Method, device and equipment for reconstructing vegetation coverage of solar power station

Abstract

The present disclosure provides a method, device and equipment for reconstructing vegetation coverage of a solar power station, and relates to the field of image analysis technology. The method includes: obtaining a multispectral remote sensing image of a target solar power station and an obstructed surface image corresponding to an obstructed area of ​​a solar energy collection device in the target solar power station; classifying the surface type in the multispectral remote sensing image based on the spectral information in the multispectral remote sensing image to identify the obstructed area; determining the actual vegetation coverage data corresponding to the obstructed area according to the obstructed surface image; and using the actual vegetation coverage data corresponding to the obstructed area to correct the vegetation coverage data corresponding to the multispectral remote sensing image to obtain the reconstructed vegetation coverage of the solar power station. The technical solution in the present disclosure can improve the accuracy and reliability of the vegetation coverage of the solar power station.

Description

Background Art

[0002] Vegetation coverage, which is the ratio of the vertical projection area of ​​vegetation on the surface in a specific area to the total area of ​​the area, can be used to characterize vegetation growth and is an important indicator of the health of the regional ecosystem. In the existing solar power plant ecological monitoring technology, vegetation coverage is an important parameter for evaluating the health level of the power plant area ecosystem.

[0003] Taking photovoltaic power stations as an example, a large number of photovoltaic panels are usually installed in the layout design of photovoltaic power stations, and these panels cover most of the surface area of ​​the power station. However, related technologies usually rely on multispectral remote sensing images to determine the vegetation coverage in the area where the photovoltaic power station is located. However, due to the low resolution characteristics of multispectral remote sensing images and the small intervals between photovoltaic panels, a single pixel in the multispectral remote sensing image often contains mixed information of the photovoltaic panel and the surface area around the panel. In addition, due to the shading effect of the photovoltaic panel, the actual vegetation conditions under the coverage of the photovoltaic panel are not considered.

[0004] Therefore, when determining the vegetation coverage of solar power stations, relevant technical means fail to take into account the lack of vegetation information and estimation bias caused by the presence of photovoltaic panels and other devices. Therefore, it is necessary to reconstruct the vegetation coverage of solar power stations to obtain more accurate and reliable ecological monitoring data.

[0005] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute relevant technology known to ordinary technicians in the field. Summary of the invention

[0006] The purpose of the embodiments of the present disclosure is to provide a method for reconstructing vegetation coverage of a solar power station, a device for reconstructing vegetation coverage of a solar power station, an electronic device and a computer-readable storage medium, thereby improving the accuracy and reliability of vegetation coverage of a solar power station at least to a certain extent.

[0007] Other features and advantages of the present disclosure will become apparent from the following detailed description, or may be learned in part by the practice of the present disclosure.

[0008] According to a first aspect of an embodiment of the present disclosure, a method for reconstructing vegetation coverage of a solar power station is provided, comprising: obtaining a multispectral remote sensing image of a target solar power station and an obstructed surface image corresponding to an obstructed area of ​​a solar energy collection device in the target solar power station; classifying the surface type in the multispectral remote sensing image based on spectral information in the multispectral remote sensing image to identify the obstructed area; determining actual vegetation coverage data corresponding to the obstructed area based on the obstructed surface image; and using the actual vegetation coverage data corresponding to the obstructed area to correct the vegetation coverage data corresponding to the multispectral remote sensing image to obtain a reconstructed vegetation coverage of the solar power station.

[0009] In some example embodiments of the present disclosure, based on the aforementioned scheme, the obstructed surface image is acquired by a remote-controlled remote sensing image acquisition device, and the actual vegetation coverage data corresponding to the obstructed area is determined based on the obstructed surface image, including: determining the normalized vegetation index corresponding to the obstructed surface image based on image information in the obstructed surface image; and determining the actual vegetation coverage data corresponding to the obstructed area based on the normalized vegetation index.

[0010] In some example embodiments of the present disclosure, based on the aforementioned scheme, the occluded surface image is acquired by a handheld image acquisition device, and the occluded surface image includes multiple groups of image pairs consisting of occluded sampling images and non-occluded sampling images. Then, determining the actual vegetation coverage data corresponding to the occluded area according to the occluded surface image includes: determining first vegetation coverage data corresponding to the occluded sampling image and second vegetation coverage data corresponding to the non-occluded sampling image; determining the coverage ratio corresponding to each group of the image pairs according to the first vegetation coverage data and the second vegetation coverage data; and performing interpolation processing using the coverage ratio to obtain the coverage ratios corresponding to all areas in the multispectral remote sensing image.

[0011] In some example embodiments of the present disclosure, based on the aforementioned scheme, the determining of the first vegetation coverage data corresponding to the blocked sampling image and the second vegetation coverage data corresponding to the unblocked sampling image includes: in response to the blocked surface image being a remote sensing image, determining the first vegetation coverage data and the second vegetation coverage data according to the ratio of the vegetation projection area to the regional surface area in the remote sensing image; or

[0012] In response to the blocked surface image being a multispectral image, determining the vegetation area according to the normalized vegetation index corresponding to the multispectral image, and determining the first vegetation coverage data and the second vegetation coverage data based on the proportion of the vegetation area to the regional surface area; or

[0013] In response to the blocked surface image being an RGB image, the first vegetation coverage data and the second vegetation coverage data are determined according to the vegetation corresponding to the RGB image.

[0014] In some example embodiments of the present disclosure, based on the aforementioned scheme, the first vegetation coverage data and the second vegetation coverage data are determined according to the vegetation corresponding to the RGB image, including: binarizing the RGB image according to the vegetation index to determine the vegetation pixels in the RGB image; and determining the first vegetation coverage data and the second vegetation coverage data according to the ratio of the number of vegetation pixels to the total number of pixels.

[0015] In some example embodiments of the present disclosure, based on the aforementioned scheme, the actual vegetation coverage data corresponding to the occluded area is used to correct the vegetation coverage data corresponding to the multispectral remote sensing image, including: using a preset convolution kernel to traverse the image data after surface type classification of the multispectral remote sensing image; in response to the central pixel of the convolution kernel being in the occluded area, determining the central pixel vegetation coverage corresponding to the central pixel according to the ratio of the vegetation coverage corresponding to the non-central pixel in the convolution kernel and the coverage corresponding to the central pixel; correcting the vegetation coverage data corresponding to the multispectral remote sensing image based on the central pixel vegetation coverage

[0016] In some example embodiments of the present disclosure, based on the above scheme, determining the central pixel vegetation coverage corresponding to the central pixel according to the ratio of the vegetation coverage corresponding to the non-central pixel in the convolution kernel and the coverage corresponding to the central pixel includes:

[0017] Based on the inverse distance weighted method, according to:

[0018]

[0019] Determine the weight corresponding to the non-central pixel of the convolution kernel; wherein, represents the weight corresponding to the non-central pixel, Represents the non-central pixel i The distance from the center pixel, n represents the number of the non-central pixels;

[0020] according to:

[0021]

[0022] Determine the vegetation coverage of the central pixel corresponding to the central pixel; wherein, Indicates the vegetation coverage corresponding to the non-central pixel. rIndicates the coverage ratio corresponding to the central pixel.

[0023] In some example embodiments of the present disclosure, based on the aforementioned scheme, the classification of the surface type in the multispectral remote sensing image based on the spectral information in the multispectral remote sensing image includes: determining the normalized vegetation index data and the normalized building index data corresponding to the multispectral remote sensing image based on the spectral information in the multispectral remote sensing image; inputting the normalized vegetation index data and the normalized building index data into a classifier composed of a random forest network, a support vector machine and a K nearest neighbor network to classify the surface type in the multispectral remote sensing image.

[0024] According to a second aspect of an embodiment of the present disclosure, a device for reconstructing vegetation coverage of a solar power station is provided, comprising: an image acquisition unit, used to acquire a multispectral remote sensing image of a target solar power station and an obstructed surface image corresponding to an obstructed area of ​​a solar energy collection device in the target solar power station; an obstruction determination unit, used to classify the surface type in the multispectral remote sensing image based on spectral information in the multispectral remote sensing image to identify the obstructed area; a coverage determination unit, used to determine actual vegetation coverage data corresponding to the obstructed area based on the obstructed surface image; and a data correction unit, used to correct the vegetation coverage data corresponding to the multispectral remote sensing image using the actual vegetation coverage data corresponding to the obstructed area to obtain the reconstructed vegetation coverage of the solar power station.

[0025] According to a third aspect of an embodiment of the present disclosure, there is provided an electronic device, comprising: a processor; and a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the above-mentioned method for reconstructing vegetation coverage of a solar power station is implemented.

[0026] According to a fourth aspect of an embodiment of the present disclosure, there is provided a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the method for reconstructing the vegetation coverage of a solar power station as described above is implemented.

[0027] The technical solution provided by the embodiments of the present disclosure may have the following beneficial effects:

[0028] The method for reconstructing the vegetation coverage of a solar power station in the example embodiment of the present disclosure, on the one hand, can obtain all the surface information of the solar power station area by acquiring multispectral remote sensing images and obstructed surface images. The multispectral remote sensing images provide rich spectral information, while the obstructed surface images help to understand the surface conditions of the obstructed area in detail, ensuring the comprehensiveness of the data and laying the foundation for subsequent vegetation coverage analysis. On the other hand, through the combined use of multispectral remote sensing images and obstructed surface images, the surface conditions under the obstruction of the solar energy collection device can be more accurately identified, thereby improving the accuracy of vegetation coverage measurement. On the other hand, by using the actual vegetation coverage data of the obstructed area to reconstruct the vegetation coverage data of the entire solar power station obstructed area, the vegetation coverage data finally obtained can be made more real and reliable. Thereby, the accuracy and reliability of the vegetation coverage of the solar power station are further improved.

[0029] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The accompanying drawings herein are incorporated into the specification and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the specification are used to explain the principles of the present disclosure. Obviously, the accompanying drawings described below are only some embodiments of the present disclosure, and for ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without creative work.

[0031] Figure 1 The schematic diagram shows a solar energy collection device blocking remote sensing signals to obtain real surface information.

[0032] Figure 2 A schematic flow chart of a method for reconstructing vegetation coverage of a solar power station according to some embodiments of the present disclosure is schematically shown.

[0033] Figure 3 A schematic diagram of setting sampling point pairs in a photovoltaic power plant according to some embodiments of the present disclosure is schematically shown.

[0034] Figure 4 The figure schematically shows the arrangement of sampling point pairs in a CSP plant according to some embodiments of the present disclosure.

[0035] Figure 5 A schematic diagram schematically illustrates the surface type classification results of local multispectral remote sensing images according to some embodiments of the present disclosure.

[0036] Figure 6 The following schematic diagram shows an image after binarization processing of an occluded surface image according to some embodiments of the present disclosure.

[0037] Figure 7 The diagram schematically shows a process of correcting vegetation coverage data using a convolution kernel in some embodiments of the present disclosure.

[0038] Figure 8 A block diagram of a device for reconstructing vegetation coverage of a solar power station according to some embodiments of the present disclosure is schematically shown.

[0039] Fig. 9 A schematic diagram of the structure of a computer system of an electronic device according to some embodiments of the present disclosure is schematically shown.

[0040] Fig.10 A schematic diagram of a computer-readable storage medium according to some embodiments of the present disclosure is schematically shown.

[0041] In the drawings, the same or corresponding reference numerals represent the same or corresponding parts. DETAILED DESCRIPTION

[0042] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments do not represent all implementations consistent with this specification. Instead, they are merely examples of devices and methods consistent with some aspects of this specification as detailed in the appended claims.

[0043] The terms used in this specification are for the purpose of describing specific embodiments only and are not intended to limit this specification. The singular forms "a", "the" and "the" used in this specification and the appended claims are also intended to include plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0044] It should be understood that although the terms first, second, third, etc. may be used in this specification to describe various information, this information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of this specification, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0045] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be more comprehensive and complete and will fully convey the concept of the example embodiments to those skilled in the art.

[0046] In addition, the described features, structures or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced without one or more of the specific details, or other methods, components, devices, steps, etc. may be adopted. In other cases, known methods, devices, implementations or operations are not shown or described in detail to avoid blurring the various aspects of the present disclosure.

[0047] Furthermore, the drawings are only schematic illustrations and are not necessarily drawn to scale. The block diagrams shown in the drawings are only functional entities and do not necessarily correspond to physically separate entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0048] As the global demand for renewable energy increases, the deployment of solar power stations, especially photovoltaic power stations and solar thermal power stations, has grown rapidly. Among them, photovoltaic power generation directly converts solar energy into electrical energy through photovoltaic panels, while solar thermal power generation reflects sunlight to a heat storage medium through a reflector, and uses the heat energy of the heat storage medium to generate electricity through steam turbines and other methods. However, although photovoltaic power generation and solar thermal power generation have different energy conversion paths, they both require the installation of a large number of solar energy collection devices such as photovoltaic panels or solar thermal heliostats on the ground, which makes it difficult for traditional remote sensing technology to obtain the true conditions of the surface below the facilities.

[0049] Since solar energy collection devices may block or reflect remote sensing signals, the data obtained by satellite or aerial remote sensing in related technologies may be inaccurate or unable to reflect the specific conditions of the ground surface below the solar energy collection device. Figure 1 As shown, when the remote sensing signal 104 is sent to the ground surface 103, the solar thermal heliostat 101 and the photovoltaic panel 102 will intercept the remote sensing signal, so that the information under the photovoltaic panel or the solar thermal heliostat cannot be detected through the remote sensing signal.

[0050] In the existing ecological monitoring technology of solar power stations, vegetation coverage is an important parameter for assessing the health level of the ecosystem in the power station area. Before and after the construction of a solar power station, the vegetation coverage in the area where it is located often changes to a certain extent. On the one hand, during the construction of a solar power station, the rolling of heavy-loaded vehicles and engineering activities will destroy the stability of the surface and affect the growth of vegetation, and the power station facilities such as booster stations and roads will change the land cover type, thereby affecting the vegetation coverage. On the other hand, after the construction of the solar power station is completed, if the affected vegetation has not been converted to other land use types, its vegetation coverage will change due to the shielding of solar radiation by photovoltaic panels or solar thermal heliostats. Shielding may cause the temperature in the shaded area to drop, reduce evaporation, and be beneficial to water retention; at night, the panels reflect long-wave radiation, which has a warming effect on the surface and reduces the risk of frost damage, thereby affecting the vegetation coverage in the area shielded by photovoltaic panels or solar thermal heliostats. Therefore, the change in vegetation coverage is not only an important indicator of the change in the ecosystem before and after the construction of a solar power station, but also one of the key parameters for assessing the impact of the power station on the surrounding environment.

[0051] However, in the related art, when determining the vegetation coverage of the solar power station area, the spatial resolution of the commonly used remote sensing is usually between tens of meters and thousands of meters. Taking a photovoltaic power station as an example, the projection width of the photovoltaic panel on the ground is about 4-5 meters, and the distance between two rows of photovoltaic panels is usually 4-6 meters. Therefore, when using remote sensing products with a spatial resolution of more than tens of meters to determine the vegetation coverage of the photovoltaic power station area, each pixel actually contains several rows of photovoltaic panels and the ground area between them. This mixed information will cause deviations in the evaluation results of vegetation coverage, and due to the shielding effect of photovoltaic panels, the actual vegetation conditions under the coverage of photovoltaic panels are not considered. Therefore, when determining the vegetation coverage of solar power stations, the related technical means failed to consider the missing vegetation information and estimation deviation caused by the presence of photovoltaic panels and other devices, resulting in poor measurement accuracy of vegetation coverage. Therefore, it is necessary to reconstruct the vegetation coverage of solar power stations to obtain more accurate and reliable ecological monitoring data.

[0052] In order to solve all or part of the technical problems in the above-mentioned related technologies, in an exemplary embodiment of the present disclosure, a method for reconstructing vegetation coverage of a solar power station is firstly provided. Figure 2 The following is a schematic diagram of a process flow of a method for reconstructing vegetation coverage of a solar power station according to some embodiments of the present disclosure. Figure 2 As shown, the method for reconstructing the vegetation coverage of the solar power station may include the following steps:

[0053] Step S210, obtaining a multispectral remote sensing image of the target solar power station and an obstructed surface image corresponding to an obstructed area of ​​a solar energy collection device in the target solar power station;

[0054] Step S220, classifying the surface type in the multispectral remote sensing image based on the spectral information in the multispectral remote sensing image to identify the blocked area;

[0055] Step S230, determining actual vegetation coverage data corresponding to the blocked area according to the blocked surface image;

[0056] Step S240, using the actual vegetation coverage data corresponding to the blocked area, the vegetation coverage data corresponding to the multispectral remote sensing image is corrected to obtain the reconstructed vegetation coverage of the solar power station.

[0057] Specifically, firstly, a multispectral remote sensing image of the target solar power station area is obtained. The multispectral remote sensing image can represent the surface information obtained in multiple electromagnetic bands, which can provide rich spectral characteristics of the ground objects. In addition, it is necessary to obtain the surface image of the area blocked by the solar energy collection device, which can be obtained by drone aerial photography or other image acquisition devices such as cameras. Then, using the spectral information in the multispectral remote sensing image, the surface type is classified through image processing and classification algorithms, and which areas are blocked by the solar energy collection device are identified, so as to determine the location and range of the blocked area. Then, using the above-obstructed surface image obtained, further image analysis technology is used to determine the actual vegetation coverage data of the blocked area for subsequent vegetation coverage correction. Finally, according to the actual vegetation coverage data of the blocked area, the vegetation coverage data in the entire multispectral remote sensing image is corrected. Through this method, the influence of photovoltaic panels or photothermal heliostats on vegetation coverage measurement can be eliminated, so as to obtain more accurate vegetation coverage data of the solar power station area.

[0058] The following further describes the method for reconstructing the vegetation coverage of the solar power station in the above exemplary embodiment.

[0059] In step S210, a multispectral remote sensing image of a target solar power station and an obstructed surface image corresponding to an obstructed area of ​​a solar energy collection device in the target solar power station are obtained.

[0060] The target solar power station can be a photovoltaic power station or a solar thermal power station. Multispectral remote sensing images can represent surface images covering multiple electromagnetic bands acquired through a remote sensing platform, wherein each band corresponds to a different electromagnetic wavelength range and can reflect the reflectivity or emissivity of surface objects at different wavelengths. This information can be used to identify and distinguish different types of surface covers, such as vegetation, soil, and artificial buildings. Solar energy collection devices can represent equipment used to collect solar radiation energy and convert it into usable energy, such as photovoltaic panels in photovoltaic power stations and heliostats in solar thermal power stations. Shielded surface images can represent images taken of surface areas shielded by solar energy collection devices. These images can be acquired through remote-controlled remote sensing image acquisition equipment or handheld image acquisition equipment to display the surface features of the shielded area in detail.

[0061] Preferably, the multispectral remote sensing image in the embodiment of the present disclosure is a sub-meter-level high-resolution multispectral satellite remote sensing image, that is, the spatial resolution of the multispectral remote sensing image is not greater than 1 meter, so that a single pixel in the multispectral remote sensing image does not contain mixed information of the solar energy collection device and the surface area around the device as much as possible, thereby improving the reliability of vegetation coverage measurement. Exemplarily, the multispectral remote sensing image can be a GF-2 remote sensing image with a spatial resolution of 0.8 meters, or a Quickbird image with a spatial resolution of 0.65 meters. In addition, the multispectral remote sensing image needs to have visible light and near-infrared bands for the subsequent calculation of vegetation indices, and the higher the spatial resolution of the multispectral remote sensing image, the more accurately the vegetation coverage of the solar power station can be reconstructed.

[0062] In some embodiments, when acquiring a multispectral remote sensing image of a target solar power station, the following steps can be used to acquire the image: preliminarily locate the solar power station according to the location coordinates of the target solar power station, and use a map tool such as Geographical Information System (GIS), Google Earth, or Aowei to delineate the spatial position of the target solar power station, and obtain the longitude and latitude range corresponding to the target solar power station. Then, using the longitude and latitude range and the commissioning time of the power station, sub-meter-level high-resolution multispectral remote sensing images are screened out from satellite remote sensing images. The spatial range of the selected image should completely cover the power station area, and the image acquisition time should be after the power station is completed and put into operation. In addition, ensure that the power station area in the selected image is cloudless or has little cloud coverage to reduce the impact of clouds on the acquisition of surface information.

[0063] In some embodiments, after obtaining the multispectral remote sensing image of the target solar power station, it is necessary to perform image preprocessing on the multispectral remote sensing image to ensure the accuracy of the power station location, equipment location and spectral information in the image. Among them, image preprocessing may include steps such as radiation correction, geometric correction and terrain correction. Among them, radiation correction can be used to eliminate the influence of factors such as atmospheric conditions and sensor characteristics on the image radiation value, and ensure that the radiation intensity of the image truly reflects the surface reflectivity or emissivity. Geometric correction can be used to correct geometric deformation caused by sensor attitude, earth curvature, terrain undulations, etc., to ensure the accuracy of the geographic coordinate position of each pixel in the image. Terrain correction can be used to eliminate the influence of terrain undulations on the image brightness value.

[0064] In some embodiments, when acquiring obstructed surface images, the acquisition can be performed through a remote-controlled remote sensing image acquisition device or a handheld image acquisition device. Among them, the remote-controlled remote sensing image acquisition device can refer to a device that is operated by remote control and is used to acquire surface images. Such devices include but are not limited to: drones, small remote-controlled aircraft, etc. The advantages of remote-controlled remote sensing image acquisition devices are that they are highly flexible and maneuverable, can respond quickly to needs, provide high-resolution surface images, and can avoid human interference to a certain extent. Handheld image acquisition devices can refer to portable devices that are directly operated by operators to capture images, including but not limited to multispectral cameras, digital cameras, and smart phones. The advantages of handheld image acquisition devices are that they are easy to operate, relatively low cost, and the shooting angle and range can be directly controlled by the operator.

[0065] Furthermore, since photovoltaic panels and solar thermal heliostats have a certain tilt angle and are a certain distance from the ground, drone oblique photography technology can be used to obtain obstructed surface images corresponding to the obstructed areas of solar energy collection devices. In order to obtain high-quality obstructed area information, drones should be equipped with sensors with red, green, and blue bands of visible light and near-infrared bands. These sensors can capture surface reflectance information in different bands, and then calculate vegetation indices to determine vegetation coverage. The images obtained by drones need to be preprocessed to ensure that their spatial coordinate system and spatial position of the objects are consistent with sub-meter multispectral remote sensing images.

[0066] In addition, when using handheld image acquisition equipment to collect obstructed surface images, due to the large coverage area of ​​the solar energy collection device and the inconvenience of manual operation, image pairs consisting of obstructed sampling images and non-obstructed sampling images can be obtained based on point pairs consisting of adjacent obstructed points and non-obstructed points. The setting of sampling point pairs in photovoltaic power plants can refer to Figure 3As shown, the shielding point is set below the photovoltaic panel 102, the non-shielding point is adjacent to its corresponding shielding point, and is located in the ground surface 103 area between the two photovoltaic panels. Exemplarily, the sampling point pair in the photovoltaic power station may include a first sampling point pair consisting of a first shielding point 105 and a first non-shielding point 106, and a second sampling point pair consisting of a second shielding point 107 and a second non-shielding point 108. The setting of the sampling point pair in the solar thermal power station can refer to Figure 4 As shown, the blocking point is set under the solar thermal heliostat 101, the non-blocking point is adjacent to its corresponding blocking point, and is located in the surface 103 area between the two solar thermal panels. For example, the sampling point pair in the solar thermal power station may include a third sampling point pair consisting of a third blocking point 109 and a third non-blocking point 110, and a fourth sampling point pair consisting of a fourth blocking point 111 and a fourth non-blocking point 112. In addition, if a multispectral camera is used, it should be ensured that the collected image contains red, green, and blue bands of visible light and near-infrared bands, and the sampling point pairs need to be as dense as possible and evenly distributed throughout the power station area to improve the representativeness of the sample image and the coverage of the data. When using a handheld image acquisition device to collect blocked surface images, the coordinates of each shooting point need to be recorded, and the coordinate information can be obtained through a handheld GPS device or other device with positioning function. The acquired image data needs to be spatially registered to ensure that it is in the same spatial coordinate system as the sub-meter multispectral remote sensing image.

[0067] In step S220, the surface type in the multispectral remote sensing image is classified based on the spectral information in the multispectral remote sensing image to identify the blocked area.

[0068] Among them, spectral information can represent the reflectivity or emissivity of surface objects in different electromagnetic bands. Multispectral remote sensing images obtain spectral characteristics of surface objects through multiple bands. These bands usually include visible light (such as red, green, blue) and near infrared, etc. Different objects have different reflectivity or emissivity in each band. These differences constitute the spectral characteristics of the objects and are an important basis for distinguishing different surface types. Surface types can represent different categories of surface cover, such as vegetation, bare soil, buildings, roads, photovoltaic panels, solar thermal heliostats and other types. The blocked area can represent the surface area where the remote sensing signal is partially or completely blocked due to the presence of solar energy collection devices. Specifically, supervised classification or unsupervised classification algorithms can be applied to classify surface types in multispectral remote sensing images.

[0069] In some embodiments, based on the spectral information in the multispectral remote sensing image, the surface type in the multispectral remote sensing image is classified, specifically including the following steps: based on the spectral information in the multispectral remote sensing image, determining the normalized vegetation index data and the normalized building index data corresponding to the multispectral remote sensing image; inputting the normalized vegetation index data and the normalized building index data into a classifier composed of a random forest network, a support vector machine and a K nearest neighbor network to classify the surface type in the multispectral remote sensing image.

[0070] Among them, the Normalized Difference Vegetation Index (NDVI) is a commonly used remote sensing vegetation index that evaluates vegetation coverage by calculating the difference between the near-infrared band and the red band. The Normalized Difference Vegetation Index can be calculated using the following formula (1):

[0071] (1)

[0072] in, NDVI represents the normalized difference vegetation index, NIR Represents the near-infrared band reflectance of multispectral remote sensing images. RED Represents the red light band reflectance of the multispectral remote sensing image.

[0073] The Normalized Difference Built-up Index (NDBI) is a remote sensing index used to distinguish built-up land from other land surface types. It evaluates the coverage of built-up land by calculating the difference between the near-infrared band and the short-wave infrared band. The Normalized Difference Built-up Index can be calculated using the following formula (2):

[0074] (2)

[0075] in, NDBI represents the normalized building index, SWIR Represents the shortwave infrared band reflectance of multispectral remote sensing images.

[0076] Furthermore, Random Forest is an ensemble learning method that improves prediction accuracy and prevents overfitting by building multiple decision trees and combining the decision results. Random Forest selects a random subset of features at each node in the network for segmentation and uses a different subset of training data in each decision tree. Finally, the final classification result is determined by a voting mechanism or an average. Support Vector Machine (SVM) is a supervised learning algorithm that maps data to a high-dimensional space and finds a boundary in this space that best distinguishes different categories. K-Nearest Neighbors (KNN) is an instance-based learning algorithm that classifies based on the majority vote of the K nearest neighbors of the input instance in the feature space.

[0077] Before using the classifier composed of random forest network, support vector machine and K nearest neighbor network to classify the surface type, it is necessary to use the training set and validation set to pre-train the classifier and verify the classification effect. Specifically, sample points are selected on the multispectral remote sensing images corresponding to each surface type to be classified. When selecting sample points, it should be ensured that each sample point can represent the land type in which it is located, that is, the sample point should be able to fully reflect the characteristics of the surface of this type, and the number of sample points should be sufficient to support the training of the machine learning algorithm. Usually, at least enough sample points are required for each surface type to ensure that the training set is sufficiently representative. Then, from the sample points of each surface type, 30% to 50% of the number of training sample points are additionally selected as validation sample points. Next, GIS tools are used to extract the spectral information of each training and validation sample point from the multispectral remote sensing image, including feature values ​​such as normalized vegetation index and normalized building index. The extracted feature values ​​are combined with the corresponding surface type labels to form training sets and validation sets, where the data format of each sample point includes spectral feature values ​​and corresponding land use type labels. During the model training process, it is necessary to adjust the input variables, number of sample points, proportion of each type of sample points, classification algorithm parameters, etc. until the optimal classification accuracy is achieved.

[0078] Exemplarily, the training set and validation set are constructed using the Gaofen-2 satellite remote sensing data with a spatial resolution of 0.8m of the target CSP station. The surface types include grassland, bare soil and heliostats, where bare soil includes artificial buildings and natural surfaces. 550 sample points are selected as the training set, and an additional 200 sample points are used as the validation set. Classification is performed using a voting classifier, which is a classification method that integrates the results of multiple classifiers. In this embodiment, the classifiers participating in the voting are composed of a random forest network, a support vector machine, and a K nearest neighbor network, and their training accuracy is shown in Table 1 below:

[0079] Table 1

[0080]

[0081] As shown in Table 1, in this embodiment, through detailed sample point selection, data extraction and classification model training, the overall classification training accuracy reached 0.958, indicating that 95.8% of the test samples were correctly classified. This result shows that the classifier has high accuracy and generalization ability, and can effectively identify and distinguish different surface types. In addition, the user accuracy of each surface type has reached a high level, which shows that the classification result is highly consistent with the actual surface type and achieves the expected classification effect.

[0082] Furthermore, the trained classifier is tested using the validation set, and the classification accuracy obtained by verification is shown in Table 2 below:

[0083] Table 2

[0084]

[0085] As shown in Table 2, in this embodiment, the overall classification accuracy of the validation set reaches 0.940, which can meet the classification of land types. Then, the classifier is used to classify the surface types in the multispectral remote sensing image to be classified. The local classification results of the image are as follows: Figure 5 shown.

[0086] In step S230, actual vegetation coverage data corresponding to the blocked area is determined according to the blocked surface image.

[0087] Specifically, the vegetation coverage corresponding to each pixel in the image can be determined according to the normalized vegetation index information in the obstructed surface image. In some embodiments, when the obstructed surface image is acquired by a remote-controlled remote sensing image acquisition device, the actual vegetation coverage data corresponding to the obstructed area is determined according to the obstructed surface image, which specifically includes the following steps: determining the normalized vegetation index corresponding to the obstructed surface image according to the image information in the obstructed surface image; and determining the actual vegetation coverage data corresponding to the obstructed area according to the normalized vegetation index.

[0088] Among them, the normalized vegetation index corresponding to each pixel can be determined according to the near-infrared band reflectance and red light band reflectance information in the obstructed surface image, and then based on the normalized vegetation index, the vegetation coverage corresponding to each pixel can be determined by the following formula (3):

[0089] (3)

[0090] in, FVC Represents vegetation coverage, Indicates the maximum NDVI value of the current area. Indicates the minimum NDVI value of the current area. Indicates the NDVI value of the current pixel. After obtaining the vegetation coverage data corresponding to each pixel in the obstructed surface image, it can be used as the actual vegetation coverage data corresponding to the obstructed area in the multispectral remote sensing image through the coordinate correspondence, thereby completing the filling and reconstruction of the vegetation coverage.

[0091] In some embodiments, when the occluded surface image is acquired by a handheld image acquisition device, the occluded surface image includes multiple groups of image pairs consisting of occluded sampling images and non-occluded sampling images. The actual vegetation coverage data corresponding to the occluded area is determined according to the occluded surface image, which specifically includes the following steps: determining the first vegetation coverage data corresponding to the occluded sampling image and the second vegetation coverage data corresponding to the non-occluded sampling image; determining the coverage ratio corresponding to each group of image pairs according to the first vegetation coverage data and the second vegetation coverage data; and performing interpolation processing using the coverage ratio to obtain the coverage ratios corresponding to all areas in the multispectral remote sensing image.

[0092] Among them, when using a handheld image acquisition device to obtain an obstructed surface image, since the handheld image acquisition device cannot cover every pixel in the obstructed area, the vegetation coverage ratio of adjacent obstructed and non-obstructed areas can be calculated through point pairs, and the vegetation coverage ratios of multiple point pairs are interpolated in the space within the power station to obtain the vegetation coverage ratio corresponding to the entire power station area, and then the vegetation coverage of the obstructed area is reconstructed based on the vegetation coverage of the multispectral remote sensing image.

[0093] Specifically, an occluded sampling image is selected according to the image of the occluded area, and an adjacent non-occluded sampling image is selected according to the image of the non-occluded area, ensuring that the occluded sampling image and the non-occluded sampling image are adjacent in space to form multiple image pairs. The first vegetation coverage corresponding to the occluded sampling image is calculated, and the second vegetation coverage corresponding to the non-occluded sampling image is calculated. According to the first vegetation coverage data and the second vegetation coverage data, the coverage ratio of each group of image pairs is calculated. Using spatial interpolation methods such as Kriging interpolation and inverse distance weighted interpolation, the coverage ratios of multiple image pairs are interpolated in the space within the power station. An interpolation grid is created to ensure that the entire power station area is covered, and the spatial position, number of rows and columns, and spatial resolution of the interpolated vegetation coverage ratio data of the occluded area and the non-occluded area are consistent with the original sub-meter multispectral remote sensing image. The interpolated coverage ratio is combined with the vegetation coverage data of the sub-meter remote sensing image to reconstruct the vegetation coverage of the occluded area.

[0094] In some embodiments, determining first vegetation coverage data corresponding to an occluded sampling image and second vegetation coverage data corresponding to an unoccluded sampling image includes: in response to the occluded surface image being a remote sensing image, determining the first vegetation coverage data and the second vegetation coverage data based on the ratio of the vegetation projection area to the regional surface area in the remote sensing image; in response to the occluded surface image being a multispectral image, determining the vegetation area based on the normalized vegetation index corresponding to the multispectral image, and determining the first vegetation coverage data and the second vegetation coverage data based on the ratio of the vegetation area to the regional surface area; in response to the occluded surface image being an RGB image, determining the first vegetation coverage data and the second vegetation coverage data based on the vegetation index corresponding to the RGB image.

[0095] Specifically, when the obstructed surface image is a remote sensing image collected by a handheld remote sensing device, since the handheld remote sensing device can clearly capture each plant and the exposed soil on the ground, the vegetation coverage can be calculated based on the ratio of the vegetation projection area to the regional surface area, that is, according to the following formula (4):

[0096] (4)

[0097] in, represents the vegetation projection area, Represents the surface area of ​​a region.

[0098] When the obstructed surface image is a multispectral image collected by a handheld multispectral imaging device, the normalized vegetation index can be calculated by the near-infrared band reflectance and the red light band reflectance, and then the vegetation area and the soil area can be distinguished according to the normalized vegetation index to determine the vegetation coverage data.

[0099] In some embodiments, first vegetation coverage data and second vegetation coverage data are determined according to the vegetation index corresponding to the RGB image, specifically including the following steps: binarizing the RGB image according to the super green index and the visible light atmospheric impedance index to determine the vegetation pixels in the RGB image; determining the first vegetation coverage data and the second vegetation coverage data according to the ratio of the number of vegetation pixels to the total number of pixels.

[0100] Among them, the vegetation index can represent a metric used to evaluate the growth and coverage of vegetation in an RGB image. In the disclosed embodiment, the vegetation index may include a super green index, a visible light atmospheric impedance index, a triangular greenness index, a green-blue difference vegetation index, a super red index, a visible light difference vegetation index, a vegetation color index, a normalized green-red difference index, a red-green ratio vegetation index, a red-green-blue ratio vegetation index, an over-blue index, and other data that can characterize the vegetation coverage in an RGB image.

[0101] For example, the first vegetation coverage data and the second vegetation coverage data can be determined according to the super green index and the visible light atmospheric impedance index corresponding to the RGB image. The super green index can be obtained by For calculation, the visible light atmospheric impedance index can be obtained by Calculate, where Excess Green index, Represents the Visible Atmospherically Resistant Index. , and Represent the red light reflectance, green light reflectance and blue light reflectance respectively. Then, the Otsu threshold method is used to determine the surface type corresponding to each pixel in the RGB image, namely, vegetation pixel and non-vegetation pixel, according to the super green index and visible light atmospheric impedance index. The RGB image is binarized according to the surface type corresponding to the pixel. For example, the vegetation pixel is assigned 1 and the non-vegetation pixel is assigned 0. The image after binarization is as follows: Figure 6 Finally, the vegetation coverage data is determined based on the ratio of the number of vegetation pixels to the total number of pixels.

[0102] In step S240, the vegetation coverage data corresponding to the multispectral remote sensing image is corrected using the actual vegetation coverage data corresponding to the blocked area to obtain the reconstructed vegetation coverage of the solar power station.

[0103] In some embodiments, when the obstructed surface image is acquired by a remote-controlled remote sensing image acquisition device, since the remote-controlled remote sensing image acquisition device can completely acquire the surface image of the obstructed area, after obtaining the vegetation coverage data corresponding to each pixel in the obstructed surface image, it can be used as the actual vegetation coverage data corresponding to the obstructed area in the multispectral remote sensing image through the coordinate correspondence, thereby completing the correction of the vegetation coverage and obtaining the reconstructed vegetation coverage of the solar power station.

[0104] In some embodiments, when an obstructed surface image is acquired by a handheld image acquisition device, the vegetation coverage data corresponding to the multispectral remote sensing image is corrected using the actual vegetation coverage data corresponding to the obstructed area, specifically including the following steps: using a preset convolution kernel to traverse the image data after surface type classification of the multispectral remote sensing image; in response to the central pixel of the convolution kernel being in the obstructed area, determining the central pixel vegetation coverage corresponding to the central pixel based on the ratio of the vegetation coverage corresponding to the non-central pixel in the convolution kernel and the coverage corresponding to the central pixel; and correcting the vegetation coverage data corresponding to the multispectral remote sensing image based on the central pixel vegetation coverage.

[0105] When the obstructed surface image is acquired by a handheld image acquisition device, the vegetation coverage of the obstructed area can be corrected using the spatial convolution method. Figure 7 As shown, the vegetation coverage data can be corrected according to the surface type data, the vegetation coverage data corresponding to the multispectral remote sensing image, and the coverage ratio. Exemplarily, a 3×3 convolution kernel is used to traverse each pixel in the image data after the surface type classification of the multispectral remote sensing image. When the surface type of the central pixel of the convolution kernel is a photovoltaic panel or a solar thermal heliostat, and the surface type of the non-central pixel of the convolution kernel is a natural surface (such as grassland, shrubs, bare land, etc.) or a filled pixel, the vegetation coverage ratio between the natural surface pixel and the central pixel of the convolution kernel is determined based on the vegetation coverage ratio data of the occluded area and the non-occluded area. Then, the vegetation coverage of the natural surface pixel is determined, the vegetation coverage of the central pixel of the convolution kernel is calculated by using the inverse distance weighted method and other methods, and the vegetation coverage data corresponding to the multispectral remote sensing image is corrected by using the vegetation coverage of the central pixel. At the same time, the surface type of the pixel corresponding to the center pixel of the convolution kernel in the multispectral remote sensing image is modified to a filled pixel. Continue to traverse the entire image and repeat the above steps until all photovoltaic panels or solar thermal heliostat mirror images become filled pixels. In other embodiments of the present disclosure, the convolution kernel can also be a convolution kernel of other odd sizes such as 3×3, 5×5, 9×9, etc., to ensure that the center pixel can be located at the center of the convolution kernel.

[0106] In some embodiments, the vegetation coverage of the central pixel corresponding to the central pixel is determined according to the ratio of the vegetation coverage corresponding to the non-central pixel in the convolution kernel to the coverage corresponding to the central pixel, which specifically includes the following steps: based on the inverse distance weighted method, the weight corresponding to the non-central pixel of the convolution kernel is determined according to the following formula (5):

[0107] (5)

[0108] in, represents the weight corresponding to the non-central pixel, Represents the non-central pixel i The distance from the center pixel, n Indicates the number of non-central pixels;

[0109] The vegetation coverage of the central pixel corresponding to the central pixel is determined according to the following formula (6):

[0110] (6)

[0111] in, Indicates the vegetation coverage corresponding to the non-central pixel. r Indicates the coverage ratio corresponding to the central pixel.

[0112] In the above example embodiment, the method for reconstructing the vegetation coverage of a solar power station can, on the one hand, obtain all the surface information of the solar power station area by acquiring multispectral remote sensing images and obstructed surface images. The multispectral remote sensing images provide rich spectral information, while the obstructed surface images help to understand the surface conditions of the obstructed area in detail, ensuring the comprehensiveness of the data and laying the foundation for subsequent vegetation coverage analysis. On the other hand, through the combined use of multispectral remote sensing images and obstructed surface images, the surface conditions under the obstruction of the solar energy collection device can be more accurately identified, thereby improving the accuracy of vegetation coverage measurement. On the other hand, by using the actual vegetation coverage data of the obstructed area to reconstruct the vegetation coverage data of the entire solar power station obstructed area, the vegetation coverage data finally obtained can be made more real and reliable. Thereby, the accuracy and reliability of the vegetation coverage of the solar power station are further improved.

[0113] In addition, through multiple indicators such as the normalized vegetation index and the normalized building index, the surface types can be classified more comprehensively, and the classifier composed of the random forest network, the support vector machine and the K nearest neighbor network can be used to classify the surface types in the multispectral remote sensing image, thereby improving the accuracy and robustness of the classification; the interpolation processing using the coverage ratio can fill the data gaps in the blocked area and improve the consistency and continuity of the overall data; whether it is a remote sensing image, a multispectral image or an RGB image, the vegetation coverage data can be determined by different methods, thereby enhancing the versatility and applicability of the method; binarization is performed using the super green index and the visible light atmospheric impedance index. Processing can effectively distinguish between vegetation and non-vegetation areas, improve the recognition accuracy of vegetation coverage, and by calculating the ratio of the number of vegetation pixels to the total number of pixels, the vegetation coverage can be quantified to provide specific data support; by traversing the image data after surface type classification in the multispectral remote sensing image through the convolution kernel, the occluded area can be locally corrected to improve the accuracy of the data; according to the ratio of the vegetation coverage of the non-central pixel and the coverage of the central pixel in the convolution kernel, the vegetation coverage of the central pixel is dynamically adjusted to ensure the consistency and rationality of the data; through the inverse distance weighted method, the weights of the non-central pixels are reasonably allocated to ensure that the calculation of the vegetation coverage of the central pixel is more accurate.

[0114] It should be noted that, although the steps of the method in the present disclosure are described in a specific order in the drawings, this does not require or imply that the steps must be performed in this specific order, or that all the steps shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps, etc.

[0115] Next, in the embodiment of the present disclosure, a device for reconstructing vegetation coverage of a solar power station is also provided. Figure 8 As shown in the figure, the reconstruction device 800 of the vegetation coverage of the solar power station can be composed of an image acquisition module 801, an occlusion determination module 802, a coverage determination module 803 and a data correction module 804, wherein: the image acquisition module can be used to obtain the multispectral remote sensing image of the target solar power station and the occluded surface image corresponding to the occluded area of ​​the solar energy collection device in the target solar power station; the occlusion determination module can be used to classify the surface type in the multispectral remote sensing image based on the spectral information in the multispectral remote sensing image to identify the occluded area; the coverage determination module can be used to determine the actual vegetation coverage data corresponding to the occluded area according to the occluded surface image; the data correction module can be used to use the actual vegetation coverage data corresponding to the occluded area to correct the vegetation coverage data corresponding to the multispectral remote sensing image to obtain the reconstructed vegetation coverage of the solar power station.

[0116] It should be noted that the specific details of each part of the above-mentioned solar power station vegetation coverage reconstruction device have been described in detail in the implementation method of the solar power station vegetation coverage reconstruction method part. The undisclosed details can be found in the implementation method part, and thus will not be repeated here.

[0117] In addition, in an exemplary embodiment of the present disclosure, an electronic device capable of implementing the above-mentioned method for reconstructing vegetation coverage of a solar power station is also provided.

[0118] Those skilled in the art will appreciate that various aspects of the present disclosure may be implemented as systems, methods or program products. Therefore, various aspects of the present disclosure may be specifically implemented in the following forms, namely: complete hardware embodiments, complete software embodiments (including firmware, microcode, etc.), or embodiments combining hardware and software aspects, which may be collectively referred to herein as "circuits", "modules" or "systems".

[0119] Refer to the following Fig. 9 The electronic device 900 according to the above-mentioned embodiment of the present disclosure is described. Fig. 9 The electronic device 900 shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0120] like Fig. 9 As shown, the electronic device 900 is in the form of a general computing device. The components of the electronic device 900 may include but are not limited to: at least one processing unit 910, at least one storage unit 920, a bus 930 connecting different system components (including the storage unit 920 and the processing unit 910), and a display unit 940.

[0121] The storage unit stores program codes, which can be executed by the processing unit 910, so that the processing unit 910 executes the steps according to various exemplary embodiments of the present disclosure described in the above “Exemplary Method” section of this specification.

[0122] The storage unit 920 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 921 and / or a cache memory unit 922 , and may further include a read-only memory unit (ROM) 923 .

[0123] The storage unit 920 may also include a program / utility 924 having a set (at least one) of program modules 925, such program modules 925 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0124] Bus 930 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0125] The electronic device 900 may also communicate with one or more external devices 970 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 900, and / or communicate with any device that enables the electronic device 900 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed through an input / output (I / O) interface 950. In addition, the electronic device 900 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) through a network adapter 960. As shown, the network adapter 960 communicates with other modules of the electronic device 900 through a bus 930. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 900, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0126] Through the description of the above embodiments, it is easy for those skilled in the art to understand that the example embodiments described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the embodiment of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes a number of instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the embodiment of the present disclosure.

[0127] In an exemplary embodiment of the present disclosure, a computer-readable storage medium is also provided, on which a program product capable of implementing the above method of the present specification is stored. In some possible embodiments, various aspects of the present disclosure can also be implemented in the form of a program product, which includes a program code. When the program product is run on a terminal device, the program code is used to enable the terminal device to execute the steps according to various exemplary embodiments of the present disclosure described in the above "Exemplary Method" section of the present specification.

[0128] refer to Fig.10 As shown, a program product 1000 for implementing the above-mentioned solar power plant vegetation coverage reconstruction method according to an embodiment of the present disclosure is described, which can adopt a portable compact disk read-only memory (CD-ROM) and include program code, and can be run on a terminal device, such as a personal computer. However, the program product of the present disclosure is not limited thereto, and in this document, a readable storage medium can be any tangible medium containing or storing a program, which can be used by or in combination with an instruction execution system, an apparatus or a device.

[0129] The program product may adopt any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, a system, device or component of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination thereof. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0130] Computer readable signal media may include data signals propagated in baseband or as part of a carrier wave, in which readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Readable signal media may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0131] The program code contained on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, electromagnetic waves, etc., or any suitable combination of the foregoing.

[0132] Program code for performing the operations of the present disclosure may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., through the Internet using an Internet service provider).

[0133] In addition, the above-mentioned figures are only schematic illustrations of the processes included in the method according to the exemplary embodiments of the present disclosure, and are not intended to be limiting. It is easy to understand that the processes shown in the above-mentioned figures do not indicate or limit the time sequence of these processes. In addition, it is also easy to understand that these processes can be performed synchronously or asynchronously, for example, in multiple modules.

[0134] Through the description of the above embodiments, it is easy for those skilled in the art to understand that the example embodiments described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the embodiment of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a touch terminal, or a network device, etc.) to execute the method according to the embodiment of the present disclosure.

[0135] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary technical means in the art that are not disclosed in the present disclosure. The specification and examples are to be considered as exemplary only, and the true scope and spirit of the present disclosure are indicated by the claims.

[0136] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A method for reconstructing vegetation coverage of a solar power station, characterized in that: include: Acquire a multispectral remote sensing image of a target solar power station and an obstructed surface image corresponding to an obstructed area of ​​a solar energy collection device in the target solar power station; Based on the spectral information in the multispectral remote sensing image, classifying the surface type in the multispectral remote sensing image to identify the blocked area; Determine actual vegetation coverage data corresponding to the blocked area according to the blocked surface image; The image data of the surface type classification of the multispectral remote sensing image is traversed by using a preset convolution kernel; in response to the central pixel of the convolution kernel being in the occlusion area, based on the inverse distance weighted method, according to: Determine the weight corresponding to the non-central pixel of the convolution kernel; wherein, represents the weight corresponding to the non-central pixel, Represents the non-central pixel i The distance from the center pixel, n represents the number of the non-central pixels; according to: Determine the vegetation coverage of the central pixel corresponding to the central pixel; wherein, Indicates the vegetation coverage corresponding to the non-central pixel, r Represents the coverage ratio corresponding to the central pixel; based on the vegetation coverage of the central pixel, the vegetation coverage data corresponding to the multispectral remote sensing image is corrected to obtain the reconstructed vegetation coverage of the solar power station.

2. The method for reconstructing vegetation coverage of a solar power station according to claim 1, characterized in that: The obstructed surface image is obtained by remote-controlled remote sensing image acquisition equipment, and the actual vegetation coverage data corresponding to the obstructed area is determined according to the obstructed surface image, including: Determining a normalized vegetation index corresponding to the obstructing surface image according to image information in the obstructing surface image; The actual vegetation coverage data corresponding to the shielded area is determined according to the normalized vegetation index.

3. The method for reconstructing vegetation coverage of a solar power station according to claim 1, characterized in that: The obstructed surface image is acquired by a handheld image acquisition device, and the obstructed surface image includes a plurality of image pairs consisting of obstructed sampling images and non-obstructed sampling images. Then, determining the actual vegetation coverage data corresponding to the obstructed area according to the obstructed surface image includes: Determine first vegetation coverage data corresponding to the blocked sampling image and second vegetation coverage data corresponding to the unblocked sampling image; Determining a coverage ratio corresponding to each group of the image pairs according to the first vegetation coverage data and the second vegetation coverage data; The coverage ratios are used to perform interpolation processing to obtain coverage ratios corresponding to all areas in the multispectral remote sensing image.

4. The method for reconstructing vegetation coverage of a solar power station according to claim 3, characterized in that: The determining of the first vegetation coverage data corresponding to the blocked sampling image and the second vegetation coverage data corresponding to the non-blocked sampling image includes: In response to the blocked surface image being a remote sensing image, determining the first vegetation coverage data and the second vegetation coverage data according to a ratio of a vegetation projection area to a regional surface area in the remote sensing image; In response to the blocked surface image being a multispectral image, determining a vegetation area according to a normalized vegetation index corresponding to the multispectral image, and determining the first vegetation coverage data and the second vegetation coverage data based on a ratio of the vegetation area to the regional surface area; In response to the obstructing surface image being an RGB image, the first vegetation coverage data and the second vegetation coverage data are determined according to a vegetation index corresponding to the RGB image.

5. The method for reconstructing vegetation coverage of a solar power station according to claim 4, characterized in that: The determining the first vegetation coverage data and the second vegetation coverage data according to the vegetation index corresponding to the RGB image includes: Binarizing the RGB image according to the vegetation index to determine vegetation pixels in the RGB image; The first vegetation coverage data and the second vegetation coverage data are determined according to the ratio of the number of vegetation pixels to the total number of pixels.

6. The method for reconstructing vegetation coverage of a solar power station according to claim 1, characterized in that: The classifying the surface type in the multispectral remote sensing image based on the spectral information in the multispectral remote sensing image comprises: Based on the spectral information in the multispectral remote sensing image, determining normalized vegetation index data and normalized building index data corresponding to the multispectral remote sensing image; The normalized vegetation index data and the normalized building index data are input into a classifier composed of a random forest network, a support vector machine and a K nearest neighbor network to classify the surface type in the multispectral remote sensing image.

7. The method for reconstructing vegetation coverage of a solar power station according to claim 1, characterized in that: Also includes: The surface type of the pixel corresponding to the center pixel of the convolution kernel in the multispectral remote sensing image is changed to a filled pixel, and the entire image is traversed until all photovoltaic panels or solar thermal heliostat mirrors become filled pixels.

8. A device for reconstructing vegetation coverage of a solar power station, characterized in that: include: An image acquisition module, used to acquire a multispectral remote sensing image of a target solar power station and an obstructed surface image corresponding to an obstructed area of ​​a solar energy collection device in the target solar power station; An occlusion determination module, configured to classify the surface type in the multispectral remote sensing image based on the spectral information in the multispectral remote sensing image to identify an occlusion area; A coverage determination module, used to determine actual vegetation coverage data corresponding to the blocked area according to the blocked surface image; The data correction module is used to traverse the image data of the surface type classification of the multispectral remote sensing image using a preset convolution kernel; in response to the central pixel of the convolution kernel being in the occlusion area, based on the inverse distance weighted method, according to: Determine the weight corresponding to the non-central pixel of the convolution kernel; wherein, represents the weight corresponding to the non-central pixel, Represents the non-central pixel i The distance from the center pixel, n represents the number of the non-central pixels; according to: Determine the vegetation coverage of the central pixel corresponding to the central pixel; wherein, Indicates the vegetation coverage corresponding to the non-central pixel, r Represents the coverage ratio corresponding to the central pixel; based on the vegetation coverage of the central pixel, the vegetation coverage data corresponding to the multispectral remote sensing image is corrected to obtain the reconstructed vegetation coverage of the solar power station.

9. An electronic device, characterized in that: include: processor; as well as A memory, configured to store executable instructions of the processor; The processor is configured to execute the method for reconstructing vegetation coverage of a solar power station according to any one of claims 1 to 7 by executing the executable instructions.

Images