Aquatic plant detection method, device, equipment and medium

Abstract

The invention relates to the technical field of water resources, and discloses an aquatic plant detection method, device and equipment and a medium. The method comprises the following steps: acquiring optical data, radar data and ground data of a target area; based on a first aquatic vegetation region corresponding to the optical data, a second aquatic vegetation region corresponding to the radar data and a third aquatic vegetation region corresponding to the ground data, performing spatial registration on the optical data, the radar data and the ground data to obtain a first registration data set; and detecting the aquatic plants in the target area based on the first registration data set to obtain a detection result. According to the invention, through fusion of multi-source data (optical data, radar data and ground data), the precision and reliability of aquatic vegetation detection are significantly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of water resources, and particularly to a detection method, device, equipment and medium for aquatic plants. Background Art

[0002] Due to its high transparency, suitable water temperature and rich nutrients, the water body of lakes and reservoirs provides an ideal environment for the growth of large aquatic plants. These aquatic plants play an important role in maintaining the ecological balance of the water body, improving water quality and providing biological habitats. However, in some specific areas, especially the upper reaches of rivers, the overgrowth of floating plants and submerged plants may cause serious ecological and engineering problems. For example, overgrown aquatic plants often gather near the intake of hydropower stations, causing blockage of the intake. Such blockage will not only reduce the power generation efficiency of the hydropower station, but also pose a threat to the normal operation and safety of the hydropower station. Therefore, it is particularly important to accurately detect the growth of aquatic plants to provide a scientific basis for the operation of hydropower stations. Summary of the Invention

[0003] In view of this, the present invention provides a detection method, device, equipment and medium for aquatic plants to accurately detect the growth of aquatic plants.

[0004] In a first aspect, the present invention provides a detection method for aquatic plants, the method comprising:

[0005] Obtaining optical data, radar data and ground data of a target area;

[0006] Performing spatial registration on the optical data, radar data and ground data based on a first aquatic vegetation area corresponding to the optical data, a second aquatic vegetation area corresponding to the radar data, and a third aquatic vegetation area corresponding to the ground data to obtain a first registered data set;

[0007] Detecting aquatic plants in the target area based on the first registered data set to obtain a detection result.

[0008] Through the method provided by this embodiment, optical data, radar data, and ground data are integrated to detect aquatic plants using multi-source data, overcoming the limitations of a single data source. Among them, optical data provides rich spectral information, radar data has all-weather observation capabilities, and ground data provides high-precision reference information. The embodiment of the present application significantly improves the accuracy and reliability of aquatic vegetation detection through the fusion of multi-source data. In addition, before detecting aquatic plants in the target area, spatial registration of optical data, radar data, and ground data can accurately correspond the information in different data sources to the same geographic spatial coordinate system. In this way, when detecting aquatic plants, the information at the same location in different data sources can be comprehensively analyzed to avoid misjudgment and missed judgment caused by inconsistent spatial positions of the data, effectively solving the problem of inconsistency caused by differences in resolution and imaging mechanisms of multi-source data, providing a high-quality data basis for subsequent detection of aquatic plants, and improving the accuracy of aquatic plant detection.

[0009] In an alternative embodiment, based on the first aquatic vegetation area corresponding to the optical data, the second aquatic vegetation area corresponding to the radar data, and the third aquatic vegetation area corresponding to the ground data, spatial registration of the optical data, radar data, and ground data is performed to obtain a first registered data set, including:

[0010] Based on the first aquatic vegetation area, the second aquatic vegetation area, and the third aquatic vegetation area, spatial registration of the optical data, radar data, and ground data is performed to obtain a second registered data set;

[0011] Based on the water flow data in the target area, the second registered data set is adjusted to obtain a first registered data set.

[0012] Through the above embodiment, the water flow data can better reflect the hydrodynamic characteristics of the target area, especially for areas with significant dynamic changes in water flow such as the upper reaches of the river. Using the water flow data to adjust the second registered data can effectively reduce the distribution error of aquatic vegetation caused by water flow movement, making the detection results more in line with the actual environment.

[0013] In an alternative embodiment, based on the water flow data in the target area, the second registered data set is adjusted to obtain a first registered data set, including:

[0014] The water flow data is input into a pre-constructed first prediction model to obtain the predicted spatial position of the target object in the target area;

[0015] Based on the predicted spatial position, the second registered data set is adjusted to obtain a first registered data set.

[0016] Through the above-described embodiments, by introducing water flow data and utilizing a pre-constructed first prediction model, it is possible to accurately predict the spatial position changes of target objects (such as floating aquatic plants and submerged aquatic plants) under the action of water flow. At the same time, based on the second registration dataset, adjustments are made in combination with the predicted spatial positions, further eliminating data biases caused by dynamic changes in water flow, making the fusion of optical data, radar data, and ground data more accurate, and providing a basis for improving the accuracy of aquatic vegetation detection.

[0017] In an alternative embodiment, based on the first aquatic vegetation area, the second aquatic vegetation area, and the third aquatic vegetation area, spatial registration is performed on the optical data, radar data, and ground data to obtain a second registration dataset, including:

[0018] Using the pyramid scale feature matching method and / or the dense manifold matching method, spatial registration is performed on the optical data, radar data, and ground data to obtain a third registration dataset;

[0019] Based on the first aquatic vegetation area, the second aquatic vegetation area, and the third aquatic vegetation area, registration is performed on the third registration dataset to obtain a second registration dataset.

[0020] Through the above-described embodiments, the pyramid scale feature matching method can capture the feature information of the light source data and radar data at different scales and is suitable for spatial registration of data with large differences in resolution. The dense manifold matching method can handle complex non-linear feature alignment problems and is particularly suitable for feature differences caused by different imaging mechanisms between optical data, radar data, and ground data, especially in areas with complex terrain and vegetation distribution. In the embodiments of the present application, through multi-level and multi-method registration optimization, the obtained second registration dataset has higher spatial accuracy and consistency, providing a reliable data basis for subsequent aquatic vegetation detection.

[0021] In an alternative embodiment, based on the first registration dataset, aquatic plants in the target area are detected to obtain detection results, including:

[0022] Determine the optical features, radar features, and ground measurement features in the first registration dataset; the optical features include vegetation indices; the radar features include backscattering coefficients and / or polarization features; the ground measurement features include biomass and / or chlorophyll concentration;

[0023] Input the optical features, radar features, and ground measurement features into a pre-constructed second prediction model to obtain detection results.

[0024] Through the above embodiments, vegetation indices (such as NDVI, EVI) in optical features can effectively reflect the growth status and coverage of aquatic plants, providing important spectral information for the detection of aquatic plants. Radar features (such as backscattering coefficient, polarization features) can penetrate clouds and provide all-weather and all-time observation data, which are particularly suitable for complex environments with limited optical data (such as cloudy weather). Ground measurement features (such as biomass, chlorophyll concentration) provide high-precision reference information for the detection results, further ensuring the accuracy of the detection results. Combining optical features (such as vegetation indices), radar features (such as backscattering coefficient, polarization features) and ground measurement features (such as biomass, chlorophyll concentration) makes full use of the complementary advantages of multi-source data and provides support for improving the accuracy and reliability of aquatic plant detection.

[0025] In an alternative embodiment, the method further includes:

[0026] Based on the optical data, determine the vegetation information of the target area; the vegetation information is used to characterize the aquatic plants in the target area;

[0027] Based on the vegetation information, determine the first aquatic vegetation area.

[0028] Through the above embodiments, optical data (such as multispectral and hyperspectral images) can provide rich spectral information. By analyzing features such as vegetation indices (such as NDVI, EVI), the vegetation information of the target area can be extracted efficiently and accurately. Further, based on the vegetation information, the distribution range of aquatic plants in the target area can be quickly identified and located, so as to accurately determine the first aquatic vegetation area.

[0029] In an alternative embodiment, the method further includes:

[0030] Based on the radar data, determine the water body information of the target area; the water body information is used to indicate the water body area in the target area;

[0031] Based on the water body information, determine the second aquatic vegetation area.

[0032] Through the above embodiments, radar data (such as synthetic aperture radar SAR) has all-weather and all-time observation capabilities, can penetrate clouds and some vegetation, and can quickly and accurately extract the water body information of the target area. Further, based on the water body information of the radar data, the water body range in the target area can be clearly identified, that is, the second aquatic vegetation area is determined.

[0033] In a second aspect, the present invention provides a detection device for aquatic plants, the device includes:

[0034] An acquisition module, configured to acquire optical data, radar data and ground data of the target area;

[0035] A registration module, configured to perform spatial registration on optical data, radar data, and ground data based on a first aquatic vegetation area corresponding to the optical data, a second aquatic vegetation area corresponding to the radar data, and a third aquatic vegetation area corresponding to the ground data, so as to obtain a first registered data set;

[0036] A detection module, configured to detect aquatic plants in a target area based on the first registered data set to obtain a detection result.

[0037] With the above device, the optical data, radar data, and ground data are integrated, overcoming the limitations of a single data source. Among them, the optical data provides rich spectral information, the radar data has all-weather observation capabilities, and the ground data provides high-precision reference information. The embodiments of the present application significantly improve the accuracy and reliability of aquatic vegetation detection through the fusion of multi-source data. In addition, by performing spatial registration on the optical data, radar data, and ground data, the information in different data sources can be accurately mapped to the same geographic spatial coordinate system. In this way, when detecting aquatic plants, the information at the same location in different data sources can be comprehensively analyzed, avoiding misjudgment and missed judgment caused by inconsistent data spatial positions, effectively solving the problem of inconsistency caused by differences in resolution and imaging mechanisms of multi-source data, and providing a high-quality data basis for subsequent analysis.

[0038] In a third aspect, the present invention provides a computer device, including: a memory and a processor, which are communicatively connected to each other. The memory stores computer instructions, and the processor executes the computer instructions to execute the detection method of aquatic plants according to the first aspect or any corresponding embodiment thereof.

[0039] In a fourth aspect, the present invention provides a computer-readable storage medium, on which computer instructions are stored, and the computer instructions are used to cause a computer to execute the detection method of aquatic plants according to the first aspect or any corresponding embodiment thereof.

[0040] In a fifth aspect, the present invention provides a computer program product, including computer instructions, and the computer instructions are used to cause a computer to execute the detection method of aquatic plants according to the first aspect or any corresponding embodiment thereof. Description of the Drawings

[0041] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0042] Figure 1 It is a schematic flow chart of a detection method for aquatic plants according to an embodiment of the present invention;

[0043] Figure 2 It is a schematic flow chart of another detection method for aquatic plants according to an embodiment of the present invention;

[0044] Figure 3 It is a structural block diagram of a detection device for aquatic plants according to an embodiment of the present invention;

[0045] Figure 4 It is a schematic hardware structure diagram of a computer device according to an embodiment of the present invention. Specific embodiments

[0046] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0047] First, an exemplary introduction to the application scenarios of the embodiments of the present application is given.

[0048] Due to its high transparency, suitable water temperature, and rich nutrients, the water body of lakes and reservoirs provides an ideal environment for the growth of large aquatic plants. According to the growth form, aquatic plants are mainly divided into three categories: submerged plants, emergent plants, and floating plants. However, in the upper reaches of the river, the overgrowth of aquatic plants often causes blockage problems at the intake of hydropower stations, seriously affecting the power generation efficiency and normal operation of hydropower stations, and even potentially causing safety hazards. Therefore, accurately detecting the growth distribution and dynamic changes of aquatic plants has important scientific significance and practical application value for preventing blockage at the intake of hydropower stations, optimizing operation management, and ensuring the safety of power supply.

[0049] In view of this, an embodiment of the present application provides a detection method for aquatic plants to achieve accurate detection of aquatic plants. It should be noted that for the detection method of aquatic plants provided by the embodiment of the present invention, the execution subject may be a detection device for aquatic plants, and this detection device for aquatic plants can be implemented as part or all of an electronic device through software, hardware, or a combination of software and hardware. Among them, the electronic device can be a server or a terminal. Among them, the server in the embodiment of the present application can be a single server or a server cluster composed of multiple servers, and the terminal in the embodiment of the present application can be other intelligent hardware devices such as a smart phone, a personal computer, a tablet computer, a wearable device, and a smart robot. In the following method embodiments, the execution subject is taken as an electronic device for illustration.

[0050] According to an embodiment of the present invention, an embodiment of a detection method for aquatic plants is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. And although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0051] In this embodiment, a detection method for aquatic plants is provided, which can be used for the above-mentioned electronic devices such as servers. Figure 1 It is a flowchart of a detection method for aquatic plants according to an embodiment of the present invention, as Figure 1 shown, and this process includes the following S101-S103:

[0052] S101: Obtain optical data, radar data, and ground data of the target area.

[0053] Specifically, optical data refers to remote sensing data on the reflection characteristics of the earth's surface to solar radiation obtained through optical sensors (such as satellites, drones, ground cameras, etc.). These data are presented in the form of images or videos and contain rich information such as the color, texture, and shape of surface objects. In the embodiment of the present application, the optical data can be the reflection values in the near-infrared band, the red light band, the green light band, etc., which can reflect the spectral characteristics of surface objects. Exemplarily, the optical data can be obtained through Sentinel-2, Landsat, etc.

[0054] Specifically, radar data is obtained by a radar system transmitting electromagnetic waves and receiving their echoes to obtain information on the target area. Radar data not only contains parameters such as the distance, speed, and direction of the target, but also can reflect the physical characteristics of surface objects, such as roughness and water content. Exemplarily, radar data can be obtained through synthetic aperture radar (SAR), airborne radar, ground radar, etc.

[0055] Specifically, ground data is data obtained through ground observations, measurements, or sampling, including water color parameters (such as chlorophyll concentration), water body transparency, environmental data (such as temperature, wind speed, rainfall), and biomass measurement data, etc. These data usually have high accuracy and reliability and are important means to supplement remote sensing data and radar data. Exemplarily, ground data is obtained by using measurement tools. To improve the prediction accuracy, the sampling point density of ground data can be increased to enrich the data volume for predicting aquatic plants. It should be noted that the embodiments of the present application do not limit the acquisition methods of optical data, radar data, and ground data.

[0056] In a possible implementation, the optical data, radar data, and ground data are respectively data after data preprocessing.

[0057] Optionally, for optical data, data preprocessing includes but is not limited to radiometric correction, atmospheric correction, image de-clouding, image fusion, etc. Among them, radiometric correction refers to converting the original digital values of the image into radiance or reflectance with physical meaning, eliminating the errors of the sensor itself and the influence of illumination conditions. The purpose of atmospheric correction is to eliminate the influence of atmospheric scattering, absorption, etc. on the image and obtain the true reflectance of the ground object. Especially in the water body environment, due to its strong specular reflection characteristics on the surface, the sun glint effect is easily generated. At the same time, the atmospheric aerosol concentration and water vapor content above the water body may also be different from those in the land area, further affecting the atmospheric scattering characteristics. Therefore, when performing atmospheric correction in the water body environment, special consideration needs to be given to the interference of the combined action of water surface reflection and atmospheric molecular scattering on the image radiation transfer process to obtain more accurate reflectance information. Exemplarily, physical models (such as the Second Simulation of the Satellite Signal in the Solar Spectrum (6S) model, Sentinel-2 Correction (Sen2Cor)) are used to perform atmospheric correction on optical data. Image de-clouding refers to removing the interference of clouds and their shadows on the image to ensure the integrity of surface information. Exemplarily, the Function of mask (Fmask) algorithm can be used to effectively identify and remove clouds and their shadows. Image fusion refers to fusing optical images with different resolutions to generate an image with both high spatial resolution and high spectral resolution. Exemplarily, the Brovey Transform, Principal Component Analysis (PCA), etc. can be used to improve the spatio-temporal resolution of optical data.

[0058] Optionally, for radar data, data preprocessing includes, but is not limited to, radiometric correction, terrain correction, polarization decomposition, etc. Among them, radiometric correction is used to eliminate radiometric distortion in radar images, ensuring that the radiometric values of the images can accurately reflect the backscattering characteristics of ground objects. Exemplarily, filtering methods such as Gamma Maximum A Posteriori (Gamma MAP) can be used to remove noise while retaining the edge and detail information of the images. Terrain correction is used to eliminate the influence of terrain undulation on radar images, enabling the images to more realistically reflect the backscattering characteristics of ground objects. Exemplarily, the Shuttle Radar Topography Mission Digital Elevation Model (SRTM DEM) is used for terrain correction to eliminate terrain shadows and geometric distortions.

[0059] Optionally, for ground data, data preprocessing includes, but is not limited to, coordinate transformation, data interpolation, etc. Among them, coordinate transformation is used to convert the coordinate system of ground measured data into a coordinate system consistent with remote sensing images, ensuring accurate spatial matching of the data. Data interpolation is used to interpolate discrete ground measured data into continuous spatially distributed data, filling in data missing areas and generating complete time series data. Exemplarily, data interpolation includes spatial interpolation and temporal interpolation. Spatial interpolation is used to generate continuous spatially distributed data, and temporal interpolation is used to fill in missing data in time series. Through the interpolation process of ground data, complete and continuous spatial and temporal data can be generated to further supplement remote sensing data.

[0060] In a possible implementation, the obtained optical data, radar data, and ground data of the target area are data after preliminary registration. Here, the specific implementation methods of preliminary registration include, but are not limited to, unifying the projection coordinate system, extracting feature points, and removing false matches. Among them, unifying the projection coordinate system is to ensure the geographical reference consistency of optical data, radar data, and ground data. For extracting feature points and removing false matches, it is to achieve the preliminary registration of optical data and radar data. Feature point extraction methods include, but are not limited to, Scale-Invariant Feature Transform (SIFT), Oriented FAST and Rotated BRIEF (ORB), Accelerated KAZE (AKAZE). False match removal can be achieved through Random Sample Consensus (RANSAC).

[0061] S102: Based on the first aquatic vegetation area corresponding to the optical data, the second aquatic vegetation area corresponding to the radar data, and the third aquatic vegetation area corresponding to the ground data, perform spatial registration on the optical data, radar data, and ground data to obtain a first registered data set.

[0062] Specifically, spatial registration is to align data from different sensors, at different times, or at different spatial positions through mathematical transformation and geometric correction so that they are in the same spatial reference system. The first registered data set includes the registered optical data, radar data, and ground data.

[0063] Specifically, the first aquatic vegetation area refers to the aquatic vegetation area obtained using optical data. The second aquatic vegetation area refers to the aquatic vegetation area obtained using radar data. The ground data refers to the aquatic vegetation area determined based on the ground data. The determination methods for the first aquatic vegetation area, the second aquatic vegetation area, and the third aquatic vegetation area will be described in subsequent embodiments and will not be elaborated here.

[0064] In a possible implementation manner, the overlapping parts of the first aquatic vegetation area, the second aquatic vegetation area, and the third aquatic vegetation area can be used to perform spatial registration on the optical data, radar data, and ground data. Exemplarily, the specific implementation process includes: First, determine the overlapping parts of the first aquatic vegetation area, the second aquatic vegetation area, and the third aquatic vegetation area. Then, based on the overlapping parts (such as the geometric center or feature points of the overlapping parts), calculate the spatial transformation model. Finally, resample the radar data and the ground data to align the radar data and the ground data with the optical data. Integrate the registered optical data, radar data, and ground data into a unified data, that is, the first registered data set.

[0065] In another possible implementation manner, the implementation manner of spatial registration can be the unified projection coordinate system, feature point extraction, and false match elimination described above. Of course, a reference data can also be determined from the optical data, radar data, and ground data, and other data are spatially registered based on this reference data to achieve alignment with the reference data.

[0066] In yet another possible implementation manner, use the geographic coordinate matching algorithm to unify the optical data corresponding to the first aquatic vegetation area, the radar data corresponding to the second aquatic vegetation area, and the ground data corresponding to the third aquatic vegetation area to the same spatial resolution and time scale.

[0067] S103: Based on the first registered data set, detect the aquatic plants in the target area to obtain a detection result.

[0068] In a possible implementation, the detection result can be the classification of floating plants and submerged plants in the target area, and can also be plant growth characteristics, etc. Plant growth characteristics include, but are not limited to, changes in vegetation coverage area, changes in plant spatial distribution, changes in vegetation biomass, vegetation growth rate, seasonal change characteristics, changes in water transparency, changes in chlorophyll concentration, changes in radar backscattering coefficient, etc. Among them, the change in vegetation coverage area is used to monitor the diffusion or recession trend of aquatic plants and evaluate the potential impact of aquatic plants on the intake of hydropower stations. The change in plant spatial distribution, such as diffusion towards the intake or retreat towards the shore, is used to evaluate the potential threat of aquatic plants to the intake of hydropower stations and guide the priority of cleaning work. The change in vegetation biomass is used to evaluate the growth state of aquatic plants and determine whether there is overgrowth. The vegetation growth rate is used to predict the future growth trend of aquatic plants and give early warnings of possible risks. Seasonal change characteristics refer to analyzing the growth laws of aquatic plants in different seasons and extracting seasonal change characteristics (such as the rapid growth period in spring and the recession period in autumn) to understand the life cycle of aquatic plants and optimize monitoring and governance strategies. The change in water transparency refers to analyzing the change in water transparency through time-series transparency index data, indirectly reflecting the growth state of submerged plants, and is used to evaluate the impact of submerged plants on the water environment. The change in chlorophyll concentration refers to analyzing the growth dynamics of phytoplankton and floating plants and is used to monitor the degree of water eutrophication and evaluate the growth trend of floating plants. The change in radar backscattering coefficient refers to analyzing the change in the area covered by floating plants through the backscattering coefficient of time-series radar images, and is used to monitor the diffusion or recession of floating plants, especially the change under the interference of clouds in optical images.

[0069] In a possible implementation, the specific implementation process of obtaining the detection result based on the first registration dataset includes: First, extract features from the first registration dataset to obtain optical features, radar features, and ground measurement features in the first registration dataset; then, based on the optical features, radar features, ground measurement features, and the second prediction model, detect aquatic plants to obtain the detection result. It should be noted that the specific implementation method of the above S103 will be described in subsequent embodiments and will not be elaborated here.

[0070] In a possible implementation, by performing principal component analysis on the optical data, radar data, and ground data in the obtained first registration dataset, the data in the first registration dataset is fused to obtain the fused data, and based on the fused data, the aquatic plants in the target area are detected to obtain the detection result.

[0071] In the embodiments of the present application, optical data, radar data, and ground data are integrated to detect aquatic plants using multi-source data, overcoming the limitations of a single data source. Among them, optical data provides rich spectral information, radar data has all-weather observation capabilities, and ground data provides high-precision reference information. The implementation mode of the present application significantly improves the accuracy and reliability of aquatic vegetation detection through the fusion of multi-source data. In addition, before detecting aquatic plants in the target area, spatial registration of optical data, radar data, and ground data can accurately correspond the information in different data sources to the same geographic spatial coordinate system. In this way, when detecting aquatic plants, the information at the same location in different data sources can be comprehensively analyzed, avoiding misjudgment and missed judgment caused by inconsistent spatial positions of the data, effectively solving the inconsistency problem caused by differences in resolution and imaging mechanisms of multi-source data, providing a high-quality data basis for subsequent detection of aquatic plants, and improving the accuracy of aquatic plant detection.

[0072] In some embodiments, the method for detecting aquatic plants provided by the present application further includes the following content:

[0073] a1: Based on the optical data, determine the vegetation information of the target area; the vegetation information is used to characterize the aquatic plants in the target area.

[0074] In a possible implementation manner, the vegetation information includes vegetation indices (such as Normalized Difference Vegetation Index (NDVI), Enhanced Vegetation Index (EVI)), Fraction of Absorbed Photosynthetically Active Radiation (FAPAR), Normalized Difference Water Index, Forel-Ule Index (FUI), etc. Optical data (such as multi-spectral and hyperspectral images) can provide rich spectral information. By analyzing features such as vegetation indices, the vegetation information of the target area can be efficiently and accurately extracted.

[0075] Exemplarily, the Normalized Difference Vegetation Index is calculated by the following formula:

[0076]

[0077] Among them, NDVI is the Normalized Difference Vegetation Index, NIR is the reflection value of the near-infrared band, and R is the reflection value of the red light band.

[0078] a2: Based on the vegetation information, determine the first aquatic vegetation area.

[0079] In a possible implementation, in the above a2, when the vegetation information includes a vegetation index, the area where the vegetation index meets the preset condition is used as the first aquatic vegetation area. Exemplarily, the area where the vegetation index is greater than the first preset threshold is used as the first aquatic vegetation area. For example, the area where NDVI > 0.2 and NDWI > 0.1 is the aquatic vegetation area.

[0080] In another possible implementation, in the above a2, a first initial area can be obtained based on the vegetation information; then the first initial area is further segmented using the k-means clustering algorithm or the watershed algorithm to determine the first aquatic vegetation area. In the embodiments of the present application, through the vegetation information and the k-means clustering algorithm / watershed algorithm, the target area is segmented twice, which can significantly improve the accuracy of the first aquatic vegetation area.

[0081] In addition, the normalized difference water index can also be calculated based on the optical data, and the first aquatic vegetation area can also be determined using the normalized difference water index. The normalized difference water index is calculated by the following formula:

[0082]

[0083] where NDWI is the normalized difference water index, NIR is the reflectance value in the near-infrared band; G is the reflectance value in the green band.

[0084] In some embodiments, the method for detecting aquatic plants provided by the present application further includes the following content:

[0085] First, based on the radar data, the water body information of the target area is determined; the water body information is used to indicate the water body area in the target area.

[0086] Specifically, the water body information includes, but is not limited to, backscattering coefficient, polarization characteristics, etc. Among them, the backscattering coefficient (σ°) is the reflection intensity after the interaction between the radar signal and the earth's surface, usually expressed in decibels (dB). The water body surface usually shows a low backscattering value because the water body surface is relatively smooth, and the radar signal will be specularly reflected, resulting in a weak echo signal. For example, the backscattering coefficient of open water bodies (such as lakes and rivers) is usually low. When there are wind waves or floating objects (such as floating aquatic plants) on the water body surface, the backscattering coefficient will increase. The mutation of the backscattering coefficient can be used to determine the second aquatic vegetation area. Radar data usually contains multiple polarization modes (such as VV, VH, HH, HV), and different polarization modes have different sensitivities to water bodies and vegetation. The water body usually shows a low backscattering value under VV polarization, while it may show a slightly higher value under VH polarization (especially in the case of floating aquatic plants or wind waves). By performing polarization decomposition on the radar data, three components, namely surface scattering, double scattering, and volume scattering, and the corresponding characteristic parameters of each component (such as scattering power, scattering entropy, etc.) can be obtained. The polarization characteristics can include the characteristic parameters corresponding to the above three classifications. For aquatic vegetation, it usually shows a high volume scattering power and certain double scattering characteristics.

[0087] Then, based on the water body information, the second aquatic vegetation area is determined.

[0088] In a possible implementation manner, when the water body information includes the backscattering coefficient, the area where the backscattering coefficient is greater than the second preset threshold can be used as the second aquatic vegetation area.

[0089] In a possible implementation manner, when the water body information includes polarization characteristics, the characteristic parameters of volume scattering and double scattering in the polarization characteristics are used to determine the second aquatic vegetation area in the target area. Specifically, the area where the polarization characteristics meet the preset conditions is used as the second aquatic vegetation area. For example, the area where the volume scattering power > the third preset threshold and the double scattering power > the fourth preset threshold is set as the water body area.

[0090] In the embodiments of the present application, radar data (such as synthetic aperture radar SAR) has all-weather and all-time observation capabilities, can penetrate clouds and some vegetation, and quickly and accurately extract the water body information of the target area. Further, based on the water body information of the radar data, the water body range in the target area can be clearly identified, that is, the second aquatic vegetation area is determined.

[0091] In this embodiment, a detection method for aquatic plants is provided, which can be used for the above-mentioned electronic devices, such as servers, etc. Figure 2 It is a flowchart of another detection method for aquatic plants according to an embodiment of the present invention, as Figure 2 shown, and the process includes the following steps:

[0092] S201: Obtain the optical data, radar data, and ground data of the target area. For details, please refer to Figure 1 Step S101 of the illustrated embodiment, which will not be elaborated herein.

[0093] S202: Based on the first aquatic vegetation area corresponding to the optical data, the second aquatic vegetation area corresponding to the radar data, and the third aquatic vegetation area corresponding to the ground data, perform spatial registration on the optical data, radar data, and ground data to obtain a first registration data set.

[0094] In the embodiment of the present application, in the above S202, the first registration data set is obtained through the following S2021-S2022:

[0095] S2021: Based on the first aquatic vegetation area, the second aquatic vegetation area, and the third aquatic vegetation area, perform spatial registration on the optical data, radar data, and ground data to obtain a second registration data set.

[0096] In a possible implementation manner, in the above S2021, the optical data, radar data, and ground data are spatially registered through the following content to obtain a second registration data set:

[0097] First, use the pyramid scale feature matching method, and / or, the dense manifold matching method, to perform spatial registration on the optical data, radar data, and ground data to obtain a third registration data set.

[0098] Specifically, the pyramid scale feature matching method is a multi-scale registration technique. By extracting feature points and performing matching at multiple scales, the registration accuracy is gradually optimized. Using the pyramid scale feature matching method, high-resolution optical data and low-resolution radar data can be aligned at different scales.

[0099] Specifically, the dense manifold matching method is a registration technique based on manifold learning. By constructing the manifold structure of the data, the corresponding relationship between different data is found. Using the dense manifold matching (Dense Feature Matching, DFM) method, the registration points are optimized layer by layer at different scales to improve the registration accuracy.

[0100] It should be noted that in the process of spatially registering optical data, radar data, and ground data using the pyramid scale feature matching method and the dense manifold matching method, the order between the pyramid scale feature matching method and the dense manifold matching method is not limited. The pyramid scale feature matching method can be first used for spatial registration, and then the dense manifold matching method can be further used for spatial registration to obtain the third registration dataset. Alternatively, the dense manifold matching method can be first used for spatial registration, and then the pyramid scale feature matching method can be further used for spatial registration to obtain the third registration dataset.

[0101] In addition, for the parameters in the pyramid scale matching method, the registration error can be calculated based on the data before and after registration, and adjusted based on the calculated registration error, so that the registration error corresponding to the parameters in the pyramid scale matching method is within the preset error range. Exemplarily, the registration error can be the root mean square error (RMSE), Hausdorff distance, etc.

[0102] Then, based on the first aquatic vegetation area, the second aquatic vegetation area, and the third aquatic vegetation area, the third registration dataset is registered to obtain the second registration dataset.

[0103] It should be noted that the spatial registration based on the first aquatic vegetation area, the second aquatic vegetation area, and the third aquatic vegetation area has been described in the above embodiments and will not be elaborated here.

[0104] In the embodiments of the present application, the pyramid scale feature matching method can capture the feature information of the light source data and the radar data at different scales and is applicable to the spatial registration of data with large resolution differences. The dense manifold matching method can handle complex non-linear feature alignment problems and is particularly applicable to the feature differences caused by different imaging mechanisms between optical data, radar data, and ground data, especially in areas with complex terrain and vegetation distribution. In the embodiments of the present application, through multi-level and multi-method registration optimization, the obtained second registration dataset has higher spatial accuracy and consistency, providing a reliable data basis for subsequent aquatic vegetation detection.

[0105] S2022: Based on the water flow data in the target area, the second registration dataset is adjusted to obtain the first registration dataset.

[0106] Specifically, the water flow data can better reflect the hydrodynamic characteristics of the target area, especially for areas with significant dynamic changes in water flow, such as the upper reaches of the river. Using the water flow data to adjust the second registration data can effectively reduce the distribution error of aquatic vegetation caused by water flow movement, making the detection results more consistent with the actual environment. Exemplarily, the water flow data includes, but is not limited to, the flow velocity and direction of the water flow, water level, etc. The water flow data can be obtained by simulating with a water flow motion field model. Among them, the water flow motion field model includes, but is not limited to, the Soil and Water Assessment Tool (SWAT), the MIKE 21 hydrodynamic model, or a water flow model based on water flow radar data.

[0107] In a possible implementation manner, in the above S2022, the specific implementation process of adjusting the second registration data set based on the water flow data in the target area includes:

[0108] First, input the water flow data into a pre-constructed first prediction model to obtain the predicted spatial position of the target object in the target area.

[0109] Optionally, the first prediction model can be a deep learning model, such as a Convolutional Neural Network (CNN), a Long Short-Term Memory (LSTM) network. Of course, the CNN and LSTM can also be combined to predict the spatial position of the target object.

[0110] Then, based on the predicted spatial position, adjust the second registration data set to obtain the first registration data set.

[0111] Optionally, the implementation manner of further adjusting the second registration data set based on the predicted spatial position to obtain the first registration data set includes: b1: Compare the position of the target object in the second registration data set with the predicted position, and calculate the deviation between the position of the target object and the predicted spatial position. b2: According to the predicted spatial position, adjust the position of the target object in the second registration data set to make the position of the target object consistent with the predicted spatial position, and obtain the second registration data set. Exemplarily, it can be achieved through spatial transformation (such as translation, rotation, scaling, etc.).

[0112] In the embodiments of the present application, by introducing water flow data and using a pre-constructed first prediction model, it is possible to accurately predict the spatial position changes of target objects (such as floating aquatic plants and submerged aquatic plants) under the action of water flow. At the same time, based on the second registration dataset, with the predicted spatial position of the target object as the benchmark, the second registration dataset is adjusted, further eliminating the data deviation caused by the dynamic changes of water flow, making the fusion of optical data, radar data, and ground data more accurate, and providing a basis for improving the accuracy of aquatic vegetation detection.

[0113] S203: Based on the first registration dataset, detect the aquatic plants in the target area to obtain a detection result. For details, please refer to Figure 1 Step S103 of the embodiment shown, which will not be elaborated here.

[0114] In some embodiments, in the above S103, the aquatic plants in the target area are detected in the following manner:

[0115] First, determine the optical features, radar features, and ground measurement features in the first registration dataset.

[0116] Among them, the optical features include vegetation indices, such as the normalized difference vegetation index, enhanced vegetation index, etc. The radar features include the backscattering coefficient, and / or, polarization features. The backscattering coefficient reflects the scattering characteristics of the ground object. Water bodies usually show a low backscattering coefficient, while vegetation shows a relatively high backscattering coefficient. The polarization features extracted through polarization decomposition (such as Freeman-Durden decomposition), such as surface scattering, secondary scattering, and volume scattering features, etc., are used to distinguish different types of ground objects. The ground measurement features include biomass, and / or, chlorophyll concentration.

[0117] Then, input the optical features, radar features, and ground measurement features into a pre-constructed second prediction model to obtain a detection result.

[0118] Optionally, the second prediction model is used to predict the classification of aquatic plants in the target area, and determine whether the plants in the target area are floating aquatic plants, submerged aquatic plants, etc. Exemplarily, the second prediction model is a classification model constructed by the RT-DETR algorithm.

[0119] Optionally, the second prediction model is used to predict the growth characteristics of aquatic plants. Exemplarily, it can be a machine learning or deep learning model, which is used to comprehensively analyze multi-source features to achieve precise detection of aquatic plants. For example, the second prediction model is a random forest, support vector machine (Support Vector Machine, SVM), real-time detection Transformer model, etc.

[0120] It is understandable that during the training process of the second prediction model, the optical data, radar data, and ground data used are also data after spatial registration. The specific implementation method of spatial registration is similar to the above method and will not be elaborated here.

[0121] In the embodiments of the present application, vegetation indices (such as NDVI, EVI) in optical features can effectively reflect the growth status and coverage of aquatic plants, providing important spectral information for the detection of aquatic plants. Radar features (such as backscattering coefficient, polarization features) can penetrate clouds and provide all-weather and all-time observation data, which are particularly suitable for complex environments with limited optical data (such as cloudy weather). Ground measurement features (such as biomass, chlorophyll concentration) provide high-precision reference information for the detection results, further ensuring the accuracy of the detection results. By combining optical features (such as vegetation indices), radar features (such as backscattering coefficient, polarization features), and ground measurement features (such as biomass, chlorophyll concentration), the complementary advantages of multi-source data are fully utilized to support the improvement of the accuracy and reliability of aquatic plant detection.

[0122] In some embodiments, the method for detecting aquatic plants provided by the present application further includes: displaying and warning the detection results.

[0123] Exemplarily, based on the detection results, a vegetation coverage map, index distribution map, growth dynamic curve map, etc. of the target area can be generated. For example, in the vegetation coverage map, different colors are used to represent different types of vegetation (such as submerged plants, floating plants, emergent plants). In the index distribution map, a color gradient is used to represent the high and low of the index value. In the growth dynamic curve map, a curve is used to represent the change of the vegetation index or biomass over time.

[0124] Optionally, when the detection result is greater than the growth threshold, a risk warning is generated in combination with the prediction result. By setting the growth threshold, it is judged whether the growth of aquatic plants exceeds the normal range. Among them, the growth threshold can be set according to the actual situation and is not limited here. For example, if the NDVI of a certain area > 0.5 and it is predicted that floating plants will drift to the intake of the hydropower station, a risk warning of intake blockage is generated.

[0125] In the embodiments of the present application, by displaying and warning the detection results, the distribution and dynamic changes of aquatic plants in the target area can be intuitively presented, providing a scientific basis and decision-making support for water resource management and ecological protection. For example, by accurately monitoring and predicting the growth trend of aquatic plants, early warning can be provided before the problem of intake blockage of the hydropower station occurs, effectively avoiding operation risks. It provides a scientific decision-making basis for the operation personnel of the hydropower station, reducing potential economic losses and safety hazards.

[0126] The above mainly introduces the solution provided by the embodiments of the present application from the perspective of methods.

[0127] In the embodiments of the present application, a detection device for aquatic plants is further provided. This device is used to implement the above embodiments and preferred implementation manners, and those that have been described will not be repeated. As used hereinafter, the term "module" can be a combination of software and / or hardware that can achieve a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware is also possible and contemplated.

[0128] This embodiment provides a detection device for aquatic plants, as Figure 3 shown, the device includes:

[0129] An acquisition module 301, configured to acquire optical data, radar data, and ground data of a target area;

[0130] A registration module 302, configured to perform spatial registration on the optical data, radar data, and ground data based on a first aquatic vegetation area corresponding to the optical data, a second aquatic vegetation area corresponding to the radar data, and a third aquatic vegetation area corresponding to the ground data, to obtain a first registered data set;

[0131] A detection module 303, configured to detect aquatic plants in the target area based on the first registered data set, to obtain a detection result.

[0132] Through the above device, the optical data, radar data, and ground data are integrated, overcoming the limitations of a single data source. Among them, the optical data provides rich spectral information, the radar data has all-weather observation capabilities, and the ground data provides high-precision reference information. The implementation manner of the present application significantly improves the accuracy and reliability of aquatic vegetation detection through the fusion of multi-source data. In addition, spatial registration of the optical data, radar data, and ground data can accurately correspond the information in different data sources to the same geographic space coordinate system. In this way, when detecting aquatic plants, the information at the same location in different data sources can be comprehensively analyzed, avoiding misjudgment and missed judgment caused by inconsistent data spatial positions, effectively solving the problem of inconsistency caused by differences in resolution and imaging mechanisms of multi-source data, and providing a high-quality data basis for subsequent analysis.

[0133] In some alternative implementation manners, in this device, the registration module 302 includes:

[0134] A registration sub-module, configured to perform spatial registration on the optical data, radar data, and ground data based on the first aquatic vegetation area, the second aquatic vegetation area, and the third aquatic vegetation area, to obtain a second registered data set;

[0135] An adjustment sub-module, configured to adjust a second registration data set based on water flow data in a target area to obtain a first registration data set.

[0136] In some alternative embodiments, the adjustment sub-module includes:

[0137] A prediction unit, configured to input the water flow data into a pre-constructed first prediction model to obtain a predicted spatial position of a target object in the target area;

[0138] An adjustment unit, configured to adjust the second registration data set based on the predicted spatial position to obtain a first registration data set.

[0139] In some alternative embodiments, the registration sub-module includes:

[0140] A first registration unit, configured to perform spatial registration on optical data, radar data, and ground data by using a pyramid scale feature matching method and / or a dense manifold matching method to obtain a third registration data set;

[0141] A second registration unit, configured to register the third registration data set based on a first aquatic vegetation area, a second aquatic vegetation area, and a third aquatic vegetation area to obtain a second registration data set.

[0142] In some alternative embodiments, the detection module 303 includes:

[0143] A determination sub-module, configured to determine optical features, radar features, and ground measurement features in the first registration data set; the optical features include vegetation indices; the radar features include backscattering coefficients and / or polarization features; the ground measurement features include biomass and / or chlorophyll concentration;

[0144] A prediction sub-module, configured to input the optical features, radar features, and ground measurement features into a pre-constructed second prediction model to obtain a detection result.

[0145] In some alternative embodiments, the apparatus further includes:

[0146] A first determination module, configured to determine vegetation information of a target area based on optical data; the vegetation information is used to characterize aquatic plants in the target area;

[0147] A second determination module, configured to determine a first aquatic vegetation area based on the vegetation information.

[0148] In some alternative embodiments, the apparatus further includes:

[0149] A third determination module, configured to determine water body information of a target area based on radar data; the water body information is used to indicate a water body area in the target area;

[0150] A fourth determination module, configured to determine a second aquatic vegetation area based on water body information.

[0151] The further function descriptions of the above-mentioned modules and units are the same as those in the corresponding embodiments above, and will not be elaborated here.

[0152] The detection device for aquatic plants in this embodiment is presented in the form of functional units. Here, the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and a memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0153] An embodiment of the present invention further provides a computer device having the above-mentioned Figure 3 shown detection device for aquatic plants.

[0154] Please refer to Figure 4 , Figure 4 which is a schematic structural diagram of a computer device provided by an optional embodiment of the present invention. As Figure 4 shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including a high-speed interface and a low-speed interface. Each component communicates with each other using different buses and can be installed on a common main board or installed in other ways according to needs. The processor can process instructions executed within the computer device, including instructions stored in the memory or on the memory to display graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a set of blade servers, or a multi-processor system). Figure 4 In

[0155] Processor 10 may be a central processing unit, a network processor, or a combination thereof. Among them, processor 10 may further include a hardware chip. The above-mentioned hardware chip may be an application specific integrated circuit, a programmable logic device, or a combination thereof. The above-mentioned programmable logic device may be a complex programmable logic device, a field programmable gate array, a general array logic, or any combination thereof.

[0156] Among them, the memory 20 stores instructions executable by at least one processor 10, so that at least one processor 10 executes the method shown in the above embodiments.

[0157] The memory 20 may include a program storage area and a data storage area. Among them, the program storage area can store an operating system and application programs required for at least one function; the data storage area can store data created according to the use of the computer device and the like. In addition, the memory 20 may include high-speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some alternative embodiments, the memory 20 may optionally include a memory remotely provided relative to the processor 10, and these remote memories can be connected to the computer device through a network. Examples of the above-mentioned network include but are not limited to the Internet, an enterprise intranet, a local area network, a mobile communication network, and combinations thereof.

[0158] The memory 20 may include volatile memory, for example, random access memory; the memory may also include non-volatile memory, for example, flash memory, a hard disk, or a solid-state drive; the memory 20 may further include a combination of the above types of memory.

[0159] The computer device further includes a communication interface 30 for the computer device to communicate with other devices or a communication network.

[0160] The embodiments of the present invention also provide a computer-readable storage medium. The methods according to the embodiments of the present invention can be implemented in hardware, firmware, or be implemented as computer code that can be recorded on a storage medium, or be implemented as computer code originally stored in a remote storage medium or a non-transitory machine-readable storage medium and downloaded through a network and will be stored in a local storage medium, so that the methods described herein can be stored in such software processed on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only memory, a random access memory, a flash memory, a hard disk, or a solid-state drive, etc.; further, the storage medium may also include a combination of the above types of memory. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code, and when the software or computer code is accessed and executed by the computer, the processor, or the hardware, the methods shown in the above embodiments are implemented.

[0161] A part of the present invention can be applied as a computer program product, for example, computer program instructions. When executed by a computer, through the operation of the computer, the methods and / or technical solutions according to the present invention can be invoked or provided. Those skilled in the art should understand that the forms of existence of computer program instructions in a computer-readable medium include but are not limited to source files, executable files, installation package files, etc. Correspondingly, the ways for a computer to execute computer program instructions include but are not limited to: the computer directly executes the instructions, or the computer compiles the instructions and then executes the corresponding compiled program, or the computer reads and executes the instructions, or the computer reads and installs the instructions and then executes the corresponding installed program. Herein, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to the computer.

[0162] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A detection method for aquatic plants, characterized in that, The method includes: Obtaining optical data, radar data, and ground data of a target area; Performing spatial registration on the optical data, the radar data, and the ground data based on a first aquatic vegetation area corresponding to the optical data, a second aquatic vegetation area corresponding to the radar data, and a third aquatic vegetation area corresponding to the ground data, to obtain a first registered data set; Detecting aquatic plants in the target area based on the first registered data set to obtain a detection result.

2. The method according to claim 1, characterized in that, The performing spatial registration on the optical data, the radar data, and the ground data based on a first aquatic vegetation area corresponding to the optical data, a second aquatic vegetation area corresponding to the radar data, and a third aquatic vegetation area corresponding to the ground data, to obtain a first registered data set, includes: Performing spatial registration on the optical data, the radar data, and the ground data based on the first aquatic vegetation area, the second aquatic vegetation area, and the third aquatic vegetation area, to obtain a second registered data set; Adjusting the second registered data set based on water flow data in the target area to obtain the first registered data set.

3. The method according to claim 2, wherein The adjusting the second registered data set based on water flow data in the target area to obtain the first registered data set, includes: Inputting the water flow data into a pre-constructed first prediction model to obtain a predicted spatial position of a target object in the target area; Adjusting the second registered data set based on the predicted spatial position to obtain the first registered data set.

4. The method according to claim 2, wherein The performing spatial registration on the optical data, the radar data, and the ground data based on the first aquatic vegetation area, the second aquatic vegetation area, and the third aquatic vegetation area, to obtain a second registered data set, includes: Performing spatial registration on the optical data, the radar data, and the ground data using a pyramid scale feature matching method, and / or, a dense manifold matching method, to obtain a third registered data set; Performing registration on the third registered data set based on the first aquatic vegetation area, the second aquatic vegetation area, and the third aquatic vegetation area to obtain the second registered data set.

5. The method according to any one of claims 1-4, characterized in that, The detecting aquatic plants in the target area based on the first registered data set to obtain a detection result, includes: Determining optical features, radar features, and ground measurement features in the first registered data set; the optical features include vegetation indices; the radar features include backscattering coefficients, and / or, polarization features; the ground measurement features include biomass, and / or, chlorophyll concentration; Inputting the optical features, the radar features, and the ground measurement features into a pre-constructed second prediction model to obtain the detection result.

6. The method according to claim 5, wherein The method further includes: Determining vegetation information of the target area based on the optical data; the vegetation information is used to characterize aquatic plants in the target area; Determining the first aquatic vegetation area based on the vegetation information.

7. The method according to claim 6, characterized in that, The method further includes: Determine the water body information of the target area based on the radar data; the water body information is used to indicate the water body area in the target area; Determine the second aquatic vegetation area based on the water body information.

8. A detection device for aquatic plants, characterized in that, The device includes: An acquisition module, configured to acquire optical data, radar data, and ground data of a target area; A registration module, configured to perform spatial registration on the optical data, the radar data, and the ground data based on the first aquatic vegetation area corresponding to the optical data, the second aquatic vegetation area corresponding to the radar data, and the third aquatic vegetation area corresponding to the ground data, to obtain a first registered data set; A detection module, configured to detect aquatic plants in the target area based on the first registered data set, to obtain a detection result.

9. A computer device, characterized in that, Including: A memory and a processor, which are communicatively connected to each other, wherein the memory stores computer instructions, and the processor executes the computer instructions to execute the detection method of aquatic plants according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, Computer instructions are stored on the computer-readable storage medium, and the computer instructions are used to cause a computer to execute the detection method of aquatic plants according to any one of claims 1 to 7.

Images