Method and system for monitoring cyanobacterial bloom

Abstract

The invention provides a cyanobacterial bloom monitoring method and system. The cyanobacterial bloom monitoring method comprises the steps that a target water body remote sensing image of a target water body area of a monitoring area is acquired; screening a chlorophyll a sensitive wave band of the target water body remote sensing image, and obtaining chlorophyll a inversion concentration based on a preset chlorophyll a concentration inversion model; calculating a normalized vegetation index according to the multi-band remote sensing reflectivity value of the target water body remote sensing image; according to the chlorophyll a inversion concentration and the normalized vegetation index, constructing a cyanobacterial bloom grading threshold dynamic calibration model, and performing cyanobacterial bloom risk grading on the target water body area; and according to the cyanobacterial bloom risk grading result of the target water body area, generating an early warning instruction corresponding to the risk grade. The cyanobacterial bloom grading threshold value can be dynamically adjusted, so that the subjective deviation of manual weighting can be eliminated, and the real-time performance, accuracy and efficiency of cyanobacterial bloom monitoring and early warning are effectively improved.

Description

Technical Field

[0001] The present application relates to the field of geographic information technology, and in particular to a method and system for monitoring cyanobacteria blooms. Background Art

[0002] Cyanobacteria bloom refers to the phenomenon of harmful algae bloom formed by the massive reproduction of cyanobacteria under suitable environmental conditions, which causes serious damage to the aquatic ecosystem and even poses a threat to human health. With global climate change and the increase of human activities, the frequency and intensity of cyanobacteria blooms have increased. In this context, it is of great significance to carry out effective monitoring and early warning of cyanobacteria.

[0003] At present, the monitoring and early warning of cyanobacteria blooms at home and abroad mainly rely on two technical paths: mathematical model method and hierarchical analysis method. The mathematical model method simulates the evolution of algae blooms by establishing a multi-parameter coupled dynamic model (such as a three-dimensional response model of nutrients, light and water temperature), and then constructs an evaluation system through hierarchical analysis method. However, the determination of its indicator weights is overly dependent on expert experience and is easily affected by subjective cognitive bias. There will be significant differences in the risk level classification of the same water area by different research teams, resulting in low efficiency and reduced accuracy in the monitoring and early warning of cyanobacteria blooms. Summary of the invention

[0004] In response to the problems in the prior art, the present application provides a method and system for monitoring cyanobacterial blooms, which can effectively improve the efficiency and accuracy of cyanobacterial bloom monitoring and early warning by utilizing the correlation characteristics of chlorophyll a concentration and normalized vegetation index to achieve dynamic adjustment of cyanobacterial bloom classification thresholds.

[0005] A method for monitoring blue algae blooms, comprising:

[0006] Acquire a remote sensing image of a monitored area and extract multi-band remote sensing reflectance values ​​of the remote sensing image;

[0007] Extracting a target water body remote sensing image containing only the target water body area based on the multi-band remote sensing reflectance values ​​of the remote sensing image and a preset water body classification model;

[0008] By performing correlation analysis on the multi-band remote sensing reflectance values ​​and chlorophyll a concentration of the target water body remote sensing image, the chlorophyll a sensitive band of the target water body remote sensing image is screened;

[0009] According to the remote sensing reflectance value of the chlorophyll a sensitive band, based on a preset chlorophyll a concentration inversion model, the chlorophyll a inversion concentration is obtained;

[0010] According to the multi-band remote sensing reflectance value of the remote sensing image of the target water body, based on a preset normalized vegetation index calculation formula, the vegetation growth condition of the target water body is analyzed to obtain a normalized vegetation index;

[0011] Construct a dynamic calibration model for the cyanobacterial bloom classification threshold based on the retrieved chlorophyll a concentration and the normalized difference vegetation index, and classify the cyanobacterial bloom risk for the target water body area;

[0012] Correspondingly send a warning instruction to the cyanobacterial bloom warning device according to the cyanobacterial bloom risk classification of the target water body area.

[0013] This application also provides a cyanobacterial bloom monitoring system, including:

[0014] Remote sensing image acquisition module: used to acquire the remote sensing image of the monitoring area and extract the multi-band remote sensing reflectance values of the remote sensing image;

[0015] Target water body remote sensing image extraction module: used to extract the target water body remote sensing image containing only the target water body area based on the multi-band remote sensing reflectance values of the remote sensing image and a preset water body classification model;

[0016] Sensitive band screening module: used to screen the chlorophyll a sensitive bands of the target water body remote sensing image by performing a correlation analysis on the multi-band remote sensing reflectance values of the target water body remote sensing image and the chlorophyll a concentration;

[0017] Chlorophyll a concentration retrieval module: used to obtain the retrieved chlorophyll a concentration based on the remote sensing reflectance values of the chlorophyll a sensitive bands and a preset chlorophyll a concentration retrieval model;

[0018] Normalized difference vegetation index calculation module: used to analyze the vegetation growth situation of the target water body based on the multi-band remote sensing reflectance values of the target water body remote sensing image and a preset normalized difference vegetation index calculation formula, and obtain the normalized difference vegetation index;

[0019] Target water body area classification module: used to construct a dynamic calibration model for the cyanobacterial bloom classification threshold based on the retrieved chlorophyll a concentration and the normalized difference vegetation index, and classify the cyanobacterial bloom risk for the target water body area;

[0020] Warning instruction sending module: used to generate a warning instruction corresponding to the risk level according to the cyanobacterial bloom risk classification result of the target water body area, and transmit the warning instruction to the cyanobacterial bloom warning device.

[0021] Compared with the prior art, the present application determines the target water body area of the monitoring area based on the remote sensing image of the monitoring area, and obtains the chlorophyll a inversion concentration and the normalized vegetation index of the target water body area according to the target water body remote sensing image of the target water body area. Secondly, by using the correlation characteristics of the chlorophyll a concentration and the normalized vegetation index to construct an unsupervised cyanobacteria bloom classification threshold dynamic calibration model, the dynamic adjustment of the cyanobacteria bloom classification threshold can be realized, thereby eliminating the subjective deviation of manual weighting and effectively improving the real-time performance, accuracy and efficiency of cyanobacteria bloom monitoring and early warning.

[0022] In order to understand the present application more clearly, the specific implementation manners of the present application will be described below in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a flowchart of a cyanobacteria bloom monitoring method of the present application;

[0024] Figure 2 It is a flowchart of a method for constructing a water body classification model in a cyanobacteria bloom monitoring method of the present application;

[0025] Figure 3 It is a flowchart of a method for evaluating the accuracy of a water body classification model in a cyanobacteria bloom monitoring method of the present application;

[0026] Figure 4 It is a flowchart of a method for screening chlorophyll a sensitive bands in a cyanobacteria bloom monitoring method of the present application;

[0027] Figure 5 It is a flowchart of a method for constructing a chlorophyll a concentration inversion model in a cyanobacteria bloom monitoring method of the present application;

[0028] Figure 6 It is a flowchart of a method for classifying the cyanobacteria bloom risk level of the target water body area in a cyanobacteria bloom monitoring method of the present application;

[0029] Figure 7 It is a reference diagram of the on-site water body with no risk in the risk level of the target water body area in a cyanobacteria bloom monitoring method of the present application;

[0030] Figure 8 It is a reference diagram of the on-site water body with a relatively low risk in the risk level of the target water body area in a cyanobacteria bloom monitoring method of the present application;

[0031] Figure 9 It is a reference diagram of the on-site water body with a low risk in the risk level of the target water body area in a cyanobacteria bloom monitoring method of the present application;

[0032] Figure 10 It is a reference diagram of the on-site water body with a medium risk in the risk level of the target water body area in a cyanobacteria bloom monitoring method of the present application;

[0033] Figure 11 This is a reference diagram of the on-site water body with a high-risk level in the target water body area in a cyanobacteria bloom monitoring method of this application;

[0034] Figure 12 This is a schematic diagram of a cyanobacteria bloom monitoring system of this application. Detailed implementation manners

[0035] Next, the technical solutions in the embodiments of this application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative efforts shall fall within the protection scope of this application.

[0036] It should be understood that the schematic drawings are not drawn to actual scale. The flowcharts used in this application illustrate the operations implemented according to some embodiments of this application. It should be understood that the operations in the flowcharts may not be implemented in sequence, and steps without logical context relationships may be reversed or implemented simultaneously. In addition, those skilled in the art can add one or more other operations to the flowchart or remove one or more operations from the flowchart under the guidance of the content of this application.

[0037] Referring to "embodiment" herein means that the specific features, structures, or characteristics described in conjunction with the embodiment may be included in at least one embodiment of this application. The phrase appears at various positions in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art explicitly and implicitly understand that the embodiments described herein can be combined with other embodiments.

[0038] Embodiment 1

[0039] Please refer to Figure 1 , Figure 1 This is a flowchart of a cyanobacteria bloom monitoring method of this application.

[0040] This application provides a cyanobacteria bloom monitoring method, which specifically includes the following steps:

[0041] S1: Obtain the remote sensing image of the monitoring area and extract the multi-band remote sensing reflectance value of the remote sensing image;

[0042] S2: Based on the multi-band remote sensing reflectance value of the remote sensing image, extract the target water body remote sensing image containing only the target water body area based on a preset water body classification model;

[0043] S3: By performing a correlation analysis on the multi - band remote sensing reflectance values and chlorophyll a concentration of the remote sensing image of the target water body, screen the chlorophyll a sensitive bands of the remote sensing image of the target water body;

[0044] S4: Based on the remote sensing reflectance values of the chlorophyll a sensitive bands and a preset chlorophyll a concentration inversion model, obtain the chlorophyll a inversion concentration;

[0045] S5: Based on the multi - band remote sensing reflectance values of the remote sensing image of the target water body and a preset normalized difference vegetation index calculation formula, analyze the vegetation growth situation of the target water body to obtain the normalized difference vegetation index;

[0046] S6: Based on the chlorophyll a inversion concentration and the normalized difference vegetation index, construct a dynamic calibration model for the cyanobacteria bloom classification threshold, and classify the cyanobacteria bloom risk of the target water body area;

[0047] S7: According to the cyanobacteria bloom risk classification result of the target water body area, generate a warning instruction corresponding to the risk level, and transmit the warning instruction to the cyanobacteria bloom warning device.

[0048] Compared with the prior art, this solution determines the target water body area of the monitoring area based on the remote sensing image of the monitoring area, and obtains the chlorophyll a inversion concentration and the normalized difference vegetation index of the target water body area according to the remote sensing image of the target water body in the target water body area. Secondly, by using the correlation characteristics of chlorophyll a concentration and the normalized difference vegetation index to construct an unsupervised dynamic calibration model for the cyanobacteria bloom classification threshold, the dynamic adjustment of the cyanobacteria bloom classification threshold can be realized, thereby eliminating the subjective deviation of manual weighting and effectively improving the real - time performance, accuracy and efficiency of cyanobacteria bloom monitoring and warning.

[0049] The cyanobacteria bloom monitoring method of the present invention can be executed by the following computer system, which includes a cyanobacteria bloom monitoring database server, a data acquisition server and a cyanobacteria bloom monitoring server. The cyanobacteria bloom monitoring database server is used to store information such as the remote sensing image of the monitoring area, the length, width and area of the target ecological corridor, and the height and base area of each building in the buffer interface, so as to construct the cyanobacteria bloom monitoring database.

[0050] The data acquisition server is used to obtain information such as the remote sensing image of the monitoring area from the cyanobacteria bloom monitoring database server and send it to the cyanobacteria bloom monitoring server for processing.

[0051] The cyanobacteria bloom monitoring server executes the cyanobacteria bloom monitoring method of the present invention, extracts the multi-band remote sensing reflectance values of the remote sensing images of the monitoring area, and based on the water body classification model, extracts the target water body remote sensing images that only contain the target water body area; through the chlorophyll a sensitive band in the target water body remote sensing image that is sensitive to the chlorophyll a concentration, the chlorophyll a inversion concentration is obtained based on the chlorophyll a concentration inversion model; the normalized difference vegetation index is calculated through the multi-band remote sensing reflectance values of the target water body remote sensing image; according to the chlorophyll a inversion concentration and the normalized difference vegetation index, a dynamic calibration model for the cyanobacteria bloom grading threshold is constructed, and the cyanobacteria bloom risk grading is carried out for the target water body area; according to the cyanobacteria bloom risk grading result of the target water body area, a warning instruction corresponding to the risk level is generated, and the warning instruction is transmitted to the cyanobacteria bloom warning device.

[0052] For step S1, in response to the call instruction of the cyanobacteria bloom monitoring server to the cyanobacteria bloom monitoring database server, the remote sensing image of the monitoring area is obtained from the cyanobacteria bloom monitoring database server and the multi-band remote sensing reflectance values of the remote sensing image are extracted, and the multi-band remote sensing reflectance values are stored in the cyanobacteria bloom monitoring database server. The remote sensing images include: unmanned aerial vehicle (UAV) remote sensing images and satellite remote sensing images. In this embodiment, the UAV remote sensing image is a multi-spectral remote sensing image with a spatial resolution better than 10 m, including: Blue band, Green band, Red band, Red Edge band, and NIR band, which can be obtained from a UAV platform or relevant scientific research institutions.

[0053] Before extracting the multi-band remote sensing reflectance values of the UAV remote sensing image, preprocessing operations such as format conversion and distortion correction can be performed on the UAV remote sensing image, and then radiometric calibration and atmospheric correction are performed on the UAV remote sensing image to complete the extraction of the multi-band remote sensing reflectance values of the UAV remote sensing image; after radiometric calibration and atmospheric correction of the UAV remote sensing image, orthorectification, image color homogenization, and image mosaicking and other processes can also be performed. Among them, format conversion refers to converting the original UAV sensor data such as RAW format into a general raster format; distortion correction is used to eliminate the optical distortion of the UAV lens to ensure the geometric fidelity of the UAV remote sensing image; radiometric calibration converts the original digital (Digital Number, DN) values of the original UAV remote sensing image into absolute radiance or apparent reflectance; atmospheric correction is to remove the influence of atmospheric scattering, absorption, etc., and convert the apparent reflectance into the remote sensing reflectance value; orthorectification is to eliminate the geometric deformation caused by the terrain undulation of the monitoring area and generate an orthoimage; image color homogenization is to adjust the radiometric consistency between multiple UAV remote sensing images such as color balance and reduce the color difference in the splicing area; image mosaicking is to splice multiple orthoimages into a complete area coverage.

[0054] The satellite remote sensing image is Sentinel-2A remote sensing image, with a temporal resolution of 5 days and a spatial resolution of 10m, 20m or 60m. The Sentinel-2A remote sensing image includes bands such as Band2, Band3, Band4, Band5, Band6, Band7, Band8 and Band10, and can be obtained from the official website of the European Space Agency.

[0055] Operations such as radiometric calibration, geometric precise correction, orthorectification, image registration and atmospheric correction can be sequentially performed on the satellite remote sensing image, so as to complete the extraction of the remote sensing reflectance value of the satellite remote sensing image. Among them, radiometric calibration is to convert the original digital value recorded by the sensor into radiance or apparent reflectance; the purpose of geometric precise correction is to eliminate geometric distortion caused by sensor attitude, platform movement or terrain undulation, and ensure the accurate spatial position of the image; orthorectification is to correct the displacement caused by terrain such as projection difference in combination with the digital elevation model of the monitoring area, and ensure that the image has the characteristics of orthographic projection; image registration is to align remote sensing images of multiple time phases, multiple sensors or different bands to a unified coordinate system, and ensure that the data is consistent in space. Atmospheric correction is to remove the influence of atmospheric scattering, absorption, etc., and convert the apparent reflectance into the true surface reflectance.

[0056] For step S2, in response to the call instruction of the cyanobacteria bloom monitoring server to the cyanobacteria bloom monitoring database server, obtain the multi-band remote sensing reflectance values of the remote sensing image from the cyanobacteria bloom monitoring database server, and based on a preset water body classification model, extract the target water body remote sensing image that only contains the target water body area, and store the target water body remote sensing image in the cyanobacteria bloom monitoring database server.

[0057] In this embodiment, please refer to Figure 2 , Figure 2 which is the method flowchart for constructing the water body classification model in a cyanobacteria bloom monitoring method of the present application. Constructing the water body classification model includes:

[0058] S21: Obtain the historical remote sensing images of the monitoring area, where the historical remote sensing images include the multi-band remote sensing reflectance values corresponding to water bodies and non-water bodies;

[0059] S22: Perform annotation on the historical remote sensing images for water bodies and non-water bodies to obtain standard historical remote sensing images;

[0060] S23: Use the multi-band remote sensing reflectance values of the historical remote sensing images and the standard historical remote sensing images as the second sample data, and divide the second sample data into a second training sample data set and a second test sample data set according to a ratio of 8:2;

[0061] S24: Use the multi-band remote sensing reflectance values of the second training sample dataset as input features, and the standard historical remote sensing images of the second training sample dataset as target variables, and construct the water body classification model based on a machine learning algorithm;

[0062] S25: Input the multi-band remote sensing reflectance values of the second test sample dataset into the water body classification model to obtain the water body classification test results corresponding to the second test sample dataset;

[0063] S26: Evaluate the accuracy of the water body classification model by counting the number of correctly classified water bodies and non-water bodies in the water body classification test results to obtain the water body classification model accuracy evaluation result;

[0064] S27: When the water body classification model accuracy evaluation result is lower than the preset water body classification model accuracy threshold, use the second optimization algorithm to adjust the hyperparameters of the water body classification model.

[0065] For step S21, the historical remote sensing images of the monitoring area refer to the UAV remote sensing images or satellite remote sensing images of the monitoring area obtained by remote sensing technology at a certain past time point or time period. Among them, the historical remote sensing images include the multi-band remote sensing reflectance values corresponding to water bodies and non-water bodies. The types of water bodies include general algal blooms, aquatic vegetation, and normal water bodies, etc. The general algal bloom refers to the surface foam or film formed in the water body due to the large reproduction of algae; the aquatic vegetation refers to the plants growing in the water; the normal water body refers to the water body that has not been significantly polluted or abnormally affected. The non-water body refers to the part of the earth's surface other than the water body, including land, vegetation, buildings, etc. The multi-band remote sensing reflectance values corresponding to water bodies and non-water bodies refer to the specific reflectance values of water bodies and non-water bodies in different bands. These values constitute the spectral characteristics of water bodies and non-water bodies in historical remote sensing images and can be used to classify water bodies and non-water bodies.

[0066] For step S22, software such as ArcGIS or QGIS can be used to label water bodies and non-water bodies in the historical remote sensing images based on manual visual interpretation, such as by information such as color differences and texture features. Specifically, water bodies are labeled as 1 and non-water bodies are labeled as 0. Of course, in other embodiments, the labels of water bodies and non-water bodies can also be generated in the historical remote sensing images by calling the multi-band remote sensing reflectance values corresponding to water bodies and non-water bodies to calculate spectral indices, such as the Normalized Difference Water Index (NDWI) and the Modified Normalized Difference Water Index (MNDWI).

[0067] For steps S23 - 24, the second training sample dataset is used to construct a water body classification model, enabling the water body classification model to learn the spectral features of water bodies and non - water bodies. The second test sample dataset is used to evaluate the performance of the water body classification model. The machine learning algorithms include Support Vector Machine (SVM) and Random Forest.

[0068] By using the multi - band remote sensing reflectance values of the second training sample dataset to form a multi - dimensional feature vector as the input feature, and the standard historical remote sensing images of the second training sample dataset as the target variable to provide a supervision signal, machine learning algorithms such as SVM or Random Forest are used to establish a mapping relationship between the input feature and the target variable. By iteratively adjusting the model hyperparameters, such as the penalty parameter and kernel function parameter in SVM or the number and depth of trees in Random Forest, etc., the difference between the predicted label and the true label is minimized to construct the water body classification model. Among them, the predicted label refers to the water body classification training result corresponding to the second training sample dataset output by the model during the training process, and the true label refers to the standard historical remote sensing images of the second training sample dataset.

[0069] For step S26, please refer to Figure 3 , Figure 3 which is the flowchart of the method for evaluating the accuracy of the water body classification model in a cyanobacteria bloom monitoring method of this application. The accuracy of the water body classification model is evaluated by counting the number of correctly classified water bodies and non - water bodies in the water body classification test results, and the water body classification model accuracy evaluation result is obtained, including:

[0070] S261: Randomly generate a number of test points in the water body classification test results;

[0071] S262: Count the number of test points with correctly classified water bodies and the number of test points with correctly classified non - water bodies in the water body classification test results;

[0072] S263: Based on the total number of test points, the number of test points with correctly classified water bodies, and the number of test points with correctly classified non - water bodies, the water body classification model accuracy evaluation result is obtained according to the following formula:

[0073]

[0074] In the formula, OA is the water body classification model accuracy evaluation result, x 11 is the number of test points with correctly classified water bodies, x 22 is the number of test points with correctly classified non - water bodies, and N is the total number of test points.

[0075] For steps S261 - S264, geographic information system software such as ArcGIS can be used to read the water body classification test results, and tools such as creating random points can be used to randomly generate the test points. In this embodiment, the total number of the test points can be set to 200. To avoid the random points being too dense, the minimum distance between two random test points can also be set.

[0076] The judgment rule can specifically be that those skilled in the art, based on the inventive concept of this application and combined with the common general knowledge in the art, manually judge whether each test point in the water body classification test results belongs to the water body or non - water body, and perform statistical calculations on the number of test points with correct water body classification and the number of test points with correct non - water body classification.

[0077] In other embodiments, coefficients such as Kappa and F1 - Score can also be calculated to evaluate the accuracy of the water body classification model, and the water body classification model accuracy evaluation result can be obtained.

[0078] For step S27, the water body classification model accuracy threshold is preferably set to 90%, and the second optimization algorithm can be iterative algorithms such as grid search method, Bayesian optimization, genetic algorithm, and gradient boosting. Of course, the water body classification model accuracy threshold can be adaptively modified according to actual needs.

[0079] For step S3, in response to the call instruction of the cyanobacteria bloom monitoring server to the cyanobacteria bloom monitoring database server, the multi - band remote sensing reflectance values of the target water body remote sensing image are obtained from the cyanobacteria bloom monitoring database server, the chlorophyll a sensitive bands of the target water body remote sensing image are screened, and the remote sensing reflectance values of the chlorophyll a sensitive bands sensitive to the chlorophyll a concentration are stored in the cyanobacteria bloom monitoring database server.

[0080] In one embodiment, please refer to Figure 4 , Figure 4 is the method flow chart for screening chlorophyll a sensitive bands in a cyanobacteria bloom monitoring method of this application. The screening of the chlorophyll a sensitive bands of the target water body remote sensing image by performing a correlation analysis on the multi - band remote sensing reflectance values of the target water body remote sensing image and the chlorophyll a concentration includes:

[0081] S31: Obtain the first sample data of the target water body area at different historical monitoring times, where the first sample data includes the historical monitoring concentration of chlorophyll a and the historical target water body remote sensing image;

[0082] S32: Calculate the Pearson correlation coefficient between the remote sensing reflectance of each band in the historical target water body remote sensing image and the historical monitoring concentration of chlorophyll a according to the following formula:

[0083]

[0084] In the formula, r j represents the Pearson correlation coefficient of the j-th band, and x ji represents the remote sensing reflectance of the j-th band in the historical target water body remote sensing image of the i-th first sample data, represents the average value of the remote sensing reflectance of the j-th band, and y i represents the historical monitoring concentration of chlorophyll a of the i-th first sample data, represents the average value of the historical monitoring concentration of chlorophyll a, and n is the number of the first sample data;

[0085] S33: According to the Pearson correlation coefficients of the remote sensing reflectances of the respective bands and the historical monitoring concentration of chlorophyll a, based on a preset sensitive band screening criterion, select the corresponding bands as the chlorophyll a sensitive bands.

[0086] For steps S31 - S33, the first sample data of the target water body area at different historical monitoring times refers to the historical monitoring concentration of chlorophyll a and the historical target water body remote sensing image obtained at the same past time point or time period, and the historical monitoring concentration of chlorophyll a and the historical target water body remote sensing image are strictly corresponding in terms of spatial position (the same geographical coordinate point) and time dimension (the same monitoring moment), and can be obtained through channels such as local environmental monitoring agencies or academic research institutions.

[0087] The specific sensitive band screening criterion is to perform a descending order sorting on each band according to the Pearson correlation coefficients of the remote sensing reflectances of the respective bands and the historical monitoring concentration of chlorophyll a, and select the top five bands as the chlorophyll a sensitive bands.

[0088] In other embodiments, a correlation threshold can also be preset according to actual requirements or statistical criteria, and select the bands whose Pearson correlation coefficients of the remote sensing reflectances of the respective bands and the historical monitoring concentration of chlorophyll a are greater than the preset correlation threshold as the chlorophyll a sensitive bands. Exemplarily, in the study of Qiandao Lake, it is found through Pearson analysis that the (B4 + B2) / B3 band combination is significantly correlated with the chlorophyll a concentration, and the determination coefficient r 2 = 0.7366, that is, the Pearson correlation coefficient r≈0.858. If the correlation threshold is set to 0.8 (corresponding to r 2 ≥0.64), then this band combination meets the screening criterion.

[0089] For step S4, in response to the call instruction of the cyanobacteria bloom monitoring server to the cyanobacteria bloom monitoring database server, obtain the remote sensing reflectance value of the chlorophyll a sensitive band sensitive to the chlorophyll a concentration from the cyanobacteria bloom monitoring database server, and based on a preset chlorophyll a concentration inversion model, obtain the chlorophyll a inversion concentration and store it in the cyanobacteria bloom monitoring database server.

[0090] In this embodiment, please refer to Figure 5 , Figure 5 which is the method flowchart for constructing the chlorophyll a concentration inversion model in a cyanobacteria bloom monitoring method of this application. Constructing the chlorophyll a concentration inversion model includes:

[0091] S41: Divide the first sample data into a first training sample data set and a first test sample data set according to a ratio of 8:2;

[0092] S42: Use the historical monitoring concentration of chlorophyll a and the remote sensing reflectance value of the sensitive band in the historical target water body remote sensing image of the first training sample data set as input features, and construct the chlorophyll a concentration inversion model based on a machine learning algorithm;

[0093] S43: Use the remote sensing reflectance of the sensitive band in the historical target water body remote sensing image of the first test sample data set as an input feature, and based on the chlorophyll a concentration inversion model, obtain the chlorophyll a inversion concentration test result corresponding to the first test sample data set;

[0094] S44: According to the chlorophyll a inversion concentration test result and the historical monitoring concentration of chlorophyll a in the first test sample data set, evaluate the accuracy of the chlorophyll a concentration inversion model to obtain the chlorophyll a concentration inversion model accuracy evaluation result.

[0095] For step S41, the first training sample data set is used to construct the chlorophyll a concentration inversion model, so that the chlorophyll a concentration inversion model learns the relationship between the remote sensing reflectance value of the sensitive band in the historical target water body remote sensing image and the chlorophyll a concentration. The first test sample data set is used to evaluate the performance of the chlorophyll a concentration inversion model.

[0096] For step S42, the machine learning model includes Support Vector Machine (SVM), Random Forest, XGBoost, etc. By constructing a multi-dimensional feature vector with the historical monitoring concentration of chlorophyll a in the first training sample dataset and the remote sensing reflectance values of sensitive bands in the historical target water body remote sensing images as input features, and iteratively adjusting the model hyperparameters based on machine learning algorithms such as SVM, Random Forest, or XGBoost, such as the penalty parameter and kernel function parameter in SVM or the number and depth of trees in Random Forest, etc., the chlorophyll a concentration inversion model is constructed.

[0097] For step S44, based on the chlorophyll a inversion concentration test results and the historical monitoring concentration of chlorophyll a in the first test sample dataset, the accuracy of the chlorophyll a concentration inversion model is evaluated to obtain the accuracy evaluation result of the chlorophyll a concentration inversion model, including:

[0098] Based on the chlorophyll a inversion concentration test results and the historical monitoring concentration of chlorophyll a in the first test sample dataset, and according to the coefficient of determination calculation formula, the accuracy evaluation result of the chlorophyll a concentration inversion model is obtained. The coefficient of determination calculation formula is:

[0099]

[0100] In the formula, R 2 is the accuracy evaluation result of the chlorophyll a concentration inversion model, p i is the predicted chlorophyll a concentration, is the historical monitoring concentration of chlorophyll a in the first training sample dataset, is the average value of the historical monitoring concentration of chlorophyll a in the first training sample dataset, and n is the size of the first sample data.

[0101] Or based on the chlorophyll a inversion concentration test results and the historical monitoring concentration of chlorophyll a in the first test sample dataset, and according to the root mean square error calculation formula, the accuracy evaluation result of the chlorophyll a concentration inversion model is obtained. The root mean square error calculation formula is:

[0102]

[0103] In the formula, RMSE is the accuracy evaluation result of the chlorophyll a concentration inversion model, z i is the chlorophyll a inversion concentration test result, is the historical monitoring concentration of chlorophyll a in the first test sample dataset, and n is the size of the first sample data.

[0104] When obtaining the accuracy evaluation result of the chlorophyll a concentration inversion model based on the determination coefficient calculation formula, the accuracy threshold of the chlorophyll a concentration inversion model can be set to 0.7. If the accuracy evaluation result of the chlorophyll a concentration inversion model is lower than 0.7, adjust the hyperparameters of the chlorophyll a concentration inversion model through the first optimization algorithm, such as the penalty parameter and kernel function parameter in the support vector machine (SVM) or the number and depth of trees in the random forest, etc.;

[0105] When obtaining the accuracy evaluation result of the chlorophyll a concentration inversion model based on the root mean square error calculation formula, the accuracy threshold of the chlorophyll a concentration inversion model can be set to 0.1. If the accuracy evaluation result of the chlorophyll a concentration inversion model is higher than 0.1, adjust the hyperparameters of the chlorophyll a concentration inversion model through the first optimization algorithm, such as the penalty parameter and kernel function parameter in the support vector machine (SVM) or the number and depth of trees in the random forest, etc.

[0106] The first optimization algorithm can select iterative algorithms such as grid search method, Bayesian optimization, genetic algorithm, etc. Of course, in other embodiments, functions such as mean squared error loss function and mean absolute error loss function can also be selected to evaluate the accuracy of the chlorophyll a concentration inversion model. According to the performance requirements of the chlorophyll a concentration inversion model and the actually selected formula for evaluating the accuracy of the chlorophyll a concentration inversion model, the accuracy threshold of the chlorophyll a concentration inversion model can be adaptively modified.

[0107] For step S5, in response to the call instruction of the cyanobacterial bloom monitoring server to the cyanobacterial bloom monitoring database server, obtain the multi-band remote sensing reflectance values of the target water body remote sensing image from the cyanobacterial bloom monitoring database server, analyze the vegetation growth situation of the target water body based on the preset normalized difference vegetation index (NDVI) calculation formula, obtain the normalized difference vegetation index, and store the normalized difference vegetation index in the cyanobacterial bloom monitoring database server.

[0108] The multi-band remote sensing reflectance values of the target water body remote sensing image include: remote sensing reflectance values in the near-infrared band and the red band. The analyzing the vegetation growth situation of the target water body based on the multi-band remote sensing reflectance values of the target water body remote sensing image and obtaining the normalized difference vegetation index according to the preset normalized difference vegetation index calculation formula includes:

[0109] According to the remote sensing reflectance values in the near-infrared band and the red band, obtain the normalized difference vegetation index according to the normalized difference vegetation index calculation formula, and the normalized difference vegetation index calculation formula is:

[0110]

[0111] wherein, NDVI is the normalized difference vegetation index, and R nir is the remote sensing reflectance value of the near-infrared band, and R red is the remote sensing reflectance value of the red light band.

[0112] For step S6, in response to the call instruction of the cyanobacteria bloom monitoring server to the cyanobacteria bloom monitoring database server, obtain the chlorophyll a inversion concentration and the normalized difference vegetation index from the cyanobacteria bloom monitoring database server, construct a dynamic calibration model for the cyanobacteria bloom grading threshold, and perform cyanobacteria bloom risk grading on the target water body area.

[0113] In this embodiment, please refer to Figure 6 , Figure 6 which is the method flow chart for performing cyanobacteria bloom risk grading on the target water body area in a cyanobacteria bloom monitoring method of the present application. The constructing a dynamic calibration model for the cyanobacteria bloom grading threshold and performing cyanobacteria bloom risk grading on the target water body area according to the chlorophyll a inversion concentration and the normalized difference vegetation index includes:

[0114] S61: Randomly collect the target water body area to obtain a number of the chlorophyll a inversion concentrations and normalized difference vegetation indices;

[0115] S62: Perform a correlation analysis on the normalized difference vegetation index and the chlorophyll a inversion concentration. If the normalized difference vegetation index and the chlorophyll a inversion concentration are linearly correlated, establish a linear model of the normalized difference vegetation index and the chlorophyll a inversion concentration as the dynamic calibration model for the cyanobacteria bloom grading threshold;

[0116] S63: Obtain a preset chlorophyll a concentration grading threshold, and input the chlorophyll a concentration grading threshold into the dynamic calibration model for the cyanobacteria bloom grading threshold to obtain the normalized difference vegetation index grading threshold;

[0117] S64: Perform cyanobacteria bloom risk grading on the target water body area according to the chlorophyll a inversion concentration and the chlorophyll a concentration grading threshold, and the normalized difference vegetation index and the normalized difference vegetation index grading threshold.

[0118] For step S61, randomly sample the target water body area to obtain a number of the retrieved chlorophyll a concentrations and normalized difference vegetation indices (NDVIs). This can be achieved using GIS software such as ArcGIS or QGIS. Specifically: Import the remote sensing image of the target water body in the target water body area into ArcGIS or QGIS; Use the remote sensing image of the target water body as the background to set the sampling area, and use the random point generation tool to randomly generate sampling points within the sampling area; According to the coordinates of the generated sampling points, extract the remote sensing reflectance values at the corresponding positions on the remote sensing image of the target water body, and based on the chlorophyll a concentration inversion model and the NDVI calculation formula, obtain the retrieved chlorophyll a concentrations and NDVIs of each sampling point.

[0119] For step S62, calculate the Pearson correlation coefficient between the NDVI and the retrieved chlorophyll a concentration for correlation analysis. When the Pearson correlation coefficient between the NDVI and the retrieved chlorophyll a concentration is greater than 0, it indicates that the NDVI and the retrieved chlorophyll a concentration are positively correlated.

[0120] Of course, a significance test can also be performed on the NDVI and the retrieved chlorophyll a concentration. For example, calculate the significance level of the NDVI and the retrieved chlorophyll a concentration through a t-test or an F-test. When the significance level of the NDVI and the retrieved chlorophyll a concentration is less than the preset significance level threshold, it is considered that the linear relationship between the NDVI and the retrieved chlorophyll a concentration is significant. Among them, the significance level threshold can be set to 0.05 or 0.01. In other embodiments, the significance level threshold can be adaptively adjusted according to actual needs.

[0121] The expression of the dynamic calibration model for the cyanobacterial bloom grading threshold is:

[0122] Chla = a × NDVI + b

[0123] In the formula, Chla is the retrieved chlorophyll a concentration, NDVI is the normalized difference vegetation index, a is the regression coefficient, indicating the expected change in the chlorophyll a concentration for every 1 unit increase in NDVI, and b is the intercept, indicating the estimated chlorophyll a concentration when NDVI is 0.

[0124] The regression coefficient a of the dynamic calibration model for the cyanobacterial bloom grading threshold can be obtained by the least squares method. The specific formula is:

[0125]

[0126] In the formula, a is the regression coefficient of the dynamic calibration model for the cyanobacterial bloom grading threshold, NDVI iis the normalized difference vegetation index (NDVI) of the i-th sampling point, is the average value of the normalized difference vegetation index, Chla i is the retrieved concentration of chlorophyll a at the i-th sampling point, is the average value of the retrieved concentration of chlorophyll a, and m is the number of sampling points.

[0127] The intercept b of the dynamic calibration model for the cyanobacterial bloom classification threshold can be obtained through the following formula:

[0128]

[0129] In the formula, b is the intercept of the dynamic calibration model for the cyanobacterial bloom classification threshold, is the average value of the normalized difference vegetation index, is the average value of the retrieved concentration of chlorophyll a, and a is the regression coefficient of the dynamic calibration model for the cyanobacterial bloom classification threshold.

[0130] After obtaining the dynamic calibration model for the cyanobacterial bloom classification threshold, the model can be tested by means of residual analysis or R-square value calculation to ensure the fitting effect and prediction ability of the model.

[0131] For step S63, the chlorophyll a concentration classification threshold is a value that divides the chlorophyll a content in the water body into different levels or ranges, and is used to evaluate the eutrophication status of the water body, the occurrence risk of cyanobacterial blooms, and the water quality status. The normalized difference vegetation index classification threshold is a value that divides the normalized difference vegetation index in the water body into different levels or ranges, and is used to evaluate the vegetation coverage of the water body. The chlorophyll a concentration classification threshold can be obtained through channels such as the water quality monitoring standards announced by the environmental protection department, such as the surface water environmental quality standards. It can be seen that by utilizing the correlation characteristics between the chlorophyll a concentration and the normalized difference vegetation index, the dynamic calibration model for the cyanobacterial bloom classification threshold can be adjusted in real time, and then the dynamic adjustment of the normalized difference vegetation index classification threshold can be realized, thereby eliminating the subjective deviation of artificial weighting and effectively improving the real-time performance, accuracy, and efficiency of cyanobacterial bloom monitoring and early warning.

[0132] For step S64, in this embodiment, the cyanobacterial bloom risk classification of the target water body area is specifically divided into no risk (Level I), lower risk (Level II), low risk (Level III), medium risk (Level IV), and high risk (Level V). The chlorophyll a concentration classification threshold and the normalized difference vegetation index classification threshold are shown in Table 1 below:

[0133] Table 1:

[0134]

[0135]

[0136] When the retrieved chlorophyll a concentration is in the range of 0 - 10 μg / L and the normalized vegetation index value of the target water body area is in the range of [-1, -0.07), it is determined that the risk level of cyanobacterial bloom outbreak in the target water body area is risk-free. At this time, the water color of the target water body area is normal, there is no algal aggregation on the water surface, and algal particles can hardly be identified in the water. As Figure 7 shown.

[0137] When the retrieved chlorophyll a concentration is in the range of 10 - 15 μg / L and the normalized vegetation index value of the target water body area is in the range of [-0.07, -0.05), it is determined that the target water body area has a low risk. At this time, the water color of the target water body area is light yellowish green, and there is no algal aggregation on the water surface. As Figure 8 shown.

[0138] When the retrieved chlorophyll a concentration is in the range of 15 - 50 μg / L and the normalized vegetation index value of the target water body area is in the range of [-0.05, 0.07), it is determined that the target water body area has a low risk. At this time, the water color of the target water body area is yellowish green, and a small amount of algal particles can be distinguished in the water. As Figure 9 shown.

[0139] When the retrieved chlorophyll a concentration is in the range of 50 - 100 μg / L and the normalized vegetation index value of the target water body area is in the range of [0.07, 0.28), it is determined that the target water body area has a medium risk. At this time, the water color of the target water body area is yellowish green, there are sporadic algal aggregations on the water surface, or suspended algae are clearly visible in the water. As Figure 10 shown.

[0140] When the retrieved chlorophyll a concentration is greater than 100 μg / L and the normalized vegetation index value of the target water body area is greater than 0.28, it is determined that the target water body area has a high risk. At this time, the water color of the target water body area is yellowish green, there are algal aggregations on the water surface, floating in patches, or suspended algae are clearly visible in the water. As Figure 11 shown.

[0141] When the judgment criteria of the retrieved chlorophyll a concentration and the normalized vegetation index are inconsistent, the higher level is selected. For example, when the retrieved chlorophyll a concentration is in the range of 15 - 50 μg / L (low risk), but the normalized vegetation index is in the range of [0.07, 0.28) (medium risk), the risk level where the normalized vegetation index is located is used as the evaluation result, that is, the target water body area has a medium risk.

[0142] Of course, in other embodiments, the chlorophyll a concentration classification threshold can be adaptively modified according to the actual water quality monitoring standards. At the same time, the normalized difference vegetation index (NDVI) classification threshold can also be adaptively modified according to the chlorophyll a concentration classification threshold and the cyanobacterial bloom classification threshold dynamic calibration model.

[0143] For step S7, in response to the call instruction of the cyanobacterial bloom monitoring server to the cyanobacterial bloom monitoring database server, according to the cyanobacterial bloom risk classification result of the target water body area, an early warning instruction corresponding to the risk level is generated, and the early warning instruction is transmitted to the cyanobacterial bloom early warning device.

[0144] In this embodiment, the early warning instruction includes the generation time, the geographical coordinates of the target water body area, the risk level, key parameters, and early warning suggestions, etc. Among them, the risk level includes no risk (Level I), lower risk (Level II), low risk (Level III), medium risk (Level IV), and high risk (Level V). The key parameters include the chlorophyll a inversion concentration and the normalized difference vegetation index, as well as the chlorophyll a concentration classification threshold and the normalized difference vegetation index classification threshold.

[0145] When the cyanobacterial bloom risk classification result of the target water body area is no risk, the early warning suggestion can be to maintain routine monitoring, record water quality data, and there is no need to initiate an emergency response. When the cyanobacterial bloom risk classification result of the target water body area is lower risk, the early warning suggestion can be to increase the inspection frequency and submit a monitoring report to the management department. When the cyanobacterial bloom risk classification result of the target water body area is low risk, the early warning suggestion can be to initiate early warning monitoring (once a day), restrict the surrounding agricultural / industrial sewage discharge, and issue a water quality safety reminder to the public. When the cyanobacterial bloom risk classification result of the target water body area is medium risk, the early warning suggestion can be to initiate emergency monitoring (once an hour), close some waters, deploy salvage equipment, and prepare for algae removal operations. When the cyanobacterial bloom risk classification result of the target water body area is high risk, the early warning suggestion can be to issue a red alert, evacuate the surrounding personnel, initiate cross-departmental linkage (environmental protection, water conservancy, emergency), and implement emergency treatment.

[0146] The cyanobacterial bloom monitoring server can push the early warning instruction to the cyanobacterial bloom early warning device through a wireless network. The cyanobacterial bloom early warning device can be a water quality monitoring buoy, a shore display screen, a mobile terminal, etc. Of course, the cyanobacterial bloom monitoring server can also be docked with the early warning device system.

[0147] In other embodiments, the early warning instruction can be adaptively modified according to the actual monitoring requirements.

[0148] Embodiment 2

[0149] Please refer toFigure 12 , Figure 12 This is a schematic diagram of a cyanobacterial bloom monitoring system of the present application.

[0150] The present application also provides a cyanobacterial bloom monitoring system, including:

[0151] Remote sensing image acquisition module 1: It is used to acquire the remote sensing image of the monitoring area and extract the multi-band remote sensing reflectance value of the remote sensing image;

[0152] Target water body remote sensing image extraction module 2: It is used to extract the target water body remote sensing image containing only the target water body area based on the multi-band remote sensing reflectance value of the remote sensing image and a preset water body classification model;

[0153] Sensitive band screening module 3: It is used to screen the chlorophyll a sensitive band of the target water body remote sensing image by performing a correlation analysis on the multi-band remote sensing reflectance value of the target water body remote sensing image and the chlorophyll a concentration;

[0154] Chlorophyll a concentration inversion module 4: It is used to obtain the chlorophyll a inversion concentration based on the remote sensing reflectance value of the chlorophyll a sensitive band and a preset chlorophyll a concentration inversion model;

[0155] Normalized difference vegetation index calculation module 5: It is used to analyze the vegetation growth situation of the target water body based on the multi-band remote sensing reflectance value of the target water body remote sensing image and a preset normalized difference vegetation index calculation formula, and obtain the normalized difference vegetation index;

[0156] Target water body area grading module 6: It is used to construct a dynamic calibration model for cyanobacterial bloom grading thresholds based on the chlorophyll a inversion concentration and the normalized difference vegetation index, and perform cyanobacterial bloom risk grading on the target water body area;

[0157] Early warning instruction issuing module 7: It is used to generate an early warning instruction corresponding to the risk level according to the cyanobacterial bloom risk grading result of the target water body area, and transmit the early warning instruction to the cyanobacterial bloom early warning device.

[0158] It should be noted that the data obtained when the cyanobacterial bloom monitoring system provided by the present application implements a cyanobacterial bloom monitoring method are stored in the storage of this system one by one. When relevant calculations are required, the data required for the calculations can be directly obtained from the storage correspondingly.

[0159] It should also be noted that when implementing a cyanobacteria bloom monitoring method using the cyanobacteria bloom monitoring system provided in the above embodiments, only the division of the above functional modules is used for illustration. In actual applications, the above functions can be allocated to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the cyanobacteria bloom monitoring system provided in the above embodiments and the cyanobacteria bloom monitoring method in Embodiment 1 belong to the same inventive concept. For the implementation process, please refer to the method embodiments and will not be elaborated here.

[0160] Based on the same inventive concept, the present application also provides an electronic device, which can be a server, a desktop computing device or a mobile computing device (for example, a laptop computing device, a handheld computing device, a tablet computer, a netbook, etc.). The device includes one or more processors and a memory. The processor is used to execute a program to implement the above cyanobacteria bloom monitoring method; the memory is used to store a computer program executable by the processor.

[0161] The present application is not limited to the above embodiments. If various changes or deformations to the present application do not depart from the spirit and scope of the present application, and if these changes and deformations fall within the scope of the claims of the present application and equivalent technical scope, then the present application also intends to include these changes and deformations.

Claims

1. A method for monitoring cyanobacterial blooms, characterized in that: The following steps are involved: Acquire a remote sensing image of a monitored area and extract multi-band remote sensing reflectance values ​​of the remote sensing image; Extracting a target water body remote sensing image containing only the target water body area based on the multi-band remote sensing reflectance values ​​of the remote sensing image and a preset water body classification model; By performing correlation analysis on the multi-band remote sensing reflectance values ​​and chlorophyll a concentration of the target water body remote sensing image, the chlorophyll a sensitive band of the target water body remote sensing image is screened; According to the remote sensing reflectance value of the chlorophyll a sensitive band, based on a preset chlorophyll a concentration inversion model, the chlorophyll a inversion concentration is obtained; According to the multi-band remote sensing reflectance value of the remote sensing image of the target water body, based on a preset normalized vegetation index calculation formula, the vegetation growth condition of the target water body is analyzed to obtain a normalized vegetation index; According to the chlorophyll a inversion concentration and the normalized difference vegetation index, a dynamic calibration model for cyanobacteria bloom classification threshold is constructed, and cyanobacteria bloom risk classification is performed on the target water area; According to the cyanobacteria bloom risk grading result of the target water body area, a warning instruction corresponding to the risk level is generated, and the warning instruction is transmitted to the cyanobacteria bloom warning device.

2. A method for monitoring cyanobacteria blooms according to claim 1, characterized in that: The step of constructing a dynamic calibration model for cyanobacteria bloom classification thresholds based on the chlorophyll a inversion concentration and the normalized difference vegetation index, and performing cyanobacteria bloom risk classification on the target water area includes: Randomly sampling the target water area to obtain a number of chlorophyll a inversion concentrations and normalized difference vegetation indexes; Performing a correlation analysis on the normalized vegetation index and the chlorophyll a inversion concentration, if the normalized vegetation index and the chlorophyll a inversion concentration are linearly correlated, establishing a linear model of the normalized vegetation index and the chlorophyll a inversion concentration as a dynamic calibration model for the cyanobacteria bloom classification threshold; Obtaining a preset chlorophyll a concentration classification threshold, and inputting the chlorophyll a concentration classification threshold into the cyanobacteria bloom classification threshold dynamic calibration model to obtain the normalized vegetation index classification threshold; The blue algae bloom risk of the target water area is graded according to the chlorophyll a inversion concentration and the chlorophyll a concentration classification threshold, as well as the normalized difference vegetation index and the normalized difference vegetation index classification threshold.

3. A method for monitoring cyanobacteria blooms according to claim 1, characterized in that: The method of screening the chlorophyll a sensitive band of the target water body remote sensing image by performing a correlation analysis on the multi-band remote sensing reflectance value and the chlorophyll a concentration of the target water body remote sensing image comprises: Acquire first sample data of the target water body area at different historical monitoring times, wherein the first sample data includes historical monitoring concentrations of chlorophyll a and historical remote sensing images of the target water body; The Pearson correlation coefficient between the remote sensing reflectance of each band in the historical target water body remote sensing image and the historical monitoring concentration of chlorophyll a is calculated according to the following formula: In the formula, r j represents the Pearson correlation coefficient of the jth band, x ji represents the remote sensing reflectance of the jth band in the i-th historical target water body remote sensing image of the first sample data, represents the average value of the remote sensing reflectance of the jth band, y i represents the historical monitoring concentration of chlorophyll a of the i-th sample data, represents the average value of the historical monitoring concentration of chlorophyll a, and n is the number of the first sample data; According to the Pearson correlation coefficient between the remote sensing reflectance of each band and the historical monitoring concentration of chlorophyll a, based on the preset sensitive band screening criteria, the corresponding band is selected as the chlorophyll a sensitive band.

4. A method for monitoring cyanobacteria blooms according to claim 3, characterized in that: Constructing the chlorophyll a concentration inversion model includes: Dividing the first sample data into a first training sample data set and a first test sample data set in a ratio of 8:2; Taking the historical monitoring concentration of chlorophyll a in the first training sample data set and the remote sensing reflectance value of the sensitive band in the historical target water body remote sensing image as input features, constructing the chlorophyll a concentration inversion model based on the machine learning algorithm; Using the remote sensing reflectance of the sensitive band in the historical target water body remote sensing image of the first test sample data set as an input feature, and based on the chlorophyll a concentration inversion model, obtaining the chlorophyll a inversion concentration test result corresponding to the first test sample data set; According to the chlorophyll a inversion concentration test result and the historical monitoring concentration of chlorophyll a in the first test sample data set, the accuracy of the chlorophyll a concentration inversion model is evaluated to obtain the chlorophyll a concentration inversion model accuracy evaluation result.

5. A method for monitoring cyanobacteria blooms according to claim 4, characterized in that: The step of evaluating the accuracy of the chlorophyll a concentration inversion model according to the chlorophyll a inversion concentration test result and the chlorophyll a historical monitoring concentration of the first test sample data set to obtain the chlorophyll a concentration inversion model accuracy evaluation result includes: According to the chlorophyll a inversion concentration test result and the chlorophyll a historical monitoring concentration of the first test sample data set, based on the determination coefficient calculation formula, the accuracy evaluation result of the chlorophyll a concentration inversion model is obtained. The determination coefficient calculation formula is: In the formula, R 2 is the accuracy evaluation result of the chlorophyll a concentration inversion model, p i is the chlorophyll a inversion concentration test result, is the historical monitoring concentration of chlorophyll a of the first test sample data set, is the average value of the historical monitoring concentration of chlorophyll a of the first test sample data set, and n is the size of the first sample data; Alternatively, the chlorophyll a concentration inversion model accuracy evaluation result is obtained based on the chlorophyll a inversion concentration test result and the chlorophyll a historical monitoring concentration of the first test sample data set, based on the root mean square error calculation formula, and the root mean square error calculation formula is: Where, RMSE is the accuracy evaluation result of the chlorophyll a concentration inversion model, z i is the chlorophyll a inversion concentration test result, is the historical monitoring concentration of chlorophyll a of the first test sample data set, and n is the size of the first sample data.

6. The method for monitoring cyanobacteria bloom according to claim 1, characterized in that: Constructing the water body classification model includes: Acquire historical remote sensing images of the monitored area, wherein the historical remote sensing images include multi-band remote sensing reflectance values ​​corresponding to water bodies and non-water bodies; annotating water bodies and non-water bodies on the historical remote sensing images to obtain standard historical remote sensing images; The multi-band remote sensing reflectance value of the historical remote sensing image and the standard historical remote sensing image are used as second sample data, and the second sample data is divided into a second training sample data set and a second test sample data set in a ratio of 8:2; The multi-band remote sensing reflectance value of the second training sample data set is used as an input feature, and the standard historical remote sensing image of the second training sample data set is used as a target variable, and the water body classification model is constructed based on a machine learning algorithm; Inputting the multi-band remote sensing reflectance value of the second test sample data set into the water body classification model to obtain a water body classification test result corresponding to the second test sample data set; By counting the number of correctly classified water bodies and non-water bodies in the water body classification test results, the accuracy of the water body classification model is evaluated to obtain the accuracy evaluation result of the water body classification model; When the accuracy evaluation result of the water body classification model is lower than a preset water body classification model accuracy threshold, a second optimization algorithm is used to adjust the hyperparameters of the water body classification model.

7. A method for monitoring cyanobacteria blooms according to claim 6, characterized in that: The water body classification model is evaluated for accuracy by counting the number of correctly classified water bodies and non-water bodies in the water body classification test results to obtain the water body classification model accuracy evaluation result, including: Randomly generate a number of inspection points in the water body classification test results; Counting the number of inspection points with correct water body classification and the number of inspection points with correct non-water body classification in the water body classification test results; According to the total number of the inspection points, the number of inspection points with correct water body classification, and the number of inspection points with correct non-water body classification, the accuracy evaluation result of the water body classification model is obtained based on the following formula: Where OA is the accuracy evaluation result of the water body classification model, x 11 The number of checkpoints that are correctly classified for water bodies, x 22 is the number of test points correctly classified as non-water bodies, and N is the total number of test points.

8. The method for monitoring cyanobacteria bloom according to claim 1, characterized in that: The multi-band remote sensing reflectance values ​​of the target water body remote sensing image include: remote sensing reflectance values ​​of near infrared band and red light band; The step of analyzing the vegetation growth of the target water body based on the multi-band remote sensing reflectance value of the target water body remote sensing image and the preset normalized vegetation index calculation formula to obtain the normalized vegetation index includes: According to the remote sensing reflectance values ​​of the near infrared band and the red light band, the normalized vegetation index is obtained according to the normalized vegetation index calculation formula. The normalized vegetation index calculation formula is: Among them, NDVI is the normalized difference vegetation index, R nir is the remote sensing reflectance value of the near-infrared band, R red is the remote sensing reflectance value of the red light band.

9. A cyanobacteria bloom monitoring system, characterized in that: include: Remote sensing image acquisition module: used to acquire remote sensing images of the monitored area and extract multi-band remote sensing reflectivity values ​​of the remote sensing images; Target water body remote sensing image extraction module: used to extract the target water body remote sensing image containing only the target water body area based on the multi-band remote sensing reflectance value of the remote sensing image and the preset water body classification model; Sensitive band screening module: used to screen the chlorophyll a sensitive band of the target water body remote sensing image by performing correlation analysis on the multi-band remote sensing reflectance value and chlorophyll a concentration of the target water body remote sensing image; Chlorophyll a concentration inversion module: used to obtain the chlorophyll a inversion concentration according to the remote sensing reflectance value of the chlorophyll a sensitive band and based on a preset chlorophyll a concentration inversion model; Normalized vegetation index calculation module: used to analyze the vegetation growth of the target water body according to the multi-band remote sensing reflectance value of the target water body remote sensing image and based on a preset normalized vegetation index calculation formula to obtain a normalized vegetation index; Target water body area classification module: used to construct a dynamic calibration model of cyanobacteria bloom classification threshold according to the chlorophyll a inversion concentration and the normalized vegetation index, and to classify the cyanobacteria bloom risk in the target water body area; The warning instruction issuing module is used to generate a warning instruction corresponding to the risk level according to the cyanobacteria bloom risk classification result of the target water body area, and transmit the warning instruction to the cyanobacteria bloom warning device.

10. An electronic device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, a method for monitoring cyanobacteria blooms as described in any one of claims 1 to 8 is implemented.

Images