-dimensional model, so that the efficiency of target object division work is remarkably improved, and the time required by the division work is effectively shortened.

Description

Technical Field

[0001] The present invention relates to the technical field of object segmentation, and in particular to a method for segmenting a target object based on a three-dimensional model. Background Art

[0002] As photovoltaic power plants become increasingly large and scalable, ground-mounted photovoltaic power plants often require the dense and orderly arrangement of hundreds or even thousands of photovoltaic strings. However, limited by existing technology, the division of photovoltaic strings still relies on manual labor, a traditional method that wastes considerable time and is inefficient. Summary of the Invention

[0003] The present invention provides a method for dividing target objects based on a three-dimensional model. By automatically dividing the longitude, latitude and height data of the target object in the three-dimensional model, the efficiency of the target object division work is significantly improved and the time required for the division work is effectively shortened.

[0004] According to one aspect of the present invention, a method for segmenting a target object based on a three-dimensional model is provided, the method comprising:

[0005] Acquire a target image corresponding to the three-dimensional model; wherein the target image is a two-dimensional image containing at least one target object;

[0006] Determining vertex information of the target object;

[0007] According to the vertex information of the target object, obtaining position information corresponding to the vertices of the target object in the three-dimensional model;

[0008] Determining the height of the vertices of the target object according to the position information corresponding to the vertices of the target object, and mapping the position information and the height corresponding to the vertices of the target object to a three-dimensional model;

[0009] Based on the position information and height corresponding to the vertices of the target object after mapping, the three-dimensional model is divided to obtain a division result. The technical solution of the embodiment of the present invention is to obtain a target image corresponding to the three-dimensional model; wherein the target image is a two-dimensional image containing at least one target object; determine the vertex information of the target object; obtain the position information corresponding to the vertices of the target object in the three-dimensional model according to the vertex information of the target object; determine the height of the target object vertex according to the position information corresponding to the target object vertex, and map the position information and height corresponding to the target object vertex to the three-dimensional model; divide the three-dimensional model based on the position information and height corresponding to the mapped target object vertex to obtain a division result. This technical solution significantly improves the efficiency of the target object division work by automatically dividing the longitude, latitude and height data of the target object in the three-dimensional model, and effectively shortens the time required for the division work. It solves the problem that the target object division work in the prior art relies on the manual operation mode, resulting in a lot of time waste and poor efficiency.

[0010] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present invention, nor is it intended to limit the scope of the present invention. Other features of the present invention will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0012] Figure 1 This is a flowchart of a method for dividing a target object based on a three-dimensional model provided in accordance with the first embodiment of the present invention;

[0013] Figure 2 is a schematic diagram of the target image provided in Example 1 of the present application;

[0014] Figure 3 This is a flowchart of the automatic division of target objects provided in Example 1 of the present application;

[0015] Figure 4 A schematic diagram of a target object segmentation process based on a three-dimensional model provided in the second embodiment of the present invention;

[0016] Figure 5 A schematic diagram of another target object segmentation process based on a three-dimensional model provided in the third embodiment of the present invention;

[0017] Figure 6A schematic diagram of another target object segmentation process based on a three-dimensional model provided in the fourth embodiment of the present invention. DETAILED DESCRIPTION

[0018] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described 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 making creative efforts should fall within the scope of protection of the present invention.

[0019] It should be noted that the terms "initial", "target", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the terms used in this way are interchangeable where appropriate so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0020] Example 1

[0021] Figure 1 This is a flow chart of a method for segmenting a target object based on a three-dimensional model according to the first embodiment of the present invention. This embodiment is applicable to the case of automatically segmenting the target object. Figure 1 As shown, the method includes:

[0022] S110. Acquire a target image corresponding to the three-dimensional model; wherein the target image is a two-dimensional image containing at least one target object.

[0023] In this solution, the target image contains at least one target object, which includes photovoltaic strings, building equipment, power devices, energy equipment, etc. Figure 2 is a schematic diagram of the target image provided in Example 1 of the present application, such as Figure 2 As shown, the target image is a two-dimensional image containing at least one photovoltaic string. A photovoltaic string is a minimum unit in a photovoltaic power generation system that connects multiple photovoltaic modules in series to form a desired DC output voltage.

[0024] In this embodiment, the target image can be obtained from the target object database; the target image can also be collected using a universal visual model; and the required target image can also be captured from an initial area presented by a preset target object.

[0025] Optionally, obtaining a target image corresponding to the 3D model includes:

[0026] Acquire an initial region; wherein the initial region is a two-dimensional region containing at least one target object;

[0027] In response to the input operation, determining parameters of the target area; wherein the parameters include the shape, size and position of the target area;

[0028] A target area is generated in the interface where the initial area is located according to the parameters, and the target area is used as a target image.

[0029] In this solution, in the visualization display interface of the target platform, the initial area is presented as a two-dimensional plane, and the initial area includes at least one group of target objects as visualization objects.

[0030] Input operations refer to the process by which a user transmits information and instructions to a computer or other electronic device through various input devices. For example, input operations include mouse clicks, drags, slides, or touch operations.

[0031] In this embodiment, after monitoring the input operation performed by the user, the target platform automatically divides the input operation to obtain the shape, size and position of the target area, and intercepts the target area in the interface where the initial area is located based on the shape, size and position of the target area.

[0032] The target region is a subregion of the initial region, and its coverage is completely contained within the initial region. The target region also contains at least one set of target objects.

[0033] Further, Figure 3 This is a flowchart of the automatic division of target objects provided in the first embodiment of the present application. Figure 3 As shown, the parameters of the target area are determined, the target area is generated in the interface where the initial area is located according to the parameters, and the target area is used as the target image.

[0034] Precisely identify the target image within the initial area, allowing intuitive operation to precisely delineate the desired image, effectively avoiding errors caused by manual cropping. Flexible adjustments to various shapes and sizes are supported, fully adapting to the actual needs of different scenarios. The effects of parameter adjustments can be viewed in real time, allowing for quick and convenient optimization of the target image.

[0035] S120: Determine vertex information of the target object.

[0036] The vertex information may refer to two-dimensional coordinate information of the vertex.

[0037] In this solution, traditional image recognition algorithms can be used to identify the target object in the target image and obtain the coordinates of the target object's vertices. Alternatively, image recognition models can be used to identify the target object in the target image and obtain the coordinates of the target object's vertices. Image recognition algorithms include edge detection algorithms and feature matching algorithms. Image recognition models include neural network models and object detection models.

[0038] Among them, the coordinates of the four vertices of the target object can be represented by (x1, y1), (x2, y2), (x3, y3), and (x4, y4).

[0039] This solution can not only obtain vertex information of the target object, but also partial vertex information of the target object. In particular, for target objects with regular shapes, unknown vertex information can be inferred from other vertex information. For example, if the target object has a regular shape of a rectangle, when the information of three vertices is known, the unknown vertex information can be inferred from the length or width of the rectangle.

[0040] S130 . Obtain position information corresponding to the vertices of the target object in the three-dimensional model according to the vertex information of the target object.

[0041] The location information may refer to latitude and longitude information.

[0042] In this embodiment, the position information of the vertices of the target object can be represented by (latx1, lng y1), (latx2, lng y2), (latx3, lng y3), and (latx4, lng y4).

[0043] In this solution, the coordinate transformation formula can be used to transform the vertex information of the target object to obtain the position information of the vertex of the target object.

[0044] Among them, the distance between the target object vertex and any vertex in the target image can also be calculated, and the position information of the target object vertex can be calculated based on the distance between the target object vertex and any vertex in the target image and the association between the vertex information and the position information in the target image.

[0045] In this embodiment, the vertex information of the target object may be calculated based on a predetermined position information calculation formula to obtain the position information of the vertex of the target object.

[0046] S140 : Determine the height of the target object vertex based on the position information corresponding to the target object vertex, and map the position information and height corresponding to the target object vertex to a three-dimensional model.

[0047] In this solution, the height corresponding to the position information of the vertex of the target object can be obtained from the database; the height corresponding to the position information of the vertex of the target object can also be obtained based on the digital elevation model; and the height corresponding to the position information of the vertex of the target object can also be measured using the global navigation satellite system.

[0048] Furthermore, after obtaining the position information and height data of the vertices of the target object, the target object can be accurately spatially divided from the three-dimensional model based on these spatial coordinate information.

[0049] Specifically, the position information and height data of the target object's vertices are obtained; a three-dimensional coordinate system corresponding to the three-dimensional model is established, where the three-dimensional coordinate system includes the X-axis and Y-axis used to represent the vertex position, and the Z-axis used to represent the vertex height; based on the position information, the vertices are mapped to the corresponding positions in the three-dimensional coordinate system; the height data is assigned to the Z-axis coordinate of the corresponding vertex to complete the position and height mapping of the vertex in the three-dimensional model.

[0050] S150 : Divide the three-dimensional model based on the mapped position information and heights corresponding to the vertices of the target object to obtain a division result.

[0051] In this solution, the distribution range of vertices in three-dimensional space is determined based on position information; the vertices within the distribution range are layered according to height information; the number of divided areas and the shape of the areas are determined according to preset division rules; the layered vertices are combined according to the number of areas and the shape of the areas to obtain the division results. The technical solution of the embodiment of the present invention obtains a target image corresponding to a three-dimensional model; wherein the target image is a two-dimensional image containing at least one target object; vertex information of the target object is determined; according to the vertex information of the target object, the position information corresponding to the vertices of the target object in the three-dimensional model is obtained; based on the position information corresponding to the vertices of the target object, the height of the vertices of the target object is determined, and the position information and height corresponding to the vertices of the target object are mapped to the three-dimensional model; based on the position information and height corresponding to the vertices of the target object after mapping, the three-dimensional model is divided to obtain the division results. By executing this technical solution, the efficiency of the target object division work is significantly improved by automatically dividing the longitude, latitude and height data of the target object in the three-dimensional model, and the time required for the division work is effectively shortened.

[0052] Example 2

[0053] Figure 4This is a schematic diagram of a target object segmentation process based on a three-dimensional model provided by the second embodiment of the present invention. The relationship between this embodiment and the above embodiment is a detailed description of the process of determining the coordinates of the target object vertices. Figure 4 As shown, the method includes:

[0054] S410. Acquire a target image corresponding to the three-dimensional model; wherein the target image is a two-dimensional image containing at least one target object.

[0055] S420: Input the target image into an image recognition model, identify the target object in the target image based on the image recognition model, and obtain vertex information of the target object.

[0056] In this program, if Figure 3 As shown, the bounding box of the target object in the target image is identified based on the image recognition model to obtain the vertex information of the target object. The image recognition model can be a large visual model, a large multimodal model, etc.

[0057] Optionally, the process of determining the image recognition model includes:

[0058] Acquire an image to be trained; wherein the image to be trained is a two-dimensional image containing at least one target object;

[0059] Determine the initial parameters of the image recognition model to be trained;

[0060] Inputting the image to be trained into the image recognition model to be trained, and outputting the predicted coordinates of the vertices of the target object;

[0061] Determining a loss function value based on the predicted coordinates of the target object vertices and the true coordinates of the target object vertices in the to-be-trained image;

[0062] The initial parameters of the image recognition model to be trained are trained according to the loss function value to obtain a trained and updated image recognition model.

[0063] In this embodiment, the image to be trained can be obtained from the target object database; the image to be trained can also be collected using a general visual large model; and the required image to be trained can also be intercepted from the initial area presented by a preset target object.

[0064] Specifically, the initial parameters of the image recognition model to be trained include network structure parameters and training hyperparameters. Network structure parameters include the number of model layers, the number of attention heads, and the embedding dimension; these parameters determine the model's feature extraction capabilities and information exchange efficiency. Training hyperparameters include key indicators such as the learning rate and the number of training rounds, which are used to control the model's training process and convergence speed. By rationally configuring these parameters, the model's performance in coordinate recognition tasks can be effectively optimized.

[0065] In this solution, the image to be trained is input into the image recognition model to be trained, which performs in-depth analysis and processing on the image to be trained, performs prediction operations, and finally outputs the predicted coordinates of the vertices of the target object.

[0066] The loss function is a function used to measure the difference between the model's predictions and the actual results. Loss functions can include mean squared error, mean absolute error, mean absolute percentage error, logarithmic loss function, etc.

[0067] Specifically, during the model training process, the predicted coordinates of the target object vertices and the actual coordinates of the target object vertices in the training image are used to calculate the loss function value. By calculating the loss function value, the difference between the model predicted coordinates and the actual coordinates can be scientifically quantified, and then the prediction performance and generalization ability of the model in the current training state can be comprehensively and accurately evaluated.

[0068] Furthermore, during the parameter update process, the training images need to be trained multiple times. The above process is repeated for each iteration. Through continuous iterative training, the model parameters are gradually adjusted, so that the model's predicted coordinates become closer and closer to the true coordinates, and the loss function value gradually decreases. When the loss function value reaches a predetermined minimum value, or when the change in the loss function value after multiple iterations is no longer significant, the model is considered to have converged. The resulting model is the image recognition model after training and updating.

[0069] By building an image recognition model, it is possible to accurately identify the vertices of target objects and provide highly accurate coordinate information. This model possesses powerful data processing capabilities, enabling rapid analysis and processing of large amounts of image data, and can determine the vertex coordinates of multiple target objects in a remarkably short time. Furthermore, using image recognition technology to determine coordinates eliminates the need for direct contact with the target object, effectively preventing damage to the target object and significantly reducing safety risks to personnel, resulting in extremely high safety and practicality.

[0070] S430: Acquire position information corresponding to the vertices of the target object in the three-dimensional model according to the vertex information of the target object.

[0071] S440 : Determine the height of the target object vertex based on the position information corresponding to the target object vertex, and map the position information and height corresponding to the target object vertex to a three-dimensional model.

[0072] S450: Divide the three-dimensional model based on the mapped position information and heights corresponding to the vertices of the target object to obtain a division result.

[0073] The technical solution of the embodiment of the present invention obtains a target image, identifies the target object in the target image using an image recognition model, and obtains the vertex information of the target object; based on the vertex information of the target object, obtains the position information corresponding to the vertices of the target object in the three-dimensional model; determines the height of the target object vertices based on the position information corresponding to the target object vertices, and maps the position information and height corresponding to the target object vertices to the three-dimensional model; and divides the three-dimensional model based on the mapped position information and height corresponding to the target object vertices to obtain the division results. By implementing this technical solution, the efficiency of the target object division work is significantly improved by automatically dividing the latitude, longitude and height data of the target object in the three-dimensional model, and the time required for the division work is effectively shortened.

[0074] Example 3

[0075] Figure 5 This is a schematic diagram of another target object segmentation process based on a three-dimensional model provided by the third embodiment of the present invention. The relationship between this embodiment and the above embodiment is a detailed description of the process of determining the position information of the target object vertices. Figure 5 As shown, the method includes:

[0076] S510: Acquire a target image corresponding to the three-dimensional model; wherein the target image is a two-dimensional image containing at least one target object.

[0077] S520: Determine vertex information of the target object.

[0078] S530 , determining position information of a target vertex of the target image; wherein the target vertex is any one of the vertices of the target image.

[0079] In this solution, the position information of the target image vertices can be accurately determined based on the geographic space coordinate system. Figure 3 As shown, the position information of the four vertices in the target image is recorded (lat0, lng0), (lat1, lng1), (lat2, lng2), and (lat3, lng3).

[0080] Furthermore, a vertex of the target image may be used as the target vertex, and the coordinates of the target vertex may be set to (0, 0) and the position information of the target vertex may be set to (lat0, lng0).

[0081] Optionally, determining the position information of the target vertex of the target image includes:

[0082] Determining geographic reference information related to the target image; wherein the geographic reference information is used to establish a mapping relationship between the image coordinate system and the geographic coordinate system;

[0083] Based on the geographic reference information, a mapping model between the image coordinate system and the geographic coordinate system is constructed;

[0084] The coordinates of the target vertex of the target image are input into the mapping model, and the position information of the target vertex of the target image is calculated.

[0085] In this solution, geographic reference information is used to establish a mapping between the image coordinate system and the geographic coordinate system. This information includes the acquisition device's location information, attitude information, and image scale information. The location information is the acquisition device's position coordinates in the geographic coordinate system, while the attitude information includes the acquisition device's pitch, yaw, and roll angles.

[0086] In this embodiment, a coordinate conversion formula from the image coordinate system to the geographic coordinate system can be established based on the position information, posture information of the acquisition device and the scale information of the image, and the coordinate conversion formula can be calibrated to obtain a mapping model.

[0087] Furthermore, the coordinates of the target vertices of the target image are input into the mapping model, and the mapping model calculates the coordinates of the target vertices of the target image to obtain the position information of the target vertices of the target image.

[0088] By constructing a mapping model, the mapping model can perform high-precision analysis and processing on the target vertices in the target image, and by combining geographic information data and image features, it can accurately determine the location information of the target vertices.

[0089] S540: Determine the height of the target object vertex based on the position information corresponding to the target object vertex, and map the position information and height corresponding to the target object vertex to the three-dimensional model. In this solution, the distance between the target object vertex and the target vertex in the target image can be calculated, and the position information of the target object vertex can be calculated based on the distance between the target object vertex and the target vertex in the target image and the correlation between the coordinates and position information of the target vertex in the target image.

[0090] In this embodiment, the coordinates of the vertices of the target object may be calculated based on a predetermined position information calculation formula according to the position information of the target vertices of the target image to obtain the position information of the vertices of the target object.

[0091] Optionally, calculating the coordinates of the vertices of the target object based on the position information of the target vertices of the target image to obtain the position information of the vertices of the target object includes:

[0092] Determining target parameters; wherein the target parameters include radius, circumferential angle and circumferential arc;

[0093] The ordinate of the target object vertex, the latitude of the target vertex of the target image and the target parameter are weightedly combined to calculate the latitude of the target object vertex; and the abscissa of the target object vertex, the longitude of the target vertex of the target image, the target parameter and the cosine value corresponding to the latitude of the target vertex of the target image are weightedly combined to calculate the longitude of the target object vertex.

[0094] Among them, the circumference angle is 360, and the circumference arc is .

[0095] Specifically, the position information of the vertex of the target object can be calculated using the following formula:

[0096] ;

[0097] ;

[0098] in, 、 are the position information of the vertex of the target object, 、 is the position information of the target vertex of the target image, 、 are the coordinates of the vertex of the target object, is the radius.

[0099] In this solution, the latitude of the target vertex, the vertical coordinate of the target object vertex and the preset radius can be weighted and combined to calculate the latitude of the target object vertex. Figure 3 As shown, specifically, Among them, the average radius of the earth rice.

[0100] In this embodiment, the longitude of the target vertex, the horizontal coordinate of the target object vertex and the preset radius can be weighted and combined to calculate the longitude of the target object vertex. Figure 3 As shown, specifically, .

[0101] By calculating the position information of the vertices of the target object, the geographical location of each target object can be accurately determined.

[0102] S550: Determine the height of the target object vertex based on the position information corresponding to the target object vertex, and map the position information and height corresponding to the target object vertex to a three-dimensional model.

[0103] S560: Divide the three-dimensional model based on the mapped position information and heights corresponding to the vertices of the target object to obtain a division result.

[0104] The technical solution of the embodiment of the present invention obtains the target image, identifies the target object in the target image, and obtains the vertex information of the target object; calculates the vertex information of the target object based on the position information of the target vertices in the target image to obtain the position information of the target object vertices; determines the height of the target object vertices based on the position information of the target object vertices, and maps the position information and height corresponding to the target object vertices to the three-dimensional model; and divides the three-dimensional model based on the position information and height corresponding to the mapped target object vertices to obtain the division result. By implementing this technical solution, the efficiency of the target object division work is significantly improved by automatically dividing the latitude, longitude and height data of the target object in the three-dimensional model, and the time required for the division work is effectively shortened.

[0105] Example 4

[0106] Figure 6 This is a schematic diagram of another target object segmentation process based on a three-dimensional model provided by the fourth embodiment of the present invention, which is a supplement to the target object segmentation process between this embodiment and the above embodiment. Figure 6 As shown, the method includes:

[0107] S610: Acquire a target image corresponding to the three-dimensional model; wherein the target image is a two-dimensional image containing at least one target object.

[0108] S620: Determine vertex information of the target object.

[0109] S630: Obtain position information corresponding to the vertices of the target object in the three-dimensional model according to the vertex information of the target object.

[0110] S640: Obtain the corresponding relationship between the position information and height of the vertex of the target object.

[0111] In this embodiment, the corresponding relationship between the position information and the height of the vertices of the target object can be established by measuring the position information and the height of the vertices of the target object.

[0112] S650: Based on the position information of the vertex of the target object, search the corresponding relationship between the position information of the vertex of the target object and the height to obtain the height corresponding to the position information of the vertex of the target object.

[0113] Specifically, based on the position information coordinates of the target object's vertices, the height value corresponding to each vertex is accurately obtained by searching the pre-established position information and height correspondence library. Figure 3 As shown, the latitude and longitude of the vertex of the target object (x1, y1) is (latx1, lngy1, hg1).

[0114] S660. Generate plotting data of the target object in a predetermined target three-dimensional model based on the position information and height of the vertices of the target object; wherein the target three-dimensional model is a real-scene model of the drone operation area.

[0115] In this embodiment, the target three-dimensional model is a real-scene model of the drone operation area. Optionally, the target three-dimensional model is a tiger-viewing three-dimensional model.

[0116] Specifically, such as Figure 3 As shown, in the target three-dimensional model, based on the position information and height data of the target object vertices, target object mapping data that meets industry mapping standards is generated, providing an accurate visualization basis for subsequent inspections.

[0117] Optionally, generating plotting data of the target object in a predetermined target three-dimensional model based on the position information and height of the vertices of the target object includes:

[0118] receiving an editing instruction for the target object; wherein the editing instruction is an instruction for adjusting the plotting data of the target object;

[0119] Parsing the editing instruction to obtain instruction parameters corresponding to the editing instruction;

[0120] The plotting data of the target object is processed according to the instruction parameters to obtain target plotting data of the target object.

[0121] The editing instructions include, but are not limited to, position adjustment instructions, size modification instructions, color change instructions, and display style switching instructions.

[0122] Furthermore, the editing instruction is parsed grammatically to extract instruction parameters in the editing instruction, where the instruction parameters include but are not limited to position adjustment parameters, size modification parameters, color change parameters, and display style switching parameters.

[0123] Specifically, if the instruction parameter is a position adjustment parameter, the coordinate data of the target object drawing is modified according to the position adjustment parameter; if the instruction parameter is a size modification parameter, the size data of the target object drawing is updated according to the size modification parameter; if the instruction parameter is a color change parameter, the color data of the target object drawing is replaced according to the color change parameter; if the instruction parameter is a display style switching parameter, the style data of the target object drawing is adjusted according to the display style switching parameter.

[0124] By optimizing the processing flow of target object plotting data, not only can the accuracy and reliability of the data be significantly improved, but the format and content of the data can also be flexibly adjusted to meet the needs of different application scenarios. The processed data not only meets the needs of various complex analysis and calculations, but also presents it in a more intuitive and aesthetically pleasing form, greatly enhancing the data visualization effect, allowing users to more clearly and accurately understand and analyze the relevant information of the target object.

[0125] The technical solution of the embodiment of the present invention is to obtain a target image corresponding to a three-dimensional model; wherein the target image is a two-dimensional image containing at least one target object; determine the vertex information of the target object; obtain the position information corresponding to the vertices of the target object in the three-dimensional model based on the vertex information of the target object; and determine the height of the target object vertices based on the position information corresponding to the vertices of the target object. Based on the position information and height of the target object vertices, the target object's mapping data is generated in a predetermined target three-dimensional model; wherein the target three-dimensional model is a real-life model of the drone's operating area. By executing this technical solution, the efficiency of the target object division work is significantly improved by automatically dividing the latitude, longitude and height data of the target object in the three-dimensional model, and the time required for the division work is effectively shortened. Based on the position information and height data of the target object vertices, the target object mapping data is generated, providing an accurate visualization basis for subsequent photovoltaic power station inspections.

[0126] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in the present invention can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved. This is not limited herein.

[0127] The above specific embodiments do not limit the scope of protection of the present invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention are intended to be included within the scope of protection of the present invention.

Claims

1. A method for segmenting a target object based on a three-dimensional model, characterized in that: include: Acquire a target image corresponding to the three-dimensional model; wherein the target image is a two-dimensional image containing at least one target object; Determining vertex information of the target object; According to the vertex information of the target object, obtaining position information corresponding to the vertices of the target object in the three-dimensional model; Determining the height of the target object's vertices based on the position information corresponding to the target object's vertices, and mapping the position information and height corresponding to the target object's vertices to a three-dimensional model; The three-dimensional model is divided based on the mapped position information and heights corresponding to the vertices of the target object to obtain a division result.

2. The method according to claim 1, characterized in that Obtain the target image corresponding to the 3D model, including: Acquire an initial region; wherein the initial region is a two-dimensional region containing at least one target object; In response to the input operation, determining parameters of the target area; wherein the parameters include the shape, size and position of the target area; A target area is generated in the interface where the initial area is located according to the parameters, and the target area is used as a target image.

3. The method according to claim 1, characterized in that Determining the vertex information division of the target object includes: The target image is input into an image recognition model, and a target object in the target image is recognized based on the image recognition model to obtain vertex information of the target object.

4. The method according to claim 3, characterized in that The process of determining the image recognition model includes: Acquire an image to be trained; wherein the image to be trained is a two-dimensional image containing at least one target object; Determine the initial parameters of the image recognition model to be trained; Inputting the image to be trained into the image recognition model to be trained, and outputting the predicted coordinates of the vertices of the target object; Determining a loss function value based on the predicted coordinates of the target object vertices and the true coordinates of the target object vertices in the to-be-trained image; The initial parameters of the image recognition model to be trained are trained according to the loss function value to obtain a trained and updated image recognition model.

5. The method according to claim 1, wherein Acquiring position information corresponding to the vertices of the target object in the three-dimensional model according to the vertex information of the target object includes: Determine position information of a target vertex of a target image; wherein the target vertex is any one of the vertices of the target image; The target object vertex information is calculated based on the position information of the target vertex of the target image to obtain the position information of the target object vertex.

6. The method according to claim 5, characterized in that Determine the position information of the target vertex of the target image, including: Determining geographic reference information related to the target image; wherein the geographic reference information is used to establish a mapping relationship between the image coordinate system and the geographic coordinate system; Based on the geographic reference information, a mapping model between the image coordinate system and the geographic coordinate system is constructed; The coordinates of the target vertex of the target image are input into the mapping model, and the position information of the target vertex of the target image is calculated.

7. The method according to claim 5, characterized in that Calculating the target object vertex information based on the target vertex position information of the target image to obtain the target object vertex position information includes: Determining target parameters; wherein the target parameters include radius, circumferential angle and circumferential arc; The ordinate of the target object vertex, the latitude of the target vertex of the target image and the target parameter are weightedly combined to calculate the latitude of the target object vertex; and the abscissa of the target object vertex, the longitude of the target vertex of the target image, the target parameter and the cosine value corresponding to the latitude of the target vertex of the target image are weightedly combined to calculate the longitude of the target object vertex.

8. The method according to claim 1, characterized in that Determining the height of the vertex of the target object according to the position information corresponding to the vertex of the target object includes: Obtain the corresponding relationship between the position information and height of the vertex of the target object; According to the position information of the vertex of the target object, a search is performed from the correspondence between the position information of the vertex of the target object and the height to obtain the height corresponding to the position information of the vertex of the target object.

9. The method according to claim 1, characterized in that After determining the height of the vertex of the target object based on the position information corresponding to the vertex of the target object, the method includes: Based on the position information and height of the vertices of the target object, the plotting data of the target object is generated in a predetermined target three-dimensional model; wherein the target three-dimensional model is a real-scene model of the drone operation area.

10. The method according to claim 9, characterized in that Generating plotting data of the target object in a predetermined target three-dimensional model based on the position information and height of the vertices of the target object, including: receiving an editing instruction for the target object; wherein the editing instruction is an instruction for adjusting the plotting data of the target object; Parsing the editing instruction to obtain instruction parameters corresponding to the editing instruction; The plotting data of the target object is processed according to the instruction parameters to obtain target plotting data of the target object.

Images