A photovoltaic power station construction progress identification method and system

Abstract

The application relates to a photovoltaic power station construction progress identification method and system, wherein the method comprises the following steps: obtaining an unmanned aerial vehicle (UAV) aerial image by UAV aerial photography of a construction area of a photovoltaic power station; obtaining key imaging parameters in embedded metadata of each image; establishing a mapping relationship from a pixel coordinate to a real geographic coordinate based on the image and the key imaging parameters through a camera imaging model; performing spatial clustering and deduplication fusion on key components of the photovoltaic power station in the image by using the real geographic coordinate through the mapping relationship to generate a regional component distribution map; and dynamically evaluating the construction and laying progress of the photovoltaic power station in a preset area based on the regional component distribution map. Through the application, the embedded metadata of the UAV aerial image is ingeniously utilized to perform spatial clustering and deduplication fusion on key photovoltaic components such as photovoltaic panels, pipe piles and supports in the image across images, so that the occurrence of component misidentification and repeated identification is effectively reduced.

Description

Technical Field

[0001] This application relates to the field of new energy engineering monitoring technology, and in particular to a method and system for identifying the construction progress of photovoltaic power plants. Background Technology

[0002] As the scale of photovoltaic power plant construction continues to expand, accurate monitoring of construction progress is crucial to ensuring timely project completion. Traditional photovoltaic installation progress monitoring mainly relies on manual inspections, which is not only inefficient and costly, but also difficult to implement in large areas or complex terrain conditions.

[0003] In recent years, with the development of drone technology and image recognition technology, it has become possible to automatically identify the installation progress of photovoltaic modules using high-resolution images captured by drones. However, existing methods often have the following shortcomings: Low geolocation accuracy: Traditional image recognition methods struggle to achieve pixel-level geographic coordinate transformation, resulting in inaccurate target location identification; Difficulty in deduplicating targets across images: Due to overlaps between adjacent aerial images, the same physical object may be detected multiple times, leading to duplicate counting; Limited automation: The lack of an effective fusion mechanism for multi-source data (such as UAV imagery, metadata, etc.) limits the overall automation level and practicality of the system.

[0004] Currently, no effective solution has been proposed for improving the accuracy of identifying the construction progress of photovoltaic power plants under drone aerial photography. Summary of the Invention

[0005] This application provides a method and system for identifying the construction progress of a photovoltaic power station, which at least solves the problem in the related art of how to improve the accuracy of identifying the construction progress of a photovoltaic power station under drone aerial photography.

[0006] In a first aspect, embodiments of this application provide a method for identifying the construction progress of a photovoltaic power station, the method comprising: Aerial photography of the construction area of ​​the photovoltaic power station was conducted using drones to obtain drone aerial images; The embedded metadata of each drone aerial image is obtained, and key imaging parameters are extracted from the embedded metadata. Based on the aerial images and key imaging parameters captured by the UAV, a mapping relationship between pixel coordinates and real geographic coordinates is established through a camera imaging model. For the key components of the photovoltaic power station in the drone aerial images, the mapping relationship is used to perform spatial clustering and deduplication fusion using real geographic coordinates to generate a regional component distribution map; Based on the regional component distribution map, the construction and installation progress of photovoltaic power stations within the preset area is dynamically evaluated.

[0007] In some embodiments, embedded metadata is acquired for each drone aerial image, and key imaging parameters are extracted from the embedded metadata, including: The embedded metadata of each drone aerial image is obtained, wherein the embedded metadata includes EXIF ​​metadata and XMP metadata; Key imaging parameters for each UAV aerial image are extracted from the embedded metadata. These key imaging parameters include the UAV camera's GNSS position, relative ground altitude, absolute flight altitude, gimbal attitude angle, camera intrinsic parameters, sensor size, and image resolution.

[0008] In some embodiments, the method includes: For drone aerial images lacking embedded metadata, spatiotemporal alignment interpolation is performed to supplement them using synchronously recorded flight logs; For RTK or PPK drones, the high-precision POS position obtained through post-processing is used to replace the GNSS position of the drone camera in the key imaging parameters.

[0009] In some embodiments, establishing a mapping relationship from pixel coordinates to real geographic coordinates based on the UAV aerial images and key imaging parameters through a camera imaging model includes: Based on the UAV aerial images and key imaging parameters, a projection relationship from image pixel coordinates to ground 3D coordinates is constructed through a camera imaging model to obtain the mapping relationship from pixel coordinates to real geographic coordinates, wherein: For flat terrain, the mapping relationship is to use affine transformation or homography matrix to approximately map local areas of the image to UTM or local plane coordinate system; For undulating terrain, the mapping relationship is to first introduce a digital surface model for orthorectification, generate a true orthophoto image, and then perform coordinate mapping.

[0010] In some embodiments, for key components of the photovoltaic power station in the drone aerial image, spatial clustering and deduplication fusion are performed using real geographic coordinates through the mapping relationship to generate a regional component distribution map, including: For each key component of the photovoltaic power station in the drone aerial image, extract the pixel coordinates of the mask centroid or the pixel coordinates of the center point of the minimum bounding rectangle of the key component; The pixel coordinates are converted into real geographic location coordinates through the mapping relationship, and a spatial tolerance threshold is set based on the ground sampling distance GSD. Under the constraint of the spatial tolerance threshold, the real geographic location coordinates are spatially clustered and deduplicated to generate a regional component distribution map covering the entire construction area.

[0011] In some embodiments, dynamically evaluating the construction progress of photovoltaic power stations within a preset area based on a regional component distribution map includes: The construction area of ​​the photovoltaic power station is divided into rule-based management units; Based on the regional component distribution map, the installation completion rate is obtained by calculating the actual number of components and the theoretical maximum number within each rule management unit. Alternatively, based on the regional component distribution map, the ratio of the total area of ​​the component mask within each rule management unit to the unit area can be calculated to obtain the land cover rate. Based on the installation completion rate or surface coverage rate, the construction and installation progress of photovoltaic power stations within the preset area is dynamically evaluated.

[0012] In some embodiments, before generating a regional-level component distribution map by spatially clustering and deduplicating the key components of the photovoltaic power station in the drone aerial image using the mapping relationship and real geographic coordinates, the method includes: Train the object detection model to obtain a trained object detection model; Based on the aerial images captured by the UAV, the key components of the photovoltaic power station in the UAV aerial images are obtained by segmentation and recognition using the trained target detection model.

[0013] In some embodiments, training the object detection model to obtain a trained object detection model includes: A labeled training dataset for photovoltaic construction scenarios is constructed, wherein the labeled training dataset covers sample images of different installation stages of photovoltaic power plants, different illumination angles, shading conditions and background interference. For recognition scenarios requiring high precision, a high-precision target detection model is constructed; for recognition scenarios deployed at the edge, a lightweight target detection model is constructed. Based on the labeled training dataset, the high-precision target detection model or the lightweight target detection model is trained to obtain a trained target detection model.

[0014] In some embodiments, aerial photography of the construction area of ​​the photovoltaic power station is conducted using drones to obtain drone aerial images, including: Based on the vector boundary of the photovoltaic power station to be monitored, a gridded aerial photography route covering the entire area is automatically generated in the ground control software of the UAV. The gridded aerial photography route takes into account environmental factors of image quality, including flight altitude, forward overlap rate, lateral overlap rate, lighting conditions and wind speed. Guided by the gridded aerial photography route, the UAV acquires aerial images and simultaneously records the flight log corresponding to each image. The flight log includes timestamps, IMU attitude, and raw GNSS observations.

[0015] Secondly, embodiments of this application provide a photovoltaic power plant construction progress identification system. The system is used to execute the method described in the first aspect above. The system includes a data acquisition module, an extraction module, a construction module, a generation module, and an evaluation module. The acquisition module is used to take aerial photos of the construction area of ​​the photovoltaic power station using a drone, and obtain drone aerial images. The extraction module is used to acquire embedded metadata for each drone aerial image and extract key imaging parameters from the embedded metadata. The construction module is used to establish a mapping relationship between pixel coordinates and real geographic coordinates based on the UAV aerial images and key imaging parameters through a camera imaging model. The generation module is used to perform spatial clustering and deduplication fusion on the key components of the photovoltaic power station in the drone aerial image using the mapping relationship and real geographic coordinates to generate a regional component distribution map. The evaluation module is used to dynamically evaluate the construction progress of photovoltaic power stations within a preset area based on the regional component distribution map.

[0016] Compared to related technologies, this application provides a method and system for identifying the construction progress of a photovoltaic power station. The method involves using a drone to take aerial photographs of the construction area of ​​the photovoltaic power station, obtaining drone aerial images; acquiring embedded metadata for each drone aerial image, and extracting key imaging parameters from the embedded metadata; establishing a mapping relationship between pixel coordinates and real geographic coordinates based on the drone aerial images and key imaging parameters using a camera imaging model; performing spatial clustering and deduplication fusion on key components of the photovoltaic power station in the drone aerial images using the mapping relationship and real geographic coordinates to generate a regional component distribution map; and dynamically evaluating the construction progress of the photovoltaic power station within a preset area based on the regional component distribution map. This method cleverly utilizes the embedded metadata of drone aerial images to perform cross-image spatial clustering and deduplication fusion on key photovoltaic components such as photovoltaic panels, piles, and supports in the images, effectively reducing the occurrence of misidentification and duplicate identification of components, and solving the problem of how to improve the accuracy of identifying the construction progress of photovoltaic power stations under drone aerial photography. Attached Figure Description

[0017] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 This is a flowchart of the steps of the photovoltaic power station construction progress identification method according to the embodiments of this application; Figure 2 This is a schematic diagram of an aerial photograph taken by a drone according to an embodiment of this application; Figure 3 This is a schematic diagram of the software interface corresponding to the photovoltaic power station construction progress identification method according to the embodiments of this application; Figure 4 This is a schematic diagram of the internal structure of an electronic device according to an embodiment of this application. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this application clearer, the application is described and illustrated below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application. All other embodiments obtained by those skilled in the art based on the embodiments provided in this application without inventive effort are within the scope of protection of this application.

[0019] Obviously, the accompanying drawings described below are merely some examples or embodiments of this application. Those skilled in the art can apply this application to other similar scenarios based on these drawings without any inventive effort. Furthermore, it is understood that although the efforts made in this development process may be complex and lengthy, for those skilled in the art related to the content disclosed in this application, any changes to design, manufacturing, or production based on the technical content disclosed in this application are merely conventional technical means and should not be construed as insufficient disclosure of the content of this application.

[0020] In this application, the reference to "embodiment" means that a specific feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment that is mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described in this application may be combined with other embodiments without conflict.

[0021] Unless otherwise defined, the technical or scientific terms used in this application shall have the ordinary meaning understood by one of ordinary skill in the art to which this application pertains. The terms “a,” “an,” “an,” “the,” and similar words used in this application do not indicate quantity limitation and may indicate singular or plural. The terms “comprising,” “including,” “having,” and any variations thereof used in this application are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may also include steps or units not listed, or may include other steps or units inherent to these processes, methods, products, or devices. The terms “connected,” “linked,” “coupled,” and similar words used in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. “Multiple” used in this application refers to two or more. “And / or” describes the relationship between related objects, indicating that three relationships may exist; for example, “A and / or B” can represent: A alone, A and B simultaneously, and B alone. The character " / " generally indicates that the preceding and following objects are in an "or" relationship. The terms "first," "second," and "third" used in this application are merely to distinguish similar objects and do not represent a specific ordering of the objects.

[0022] This application provides a method for identifying the construction progress of a photovoltaic power station. Figure 1 This is a flowchart illustrating the steps of the photovoltaic power plant construction progress identification method according to an embodiment of this application, as follows: Figure 1 As shown, the method includes the following steps: Step S102: Take aerial photos of the construction area of ​​the photovoltaic power station using a drone to obtain drone aerial images; Step S102 specifically includes the following steps: Step S1021: Based on the vector boundary of the photovoltaic power station to be monitored, a gridded aerial photography route covering the entire area is automatically generated in the UAV's ground control software. The gridded aerial photography route takes into account environmental factors of image quality, including flight altitude, forward overlap rate, lateral overlap rate, lighting conditions, and wind speed. It should be noted that, based on the vector boundary of the photovoltaic power station to be monitored (such as GeoJSON or Shapefile format) or by manually demarcating electronic fences, a gridded aerial flight path covering the entire area is automatically generated in ground control software (such as DJI Pilot, Pix4Dcapture, or a self-developed flight control platform). When planning the flight path, environmental factors such as flight altitude (typically 30–120 meters), forward overlap (recommended ≥80%), lateral overlap (recommended ≥70%), lighting conditions (avoiding strong midday sunlight or low-contrast periods on cloudy days), and wind speed are comprehensively considered to ensure that the image quality meets the requirements for subsequent identification and geolocation accuracy.

[0023] In step S1022, guided by the gridded aerial photography route, the drone acquires aerial images and records the flight log corresponding to each image simultaneously. The flight log includes timestamp, IMU attitude, and raw GNSS observations.

[0024] It should be noted that, Figure 2 This is a schematic diagram of an aerial photograph taken by a drone according to an embodiment of this application, such as... Figure 2 As shown, a high-precision RTK / PPK positioning UAV (such as the DJI M300 RTK) is used to automatically trigger the camera shutter at fixed time intervals (e.g., every 2 seconds) or fixed spatial intervals (e.g., every 10 meters) during flight, acquiring a sequence of orthophotos with high spatial consistency. Simultaneously, a flight log corresponding to each image is recorded, including timestamps, IMU attitude, raw GNSS observations, and other auxiliary data, providing support for subsequent metadata analysis and error correction.

[0025] Step S104: Obtain the embedded metadata of each drone aerial image and extract key imaging parameters from the embedded metadata; Step S104 specifically includes the following steps: Step S1041: Obtain the embedded metadata of each drone aerial image, wherein the embedded metadata includes EXIF ​​metadata and XMP metadata; Step S1042: Extract key imaging parameters for each UAV aerial image from embedded metadata. These key imaging parameters include the UAV camera's GNSS position, relative ground altitude, absolute flight altitude, gimbal attitude angle, camera intrinsic parameters, sensor size, and image resolution.

[0026] It should be noted that parsing the EXIF / XMP metadata of the image extracts key imaging parameters, including: camera position (WGS84 latitude and longitude and ellipsoidal altitude), relative ground altitude (AGL) or absolute flight altitude (MSL), gimbal attitude angles (pitch, yaw, roll), and camera intrinsic parameters (focal length f). x , fy Main point c x , c y The parameters include distortion coefficients k1~k3, p1, p2, sensor size, and image resolution. For drone aerial images with missing or insufficient original metadata (such as those from consumer drones), spatiotemporal alignment interpolation is performed using synchronously recorded flight logs (.csv / .bin files). For RTK or PPK drones, the high-precision POS position (centimeter-level) obtained through post-processing is used to replace the drone camera GNSS position in key imaging parameters, significantly improving geolocation accuracy.

[0027] Step S106: Based on the drone aerial images and key imaging parameters, establish the mapping relationship between pixel coordinates and real geographic coordinates through the camera imaging model; Specifically, step S106 involves constructing a projection relationship from image pixel coordinates to ground 3D coordinates using a camera imaging model, based on UAV aerial images and key imaging parameters, to obtain the mapping relationship from pixel coordinates to real geographic coordinates. It should be noted that, through the camera imaging model (pinhole camera model and collinearity equation), the projection relationship from image pixel coordinates (u, v) to the three-dimensional coordinates of the ground is constructed:

[0028] Where K is the intrinsic parameter matrix, R and t are derived from the gimbal attitude and position, and Z is the ground elevation (which can be assumed to be a constant or derived from the DEM). For flat terrain, the mapping relationship is to use affine transformation or homography to approximately map the local area of ​​the image to the UTM or local plane coordinate system; for undulating terrain, the mapping relationship is to first introduce the Digital Surface Model (DSM) for orthorectification, generate a true orthophoto image (TDOM), and then perform coordinate mapping.

[0029] Step S108: For the key components of the photovoltaic power station in the drone aerial image, spatial clustering and deduplication fusion are performed using real geographic coordinates through mapping relationships to generate a regional component distribution map. Step S108 specifically includes the following steps: Step S1081: Train the target detection model to obtain a trained target detection model; based on the UAV aerial image, perform segmentation and recognition using the trained target detection model to obtain the key components of the photovoltaic power station in the UAV aerial image.

[0030] Specifically, step S1081 involves constructing a labeled training dataset for photovoltaic construction scenarios. This dataset covers sample images from different installation stages of photovoltaic power plants, under varying illumination angles, shading conditions, and background interference. For recognition scenarios requiring high precision, a high-precision target detection model is constructed. For recognition scenarios involving edge deployment, a lightweight target detection model is constructed. Based on the labeled training dataset, the high-precision target detection model or the lightweight target detection model is trained to obtain a well-trained target detection model.

[0031] It should be noted that the AI ​​object detection model is used to automatically identify and segment key components such as photovoltaic panels, piles, and supports in the images. First, a dedicated labeled dataset for photovoltaic construction scenarios is constructed, covering sample images under different installation stages (e.g., pile driving only, support installation, component laying), different lighting angles, shading conditions, and background interference (e.g., bare soil, vegetation, temporary facilities). Second, a two-stage or one-stage deep learning model is used for multi-class instance segmentation: for high-precision requirements, Mask R-CNN or its improved versions (e.g., Cascade Mask R-CNN, HTC) are selected, outputting the bounding box, class label, and pixel-level mask for each target; for edge deployment scenarios with high real-time requirements, lightweight models such as YOLOv8-seg, YOLO-NAS-Seg, or RT-DETR can be used to reduce inference latency while maintaining high mAP. Furthermore, data augmentation strategies (such as Mosaic, CutMix, random rotation, and color perturbation) and class-balanced sampling are introduced during model training to improve generalization ability; COCO or custom evaluation metrics (such as component detection rate, false positive rate, and segmentation accuracy under IoU threshold) are used for validation. In addition, for the problem of small targets (such as pipe pile diameter <10 pixels), feature pyramid networks (FPN / PANet) or multi-scale training strategies can be combined to enhance low-level feature representation.

[0032] Step S1082: For each key component of the photovoltaic power station in the drone aerial image, extract the pixel coordinates of the mask centroid or the pixel coordinates of the center point of the minimum bounding rectangle of the key component. Step S1083: Convert pixel coordinates into real geographic location coordinates through mapping relationship, and set spatial tolerance threshold based on ground sampling distance GSD; Step S1084: Under the constraint of spatial tolerance threshold, perform spatial clustering and deduplication fusion on the real geographic positioning coordinates to generate a regional component distribution map covering the entire construction area.

[0033] It should be noted that step S1082, based on visual detection and mapping transformation, automates GPS location calculation and quantity statistics for photovoltaic modules and pipe piles. For each target instance identified by the AI ​​target detection model, the pixel coordinates (u) of its mask centroid or the center point of its minimum bounding rectangle are extracted. c , v c ); Step S1083 utilizes the pixel-geographic coordinate mapping relationship of the above framework to map (u) c , v c The data is converted to WGS84 latitude and longitude coordinates as the geographic location result of the component / pipe pile; combined with the target category and geometric features (such as the component aspect ratio and area), false detections (such as non-photovoltaic targets such as reflectors and vehicles) are eliminated; the number of each type of target is automatically counted and spatially aggregated according to the map sheet or preset grid (such as 10m×10m), and the output is a vector point layer with attributes (such as GeoJSON format), which includes fields such as ID, category, coordinates, confidence, and detection time.

[0034] It should be further explained that spatial clustering and deduplication fusion are performed on the geographic coordinates of targets identified in multiple aerial images to generate a regional component distribution map. Since overlapping areas exist in adjacent aerial images, the same physical target may be detected multiple times, leading to duplicate counting. Therefore, a deduplication mechanism based on geographic coordinates is introduced: based on the ground sampling distance... (Used to evaluate image spatial resolution and as a basis for judging the reasonableness of target size), set a spatial tolerance threshold (e.g., 0.5–1.0 meters); use DBSCAN, OPTICS, or KD-Tree-based nearest neighbor search algorithm to spatially cluster all detected targets; treat each cluster as a unique entity, retaining the detection result with the highest confidence or the most complete geometric features as the representative; integrate the deduplicated target set into a unified geographic database to generate a heat map or density map of photovoltaic modules / pipe piles covering the entire construction area; also supports the overlay of multiple aerial photography results by time series to achieve change detection and installation trajectory backtracking.

[0035] Step S110: Based on the regional component distribution map, dynamically evaluate the construction and installation progress of the photovoltaic power station within the preset area.

[0036] Step S110 specifically includes the following steps: Step S1101: Divide the construction area of ​​the photovoltaic power station into rule management units; Step S1102: Based on the regional component distribution map, calculate the actual number of components in each rule management unit and the theoretical maximum number to obtain the installation completion rate; Step S1101: Alternatively, based on the regional component distribution map, calculate the ratio of the total area of ​​the component mask in each rule management unit to the unit area to obtain the surface coverage rate. Step S1103: Based on the installation completion rate or surface coverage rate, dynamically assess the construction and installation progress of the photovoltaic power station within the preset area.

[0037] It should be noted that the system dynamically assesses the progress of photovoltaic (PV) installation by statistically analyzing the number of installed modules / coverage within a preset area (e.g., per unit area). The construction area is divided into regular management units (e.g., sub-arrays, array blocks), or the theoretical installation range is extracted from the design CAD drawings as a reference surface. The actual number of modules (N_actual) and the theoretical maximum number (N_design) within each unit are calculated to determine the installation completion rate, or the coverage ratio is calculated based on the ratio of the total area of ​​the module mask to the unit area. A progress-time curve is constructed using timestamp information, supporting weekly / monthly trend analysis. Visual reports are output, including a progress dashboard, highlighted areas of non-installed modules, and alerts for abnormal missing modules (e.g., prolonged lack of progress in a certain area), providing decision-making support for project management. Figure 3 This is a schematic diagram of the software interface corresponding to the photovoltaic power station construction progress identification method according to the embodiments of this application.

[0038] It should be noted that the steps shown in the above process or in the flowchart of the accompanying figures can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0039] This application provides a photovoltaic power plant construction progress identification system. The system is used to execute the method provided in the above embodiments. The system includes a data acquisition module, an extraction module, a construction module, a generation module, and an evaluation module. The data acquisition module is used to take aerial photos of the construction area of ​​the photovoltaic power station using drones, and obtain drone aerial images; The extraction module is used to acquire embedded metadata for each drone aerial image and extract key imaging parameters from the embedded metadata. The module is used to establish a mapping relationship between pixel coordinates and real geographic coordinates based on UAV aerial images and key imaging parameters through a camera imaging model. The generation module is used to perform spatial clustering and deduplication fusion of key components of photovoltaic power plants in drone aerial images using real geographic coordinates through mapping relationships, and generate regional component distribution maps. The evaluation module is used to dynamically evaluate the construction and installation progress of photovoltaic power stations within a preset area based on the regional component distribution map.

[0040] It should be noted that the above modules can be functional modules or program modules, and can be implemented through software or hardware. For modules implemented through hardware, the above modules can reside in the same processor; or the above modules can be located in different processors in any combination.

[0041] This embodiment provides an electronic device including a memory and a processor. The memory stores a computer program, and the processor is configured to run the computer program to perform the steps in any of the above method embodiments.

[0042] Optionally, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor and the input / output device is connected to the processor.

[0043] Optionally, the electronic device may further include a processor, memory, network interface, display screen, and input device connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for identifying the construction progress of a photovoltaic power station. The display screen may be a liquid crystal display (LCD) or an e-ink display. The input device may be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the device's casing, or an external keyboard, touchpad, or mouse.

[0044] It should be noted that the specific examples in this embodiment can refer to the examples described in the above embodiments and optional implementations, and will not be repeated here.

[0045] Furthermore, in conjunction with the photovoltaic power plant construction progress identification method in the above embodiments, this application embodiment can provide a storage medium for implementation. This storage medium stores a computer program; when executed by a processor, the computer program implements any of the photovoltaic power plant construction progress identification methods in the above embodiments.

[0046] In one embodiment, Figure 4 This is a schematic diagram of the internal structure of an electronic device according to an embodiment of this application, such as... Figure 4 As shown, an electronic device is provided, which can be a server, and its internal structure diagram can be as follows. Figure 4As shown, the electronic device includes a processor, a network interface, internal memory, and non-volatile memory connected via an internal bus. The non-volatile memory stores the operating system, computer programs, and a database. The processor provides computing and control capabilities, the network interface communicates with external terminals via a network, the internal memory provides an environment for the operating system and computer programs to run, the computer programs are executed by the processor to implement a method for identifying the construction progress of a photovoltaic power station, and the database stores data.

[0047] Those skilled in the art will understand that Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the electronic device to which the present application is applied. A specific electronic device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0048] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0049] Those skilled in the art should understand that the technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments have been described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0050] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for identifying the construction progress of a photovoltaic power station, characterized in that, The method includes: Aerial photography of the construction area of ​​the photovoltaic power station was conducted using drones to obtain drone aerial images; The embedded metadata of each drone aerial image is obtained, and key imaging parameters are extracted from the embedded metadata. Based on the aerial images and key imaging parameters captured by the UAV, a mapping relationship between pixel coordinates and real geographic coordinates is established through a camera imaging model. For the key components of the photovoltaic power station in the drone aerial images, the mapping relationship is used to perform spatial clustering and deduplication fusion using real geographic coordinates to generate a regional component distribution map; Based on the regional component distribution map, the construction and installation progress of photovoltaic power stations within the preset area is dynamically evaluated.

2. The method according to claim 1, characterized in that, The embedded metadata of each drone aerial image is obtained, and key imaging parameters are extracted from the embedded metadata, including: The embedded metadata of each drone aerial image is obtained, wherein the embedded metadata includes EXIF ​​metadata and XMP metadata; Key imaging parameters for each UAV aerial image are extracted from the embedded metadata. These key imaging parameters include the UAV camera's GNSS position, relative ground altitude, absolute flight altitude, gimbal attitude angle, camera intrinsic parameters, sensor size, and image resolution.

3. The method according to claim 2, characterized in that, The method includes: For drone aerial images lacking embedded metadata, spatiotemporal alignment interpolation is performed to supplement them using synchronously recorded flight logs; For RTK or PPK drones, the high-precision POS position obtained through post-processing is used to replace the GNSS position of the drone camera in the key imaging parameters.

4. The method according to claim 1, characterized in that, Based on the aforementioned UAV aerial images and key imaging parameters, establishing a mapping relationship from pixel coordinates to real geographic coordinates through a camera imaging model includes: Based on the UAV aerial images and key imaging parameters, a projection relationship from image pixel coordinates to ground 3D coordinates is constructed through a camera imaging model to obtain the mapping relationship from pixel coordinates to real geographic coordinates, wherein: For flat terrain, the mapping relationship is to use affine transformation or homography matrix to approximately map local areas of the image to UTM or local plane coordinate system; For undulating terrain, the mapping relationship is to first introduce a digital surface model for orthorectification, generate a true orthophoto image, and then perform coordinate mapping.

5. The method according to claim 1, characterized in that, For the key components of the photovoltaic power station in the drone aerial images, spatial clustering and deduplication fusion are performed using real geographic coordinates through the mapping relationship to generate a regional component distribution map, including: For each key component of the photovoltaic power station in the drone aerial image, extract the pixel coordinates of the mask centroid or the pixel coordinates of the center point of the minimum bounding rectangle of the key component; The pixel coordinates are converted into real geographic location coordinates through the mapping relationship, and a spatial tolerance threshold is set based on the ground sampling distance GSD. Under the constraint of the spatial tolerance threshold, the real geographic location coordinates are spatially clustered and deduplicated to generate a regional component distribution map covering the entire construction area.

6. The method according to claim 1, characterized in that, Based on the regional component distribution map, the dynamic evaluation of the construction and installation progress of photovoltaic power stations within the preset area includes: The construction area of ​​the photovoltaic power station is divided into rule-based management units; Based on the regional component distribution map, the installation completion rate is obtained by calculating the actual number of components and the theoretical maximum number within each rule management unit. Alternatively, based on the regional component distribution map, the ratio of the total area of ​​the component mask within each rule management unit to the unit area can be calculated to obtain the land cover rate. Based on the installation completion rate or surface coverage rate, the construction and installation progress of photovoltaic power stations within the preset area is dynamically evaluated.

7. The method according to claim 1, characterized in that, Before generating a regional-level component distribution map by spatially clustering and deduplicating the key components of the photovoltaic power station in the drone aerial images using the mapping relationship and real geographic coordinates, the method includes: Train the object detection model to obtain a trained object detection model; Based on the aerial images captured by the UAV, the key components of the photovoltaic power station in the UAV aerial images are obtained by segmentation and recognition using the trained target detection model.

8. The method according to claim 7, characterized in that, The object detection model is trained to obtain a trained object detection model, including: A labeled training dataset for photovoltaic construction scenarios is constructed, wherein the labeled training dataset covers sample images of different installation stages of photovoltaic power plants, different illumination angles, shading conditions and background interference. For recognition scenarios requiring high precision, a high-precision target detection model is constructed; for recognition scenarios deployed at the edge, a lightweight target detection model is constructed. Based on the labeled training dataset, the high-precision target detection model or the lightweight target detection model is trained to obtain a trained target detection model.

9. The method according to claim 1, characterized in that, Aerial photography of the photovoltaic power station construction area using drones yielded the following images: Based on the vector boundary of the photovoltaic power station to be monitored, a gridded aerial photography route covering the entire area is automatically generated in the ground control software of the UAV. The gridded aerial photography route takes into account environmental factors of image quality, including flight altitude, forward overlap rate, lateral overlap rate, lighting conditions and wind speed. Guided by the gridded aerial photography route, the UAV acquires aerial images and simultaneously records the flight log corresponding to each image. The flight log includes timestamps, IMU attitude, and raw GNSS observations.

10. A photovoltaic power station construction progress identification system, characterized in that, The system is used to perform the method according to any one of claims 1 to 9, and the system includes an acquisition module, an extraction module, a construction module, a generation module, and an evaluation module; The acquisition module is used to take aerial photos of the construction area of ​​the photovoltaic power station using a drone, and obtain drone aerial images. The extraction module is used to acquire embedded metadata for each drone aerial image and extract key imaging parameters from the embedded metadata. The construction module is used to establish a mapping relationship between pixel coordinates and real geographic coordinates based on the UAV aerial images and key imaging parameters through a camera imaging model. The generation module is used to perform spatial clustering and deduplication fusion on the key components of the photovoltaic power station in the drone aerial image using the mapping relationship and real geographic coordinates to generate a regional component distribution map. The evaluation module is used to dynamically evaluate the construction progress of photovoltaic power stations within a preset area based on the regional component distribution map.

Images