A remote sensing identification method for fish-light complementary photovoltaic breeding mode

Abstract

The application discloses a kind of fish-light complementary photovoltaic aquaculture mode remote sensing identification methods, including obtaining the remote sensing image of target area and pre-processing, the spectral feature index and texture feature of the image after pre-processing are calculated, and based on normalized vegetation index, short-wave infrared band texture mean, normalized water body index and normalized red edge index, corresponding to exclude vegetation, building and natural water body, based on difference index and photovoltaic power station feature index water photovoltaic power station pixel recognition result is obtained, determine the contour boundary of water photovoltaic power station and its minimum circumscribed polygon;Determine aquaculture water body polygon by minimum circumscribed polygon equidistant expansion, based on the relationship between the first proportion index of the number of photovoltaic power station pixels in minimum circumscribed polygon relative to its total pixels and the second proportion index of the number of aquaculture water body pixels in aquaculture water body polygon relative to its total pixels, fish-light complementary photovoltaic aquaculture mode is identified.The present application can efficiently and accurately identify fish-light complementary photovoltaic aquaculture mode.

Description

Technical Field

[0001] This invention relates to the field of remote sensing identification of ground features, and more particularly to a remote sensing identification method for a solar-fishery hybrid photovoltaic aquaculture model. background

[0002] "Solar-fishery complementarity" is a novel, eco-friendly, and low-carbon aquaculture model. It combines aquaculture with photovoltaic power generation. Photovoltaic modules are installed above fishponds, while aquaculture can be carried out in the water below, creating a new aquaculture model where "power is generated above, and fish are raised below." Fishermen can utilize fishpond resources to build photovoltaic power stations above the ponds, thus gaining income from both fish farming and photovoltaic power generation—a win-win situation. Effective monitoring of the "solar-fishery complementarity" photovoltaic aquaculture model is crucial for agricultural management departments to formulate control policies. Compared to traditional survey and statistical methods, remote sensing technology offers advantages such as timeliness, efficiency, large-scale monitoring, and low cost, making it an important means of obtaining information on the regional distribution and dynamic changes of photovoltaic aquaculture models.

[0003] Existing technologies include research on the identification of terrestrial photovoltaic power stations, as well as technical solutions for remote sensing identification of aquaculture ponds. In the identification of terrestrial photovoltaic power stations, it is necessary to exclude water surface information through spectral and texture multi-feature images or other technical means. For example, Wang Wei et al. constructed multi-scale multi-feature images by fusing spectral and texture multi-feature images and combining them with multi-scale segmentation images to identify photovoltaic power stations in different background environments in northern Guangdong; Zhou Shufang et al. proposed a two-branch deep learning network that integrates a pixel-by-pixel confidence module for identifying terrestrial photovoltaic power stations in Xinjiang; and the Chinese patent application number 202210094492.X proposes an automatic remote sensing identification method for photovoltaic panels, using the Longyangxia Solar Photovoltaic Power Station in Qinghai Province as the research area, to distinguish water surface information from photovoltaic panel information.

[0004] In identifying aquaculture ponds, it is necessary to extract water surface information through spectral, spatial morphology, and texture features. For example, Wang Fang et al. used Gaofen-1 as the data source and extracted water surface information for four marine aquaculture models (pond culture, cage culture, intertidal culture, and floating raft culture) in the core aquaculture area of ​​Zhelin Bay in eastern Guangdong Province, using the spectral, spatial morphology, and texture features of different aquaculture models and their correlations as transaction data; Ma Yanjuan et al. used the method of constructing indices through band operations to extract near-shore marine aquaculture areas in ASTER images; Liu Zhijun et al. and Xu Shan et al. respectively used object-oriented methods to classify and extract three marine aquaculture models: cage culture, ordinary ponds, and high-density shrimp ponds; Li Ying et al. used OLI image data to propose a method for distinguishing aquaculture water bodies from natural water bodies by integrating spectral and texture information from remote sensing images.

[0005] However, existing technologies have not conducted separate research on floating photovoltaic power stations, nor do they offer remote sensing identification solutions for the novel environmentally friendly aquaculture model of "fishery-solar complementarity." Clearly, the identification technologies for terrestrial photovoltaic power stations and aquaculture ponds cannot be directly applied to remote sensing identification of the "fishery-solar complementarity" aquaculture model.

[0006] The above background information is provided only to assist in understanding the inventive concept and technical solution of this invention. It does not necessarily belong to the prior art of this patent application, nor does it necessarily provide technical teaching. In the absence of clear evidence that the above information was disclosed before the filing date of this patent application, the above background information should not be used to evaluate the novelty and inventiveness of this application. Summary of the Invention

[0007] The purpose of this invention is to provide a remote sensing identification method for a fishery-solar complementary photovoltaic aquaculture mode, which can efficiently and accurately identify a fishery-solar complementary photovoltaic aquaculture mode with a photovoltaic power station above the aquaculture water body.

[0008] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0009] A remote sensing identification method for a fish-solar hybrid aquaculture model is provided for remote sensing identification of a fish-solar hybrid aquaculture model with a photovoltaic power station located above the aquaculture water body. The method includes the following steps:

[0010] Acquire remote sensing images of the target area, including images in the blue, green, red, near-infrared, and short-wave infrared bands;

[0011] The acquired remote sensing images of the target area are preprocessed, including band synthesis, atmospheric correction, geometric correction, cropping, and mosaicking.

[0012] The spectral feature indices and texture features of the preprocessed image are calculated. The spectral feature indices include normalized vegetation index, normalized water index, normalized red edge index, difference index and photovoltaic power station feature index. The texture features include the texture mean of the shortwave infrared band.

[0013] Extracting images of floating photovoltaic power stations and conventional aquaculture water bodies, where conventional aquaculture water bodies are those without floating photovoltaic power stations, includes identifying whether a pixel is a vegetation pixel based on the normalized vegetation index, identifying whether a pixel is a building pixel based on the shortwave infrared band texture mean, identifying whether a pixel is a natural water body pixel based on the normalized water body index and the normalized red edge index, where natural water bodies include rivers and lakes, and identifying the pixels of floating photovoltaic power stations and conventional aquaculture water bodies based on the difference index and photovoltaic power station feature index;

[0014] The identification results of the pixels of the floating photovoltaic power station and the pixels of conventional aquaculture water are classified and processed. The classification and processing includes one or more of Majority analysis, clustering and filtering. The pixels of the floating photovoltaic power station are converted from raster to vector, and the outline boundary of the floating photovoltaic power station and its minimum bounding polygon are determined.

[0015] Based on spatial neighborhood relationships, the minimum bounding polygon is expanded by polygon equidistant expansion to determine the aquaculture water body polygon. The first proportion index of the number of photovoltaic power station pixels in the minimum bounding polygon relative to its total number of pixels, and the second proportion index of the number of aquaculture water body pixels in the aquaculture water body polygon relative to its total number of pixels are calculated. Based on the calculation results of the first proportion index and the second proportion index, the aquaculture-solar complementary photovoltaic aquaculture mode is identified.

[0016] Furthermore, based on any one or a combination of the aforementioned technical solutions, if the first proportion index is greater than or equal to the first proportion index threshold, and the second proportion index is greater than or equal to the second proportion index threshold and less than or equal to the third proportion index threshold, then it is determined that the aquaculture water polygon corresponds to the fish-solar complementary photovoltaic aquaculture mode; otherwise, it is a non-fish-solar complementary photovoltaic aquaculture mode.

[0017] The formula for calculating the first proportion index is:

[0018]

[0019] Among them, K P1 The first proportion index is N1, which is the total number of pixels within the smallest bounding polygon of the outline boundary of the floating photovoltaic power station. P1 The number of photovoltaic power station pixels within the smallest bounding polygon of the outline boundary of the floating photovoltaic power station;

[0020] The formula for calculating the second percentage index is:

[0021]

[0022] Among them, K P2 The first proportion index is N0, which is the total number of pixels within the polygon of the aquaculture water body. P1 This represents the number of photovoltaic power station pixels within the polygon of the aquaculture water body.

[0023] Furthermore, based on any one or a combination of the aforementioned technical solutions, the first percentage index threshold is 85%, and / or the second percentage index threshold is 45%, and / or the third percentage index threshold is 55%.

[0024] Furthermore, following any one or a combination of the aforementioned technical solutions, the normalized vegetation index (NWRI) is configured to exclude vegetation pixels. If the NWRI is less than a preset NWRI threshold, the pixel is determined to be a vegetation pixel; otherwise, the pixel is determined to be a vegetation pixel. The formula for calculating the NWRI is:

[0025]

[0026] Wherein, NDVI is the normalized vegetation index, ρ nir ρ represents the near-infrared pixel reflectance value. red This represents the reflectance value of a pixel in the red light band.

[0027] And / or,

[0028] The shortwave infrared band texture mean is configured to exclude building pixels. If the shortwave infrared band texture mean corresponding to a pixel is less than a preset shortwave infrared band texture mean threshold, the pixel is determined not to be a building pixel; otherwise, the pixel is determined to be a building pixel. The formula for calculating the shortwave infrared band texture mean is as follows:

[0029]

[0030] Among them, Mean swir Let P(i,j) be the mean value of the texture in the shortwave infrared band, P(i,j) be the element in the i-th row and j-th column of the gray-level joint matrix of the image, u be the mean value of P(i,j), and N be the number of pixels.

[0031] Furthermore, following any one or a combination of the aforementioned technical solutions, the normalized water body index and the normalized red edge index are configured to exclude natural water body pixels. If the normalized water body index is less than a preset normalized water body index threshold and the normalized red edge index is less than a preset normalized red edge index threshold, then the pixel is determined to be a natural water body pixel; otherwise, the pixel is determined to be a natural water body pixel.

[0032] The formula for calculating the normalized water index is as follows:

[0033]

[0034] Wherein, NDWI is the normalized water index, ρ green ρ represents the reflectance value of a pixel in the green light band. nir This represents the reflectance value of a pixel in the near-infrared band.

[0035] The formula for calculating the normalized red edge index is as follows:

[0036]

[0037] Where NDGRE is the normalized red-edge exponent, ρ green ρ represents the reflectance value of a pixel in the green light band. re1 This represents the reflectance value of the red-edge band pixels.

[0038] Furthermore, following any one or a combination of the aforementioned technical solutions, the difference index and the photovoltaic power station characteristic index are configured to distinguish between pixels of a floating photovoltaic power station and pixels of aquaculture water bodies. If the difference index is greater than a first difference index threshold and less than a second difference index threshold, and the photovoltaic power station characteristic index is greater than the photovoltaic power station characteristic index threshold, then the pixel is determined to be a photovoltaic power station pixel; otherwise, the pixel is determined to be an aquaculture water body pixel.

[0039] The formula for calculating the difference index is as follows:

[0040] DVI = ρ blue -ρ green

[0041] Where DVI is the difference index, ρ blue ρ represents the reflectance value of a pixel in the blue light band. green This represents the reflectance value of a pixel in the green light band.

[0042] The formula for calculating the characteristic index of the photovoltaic power station is as follows:

[0043] PSI=(ρ blue +ρ swir )-(ρ red +ρ nir )

[0044] Wherein, PSI is the characteristic index of photovoltaic power plants, ρ blue ρ represents the reflectance value of a pixel in the blue light band. swir ρ represents the reflectance value of a pixel in the shortwave infrared band. nir This represents the reflectance value of a pixel in the near-infrared band.

[0045] Furthermore, following any one or a combination of the aforementioned technical solutions, the normalized vegetation index threshold is 0.05; and / or, the shortwave infrared band texture mean threshold is 20.

[0046] Furthermore, following any one or a combination of the aforementioned technical solutions, the normalized water index threshold is 0; and / or,

[0047] The normalized red-edge index threshold is 0.1.

[0048] Furthermore, following any one or a combination of the aforementioned technical solutions, the first difference index threshold is -150; and / or,

[0049] The second difference index threshold is 200; and / or,

[0050] The characteristic index threshold of the photovoltaic power station is 400.

[0051] Furthermore, based on any one or a combination of the aforementioned technical solutions, the minimum circumscribed polygon of the outline boundary of the floating photovoltaic power station is determined to be a rectangle;

[0052] The remote sensing images of the target area were acquired in March.

[0053] The beneficial effects of the technical solution provided by this invention are as follows:

[0054] a. This invention fully utilizes the unique differences in reflectivity of floating photovoltaic power stations in the blue light band, green light band, and short-wave infrared band to construct a characteristic index and a difference index for photovoltaic power stations, thereby effectively distinguishing floating photovoltaic power stations from conventional aquaculture water bodies;

[0055] b. This invention is based on the spatial pixel ratio relationship between a floating photovoltaic power station and the surrounding aquaculture water. Specifically, it calculates the first ratio index of the number of photovoltaic power station pixels within the smallest bounding polygon of the floating photovoltaic power station's outline boundary relative to its total number of pixels, and the second ratio index of the number of aquaculture water pixels within the polygon of the surrounding aquaculture water relative to its total number of pixels. Based on the range of the first and second ratio indices, it determines the aquaculture-solar complementary photovoltaic aquaculture mode, achieving efficient and accurate identification of the aquaculture-solar complementary photovoltaic aquaculture mode and filling the gap in current algorithms. Attached Figure Description

[0056] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0057] Figure 1 A flowchart illustrating a remote sensing identification method for a solar-fish integrated aquaculture model, provided as an exemplary embodiment of the present invention;

[0058] Figure 2 A schematic diagram of a solar-aquaculture hybrid farming model provided as an exemplary embodiment of the present invention;

[0059] Figure 3 A schematic diagram of a floating photovoltaic power station for remote sensing monitoring, provided as an exemplary embodiment of the present invention;

[0060] Figure 4This is a schematic diagram of a remote sensing-monitored photovoltaic aquaculture model with complementary fish and solar power, compared to a conventional aquaculture pond, provided as an exemplary embodiment of the present invention. Detailed Implementation

[0061] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0062] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, apparatus, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0063] This invention proposes a remote sensing identification method for aquaculture-solar hybrid photovoltaic aquaculture patterns based on spectral features, texture features, and spatial correlation features. This method is used to remotely identify aquaculture-solar hybrid photovoltaic aquaculture patterns with photovoltaic power stations located above the aquaculture water body. First, buildings, vegetation, and natural rivers are excluded. Then, the method identifies the photovoltaic power station on the water and the conventional aquaculture water body based on spectral and texture features. Finally, the aquaculture-solar hybrid photovoltaic aquaculture pattern is determined based on the spatial relationship between the photovoltaic power station and the aquaculture pond water body.

[0064] In one embodiment of the present invention, see Figure 1 This paper provides a remote sensing identification method for a solar-fish integrated aquaculture model, including the following steps:

[0065] Remote sensing images of the target area were acquired, including images in the blue, green, red, near-infrared, and shortwave infrared bands. The images were taken in March, a period when the pond water was abundant, the boundaries were clear, and vegetation information was sparse. (See also...) Figure 2 In this embodiment, the target area is photovoltaic aquaculture in Ninghai County, Zhejiang Province, and the data source is the Level-2A Sentinel-2 image after atmospheric correction. The image time is March 11, 2022.

[0066] The acquired remote sensing images of the target area are preprocessed, including band compositing of the remote sensing images, with composite bands including blue, green, red, red edge, near-infrared (NIR), and shortwave infrared (SWIR); and atmospheric correction, geometric correction, cropping, and mosaicking are performed on the composite images.

[0067] The spectral feature indices and texture features of the preprocessed image are calculated. The spectral feature indices include Normalized Difference Vegetation Index (NDVI), Normalized Difference Water Index (NDWI), Normalized Red Edge Index (NDGRE), Difference Index (DVI), and Photovoltaic Power Plant Feature Index (PSI). The texture features include the shortwave infrared band texture mean.

[0068] The normalized vegetation index (NVI) is used to exclude vegetation. The NVI is defined as the ratio of the difference in reflectance between the near-infrared band and the red band to their sum, and its calculation formula is as follows:

[0069]

[0070] Wherein, NDVI is the normalized vegetation index, ρ nir ρ represents the near-infrared pixel reflectance value. red This represents the reflectance value of a pixel in the red light band.

[0071] The normalized water body index and the normalized red edge index are used to exclude natural water bodies such as rivers and lakes. The natural water bodies include rivers and lakes. The normalized water body index is defined as the ratio of the difference between the reflectance of the green light band and the near-infrared band to the sum of the two, and its calculation formula is as follows:

[0072]

[0073] Wherein, NDWI is the normalized water index, ρ green ρ represents the reflectance value of a pixel in the green light band. nir This represents the reflectance value of a pixel in the near-infrared band.

[0074] The formula for calculating the normalized red edge index is as follows:

[0075]

[0076] Where NDGRE is the normalized red-edge exponent, ρ green ρ represents the reflectance value of a pixel in the green light band. re1 This represents the reflectance value of the red-edge band pixels.

[0077] Based on the characteristic that floating photovoltaic power stations have higher reflectivity in the blue light band than in the green light band, while the reflectivity of conventional aquaculture water bodies (such as ordinary aquaculture water bodies) is the opposite, a difference index is constructed. In this invention, conventional aquaculture water bodies refer to aquaculture water bodies without floating photovoltaic power stations. The calculation formula for the difference index is as follows:

[0078] DVI = ρ blue -ρ green

[0079] Where DVI is the difference index, ρ blue ρ represents the reflectance value of a pixel in the blue light band. green This represents the reflectance value of a pixel in the green light band.

[0080] Based on the characteristic that the reflectivity of floating photovoltaic power stations shows peak values ​​in the blue light and short-wave infrared bands, significantly higher than that of conventional aquaculture water bodies, a photovoltaic power station characteristic index is constructed. The calculation formula for the photovoltaic power station characteristic index is as follows:

[0081] PSI=(ρ blue +ρ swir )-(ρ red +ρ nir )

[0082] Wherein, PSI is the characteristic index of photovoltaic power plants, ρ blue ρ represents the reflectance value of a pixel in the blue light band. swir ρ represents the reflectance value of a pixel in the shortwave infrared band. nir This represents the reflectance value of a pixel in the near-infrared band.

[0083] The shortwave infrared band texture mean is used to extract building information. Its calculation window size is 3×3, and its calculation formula is as follows:

[0084]

[0085] Among them, Mean swir Let P(i,j) be the mean value of the texture in the shortwave infrared band, P(i,j) be the element in the i-th row and j-th column of the gray-level joint matrix of the image, u be the mean value of P(i,j), and N be the number of pixels.

[0086] Further, the extraction of floating photovoltaic power stations and conventional aquaculture water bodies based on the spectral feature index and texture features includes: identifying whether a pixel is a vegetation pixel based on the normalized vegetation index (NDVI); if the NDVI is less than a preset NDVI threshold, the pixel is determined not to be a vegetation pixel; otherwise, the pixel is determined to be a vegetation pixel. Preferably, the NDVI threshold is 0.05. That is, if NDVI < 0.05, it is a non-plant pixel, and the pixel is further identified; otherwise, the pixel is identified as a plant pixel and excluded.

[0087] Further, based on the mean short-wave infrared band texture, other pixels of non-vegetation pixels are identified as building pixels. The mean short-wave infrared band texture is configured to exclude building pixels. If the mean short-wave infrared band texture corresponding to a pixel is less than a preset threshold, the pixel is determined not to be a building pixel; otherwise, the pixel is determined to be a building pixel. Preferably, the threshold for the mean short-wave infrared band texture is 20. That is, if Mean swir If the value is less than 20, it is a non-building pixel, and the pixel is further identified; otherwise, the pixel is identified as a non-building pixel and excluded.

[0088] Furthermore, based on the Normalized Water Index (NDWI) and Normalized Red Edge Index (NDGRE), other pixels besides non-vegetation and building pixels are identified as natural water body pixels. If the NDWI is less than a preset NDWI threshold and the NDGRE is less than a preset NDGRE threshold, the pixel is determined not to be a natural water body pixel; otherwise, the pixel is determined to be a natural water body pixel. Preferably, the NDWI threshold is 0, and the NDGRE threshold is 0.1. That is, if NDWI < 0 and NDGRE < 0.1, it is a non-natural water body, and the pixel is further identified; otherwise, the pixel is identified as a natural water body and excluded.

[0089] It should be noted that the optimal identification efficiency is achieved by sequentially excluding vegetation pixels, building pixels, and natural water body pixels in the above identification order, followed by the identification of floating photovoltaic power station pixels and conventional aquaculture water body pixels. This invention is not limited to this specific order; other reasonable identification orders also fall within the scope of this application. For example, it is also feasible to exclude buildings first, then vegetation and natural water bodies; or to exclude natural water bodies first, then vegetation and buildings.

[0090] Since the reflectance of the floating PV power station peaks in the blue light band and the short-wave infrared band, which is significantly higher than that of conventional aquaculture water bodies. And the reflectance in the blue light band is higher than that in the green light band, while the reflectance of conventional aquaculture water bodies is opposite to this. These characteristics are the main basis for the identification of PV power stations. The conventional aquaculture water body is an aquaculture water body without a floating PV power station. Therefore, the floating PV power station pixels and the conventional aquaculture water body pixels can be identified based on the difference index and the PV power station characteristic index, and the identification results of the floating PV power station pixels and the conventional aquaculture water body pixels can be obtained. If the difference index is greater than the first difference index threshold and less than the second difference index threshold, and the PV power station characteristic index is greater than the PV power station characteristic index threshold, then the pixel is determined to be a PV power station pixel; otherwise, the pixel is determined to be a conventional aquaculture water body pixel. Preferably, the first difference index threshold is -150; the second difference index threshold is 200; the PV power station characteristic index threshold is 400. That is, if -150 < DVI < 200 and NDGRE > 400, then the pixel is determined to be a floating PV power station pixel, see Figure 3 ; otherwise, the pixel is determined to be a conventional aquaculture water body. The present invention makes full use of the unique differences in the reflectance of the floating PV power station in the blue light band, green light band, and short-wave infrared band, constructs the PV power station characteristic index and the difference index, and effectively distinguishes the floating PV power station from the conventional aquaculture water body.

[0091] Further, after-classification processing is performed on the identification results of the floating PV power station pixels and the conventional aquaculture water body pixels. The after-classification processing includes one or more of Majority analysis, clustering processing, and filtering processing. Through the above operations, small and fragmented spots can be removed to obtain a relatively regular grid result. Convert the floating PV power station pixel grid to a vector to determine the contour boundary A of the floating PV power station; and use the feature management tool to determine its minimum bounding polygon A1. Preferably, the minimum bounding polygon for determining the contour boundary of the floating PV power station is a rectangle.

[0092] Further, based on the spatial neighborhood relationship, the minimum bounding polygon A1 is expanded equidistantly to determine the aquaculture water body polygon A2. Preferably, the minimum bounding polygon A1 is expanded equidistantly by 10 meters. And calculate the first occupancy ratio index of the number of floating PV power station pixels in the minimum bounding polygon A1 relative to its total number of pixels, and the second occupancy ratio index of the number of aquaculture water body pixels in the aquaculture water body polygon A2 relative to its total number of pixels, and identify the fishery-solar complementary photovoltaic aquaculture mode based on the calculation results. Among them, the calculation formula for the first occupancy ratio index is:

[0093]

[0094] Among them, K P1The first proportion index is N1, which is the total number of pixels within the smallest bounding polygon of the outline boundary of the floating photovoltaic power station. P1 The number of photovoltaic power station pixels within the smallest bounding polygon of the outline boundary of the floating photovoltaic power station;

[0095] The formula for calculating the second percentage index is:

[0096]

[0097] Among them, K P2 The first proportion index is N0, which is the total number of pixels within the polygon of the aquaculture water body. P2 This represents the number of aquaculture water pixels within the polygon of the aquaculture water body.

[0098] If the first proportion index is greater than or equal to the first proportion index threshold, and the second proportion index is greater than or equal to the second proportion index threshold and less than or equal to the third proportion index threshold, then the aquaculture water polygon is determined to correspond to a fish-solar complementary photovoltaic aquaculture mode; otherwise, it is a non-fish-solar complementary photovoltaic aquaculture mode. Preferably, the first proportion index threshold is 85%, the second proportion index threshold is 45%, and the third proportion index threshold is 55%. That is, if K is satisfied... P1 ≥85% and 45%≤K P2 If the percentage is ≤55%, it falls under the category of solar-aquaculture hybrid farming; otherwise, it falls under other non-solar-aquaculture hybrid farming models. See [link / reference]. Figure 4 This invention determines the solar-aquaculture complementary farming model based on the spatial pixel ratio between the floating photovoltaic power station and the surrounding aquaculture water, i.e., the range of the first ratio index and the second ratio index. This achieves efficient and accurate identification of the solar-aquaculture complementary farming model, filling the gap in current algorithms.

[0099] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0100] The above description is only a specific embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A remote sensing identification method for a fish-light complementary photovoltaic aquaculture mode, for remotely sensing and identifying a fish-light complementary photovoltaic aquaculture mode in which a photovoltaic power station is arranged above a culture water body, characterized in that, The method comprises the following steps: acquiring remote sensing images of a target area, the remote sensing images comprising images of blue light, green light, red light, near-infrared, and short-wave infrared bands; preprocessing the acquired remote sensing images of the target area, the preprocessing comprising band synthesis, atmospheric correction, geometric correction, cropping, and mosaicking of the remote sensing images; calculating spectral feature indexes and texture features of the preprocessed images, the spectral feature indexes comprising normalized vegetation index, normalized water index, normalized red edge index, difference index, and photovoltaic power station feature index, and the texture features comprising short-wave infrared band texture mean value; extracting water-based photovoltaic power stations and conventional aquaculture water bodies, the conventional aquaculture water bodies being aquaculture water bodies without water-based photovoltaic power stations, comprising identifying whether a pixel is a vegetation pixel based on the normalized vegetation index, identifying whether a pixel is a building pixel based on the short-wave infrared band texture mean value, identifying whether a pixel is a natural water body pixel based on the normalized water index and the normalized red edge index, the natural water body comprising rivers and lakes, and identifying water-based photovoltaic power station pixels and conventional aquaculture water body pixels based on the difference index and the photovoltaic power station feature index; performing classification post-processing on the identification results of the water-based photovoltaic power station pixels and the conventional aquaculture water body pixels, the classification post-processing comprising one or more of Majority analysis, clustering processing, and filtering processing, converting the water-based photovoltaic power station pixel raster into a vector, and determining the contour boundary of the water-based photovoltaic power station and its minimum circumscribed polygon; based on spatial neighborhood relations, performing polygon equidistant expansion on the minimum circumscribed polygon to determine an aquaculture water body polygon, calculating a first proportion index of the number of photovoltaic power station pixels within the minimum circumscribed polygon relative to the total number of pixels thereof, and a second proportion index of the number of aquaculture water body pixels within the aquaculture water body polygon relative to the total number of pixels thereof, and identifying a fish-light complementary photovoltaic aquaculture mode based on the calculation results of the first proportion index and the second proportion index; the calculation formula of the normalized red edge index is: ； Wherein, NDGRE is the normalized red edge index, ρ green is the green light band pixel reflectivity value, ρ re1 is the red edge band pixel reflectivity value; the difference index and the photovoltaic power station feature index are configured to distinguish water-based photovoltaic power station pixels from aquaculture water body pixels, if the difference index is greater than a first difference index threshold and less than a second difference index threshold, and the photovoltaic power station feature index is greater than a photovoltaic power station feature index threshold, then the pixel is determined to be a photovoltaic power station pixel, otherwise the pixel is determined to be an aquaculture water body pixel; the calculation formula of the difference index is: ； Wherein, DVI is the difference index, p blue is the blue light band pixel reflectivity value, p green is the green light band pixel reflectivity value; the calculation formula of the photovoltaic power station feature index is: PSI=( )-( ) Wherein, PSI is the photovoltaic power station characteristic index, ρ blue is the blue light band pixel reflectivity value, ρ swir is the short-wave infrared band pixel reflectivity value, ρ nir is the near-infrared band pixel reflectivity value, ρ red is the red light band pixel reflectivity value. 2.The method according to claim 1, wherein, if the first proportion index is greater than or equal to a first proportion index threshold, and the second proportion index is greater than or equal to a second proportion index threshold and less than or equal to a third proportion index threshold, then the aquaculture water body polygon corresponds to a fish-light complementary photovoltaic aquaculture mode, otherwise it is a non-fish-light complementary photovoltaic aquaculture mode; the calculation formula of the first proportion index is: *100%； Wherein, K P1 is the first proportion index, N1 is the total number of pixels within the minimum circumscribed polygon of the profile boundary of the water-based photovoltaic power station, N P1 is the number of photovoltaic power station pixels within the minimum circumscribed polygon of the profile boundary of the water-based photovoltaic power station. the calculation formula of the second proportion index is: *100%； Wherein, K P2 is the first proportion index, N0 is the total number of pixels in the polygon of the aquaculture water body, and N P1 is the number of aquaculture water body pixels in the polygon of the aquaculture water body. 3.The method according to claim 2, wherein, the first proportion index threshold is 85%, and / or the second proportion index threshold is 45%, and / or the third proportion index threshold is 55%. 4.The method of claim 1, wherein, The normalized vegetation index is configured to exclude vegetation pixels, if the normalized vegetation index is less than a preset normalized vegetation index threshold, it is determined that the pixel is not a vegetation pixel, otherwise it is determined that the pixel is a vegetation pixel; the calculation formula of the normalized vegetation index is: ； Wherein, NDVI is the normalized vegetation index, ρ nir is the near-infrared band pixel reflectivity value, ρ red is the red light band pixel reflectivity value; And / or, The short-wave infrared band texture mean value is configured to exclude building pixels, if the short-wave infrared band texture mean value corresponding to the pixel is less than a preset short-wave infrared band texture mean value threshold, it is determined that the pixel is not a building pixel, otherwise it is determined that the pixel is a building pixel; the calculation formula of the short-wave infrared band texture mean value is: ； where Mean swir is the short-wave infrared band texture mean, P(i,j) represents the element of the i row j column in the gray level joint matrix of the image, u represents the mean of P(i,j), and N is the number of pixels. 5.The method of claim 1, wherein, The normalized water body index and the normalized red edge index are configured to exclude natural water body pixels, if the normalized water body index is less than a preset normalized water body index threshold and the normalized red edge index is less than a preset normalized red edge index threshold, it is determined that the pixel is not a natural water body pixel, otherwise it is determined that the pixel is a natural water body pixel; The calculation formula of the normalized water body index is: ； Wherein, NDWI is the normalized water body index, ρ green is the green light band pixel reflectivity value, ρ nir is the near-infrared band pixel reflectivity value. 6.The method of claim 4, wherein the method further comprises: determining the location of the fish farming area based on the obtained image data. The normalized vegetation index threshold is 0.05; and / or, the short-wave infrared band texture mean value threshold is 20. 7.The method of claim 5, wherein the method further comprises: determining the location of the fish farming area based on the received image data. The normalized water body index threshold is 0; and / or, The normalized red edge index threshold is 0.

1. 8.The method of claim 1, wherein the method further comprises: determining a distance between the fish and the light source based on the first and second images. The first difference index threshold is -150; and / or, The second difference index threshold is 200; and / or, The photovoltaic power station feature index threshold is 400. 9.The method of claim 1, wherein, The minimum circumscribed polygon of the water-based photovoltaic power station contour boundary is a rectangle; and / or, The time period for acquiring the remote sensing image of the target area is March.

Images