Online water quality analysis method and system based on multispectral image analysis model

Abstract

The invention discloses an online water quality analysis method and system based on a multispectral image analysis model, and belongs to the technical field of water quality analysis. And constructing a multispectral image analysis model comprising an analysis model, a classification model, a water body plankton coverage degree inversion model, a water body transparency inversion model, a water surface eutrophication inversion model and a key water quality parameter analysis model. And extracting water quality characteristics by using the analytical model, extracting a water surface line by using the classification model, then inputting current water quality data and the water surface line into an inversion model for water plankton coverage, water transparency and water surface eutrophication inversion and key water quality parameter analysis, and finally fusing an analysis result to output comprehensive water quality evaluation. According to the online water quality analysis method provided by the invention, efficient and real-time water quality monitoring and analysis can be realized, and a scientific basis is provided for water environment management.

Description

Technical Field

[0001] The present invention relates to the technical field of water quality analysis, and more specifically to an online water quality analysis method and system based on a multi-spectral image analysis model. Background Art

[0002] With the rapid development of industrialization and urbanization, the surface water quality of inland rivers, lakes and reservoirs has deteriorated, and environmental problems such as algae blooms and eutrophication have occurred on a large scale, directly affecting the ecological security of water bodies, the safety of drinking water for residents, and social and economic development. Water resources and environmental monitoring are important prerequisites for the planning, management and protection of water bodies and aquatic biological resources. Against the background of global warming and increasing human interference, rapid, efficient and real-time monitoring of water resources and water quality is of great significance to water resource assessment, water pollution control and water safety.

[0003] Traditional water quality monitoring collects samples from the field and then analyzes them in the laboratory to determine their physical and chemical properties. The measurement accuracy is high but time-consuming and labor-intensive. Moreover, these measurement data can only represent local sample point information and cannot quickly reflect the overall water quality status of the water body, and cannot meet the requirements of real-time dynamic water quality monitoring.

[0004] In recent years, the development of UAV remote sensing technology has provided new possibilities for water quality monitoring. As an advanced remote sensing method, multispectral imaging technology can capture the spectral characteristics of water bodies in different bands, which are closely related to various chemical and physical parameters in the water body. By analyzing these spectral data, the water quality of the water body can be inferred.

[0005] However, the current application of multispectral image analysis technology in water quality analysis still has some limitations. First, a single spectral analysis model is difficult to simultaneously deal with different types of water quality problems, such as plankton coverage in water bodies, water transparency inversion, and water surface eutrophication. Second, existing water quality monitoring methods can often only obtain limited water quality parameters and are difficult to fully reflect the complex conditions of water bodies.

[0006] Therefore, how to provide an online water quality analysis method that can comprehensively reflect the complex conditions of water bodies is an urgent problem that technicians in this field need to solve. Summary of the invention

[0007] In view of this, the present invention provides an online water quality analysis method and system based on a multispectral image analysis model, which realizes real-time and accurate analysis of the water quality of the water body by collecting and analyzing the multispectral images of the target water body.

[0008] In order to achieve the above object, the present invention adopts the following technical solution:

[0009] An online water quality analysis method based on a multispectral image analysis model, comprising:

[0010] Collecting a multispectral image of a target water body and preprocessing the multispectral image;

[0011] Constructing a multispectral image analysis model, the multispectral image analysis model includes: an analysis model, a classification model, a water body plankton coverage inversion model, a water body transparency inversion model, a water surface eutrophication inversion model and a water quality parameter analysis model;

[0012] Using the analytical model to extract features from the preprocessed multispectral image, obtaining current water quality data and water body data of the target water body;

[0013] Inputting the water body data into a classification model to extract the water surface line of the target water body;

[0014] Inputting the current water quality data and the water surface line into the water body plankton coverage inversion model, the water body transparency inversion model and the water surface eutrophication inversion model, respectively performing water body plankton coverage inversion, water body transparency inversion and water surface eutrophication inversion to obtain water body plankton coverage inversion result, water body transparency inversion result and water surface eutrophication inversion result;

[0015] The water quality parameter analysis model is used to fuse the water body plankton coverage inversion result, the water body transparency inversion result and the water surface eutrophication inversion result to output the final water quality analysis result.

[0016] Preferably, preprocessing of the multispectral image includes: geometric correction and radiation correction, multispectral image fusion, cloud and shadow removal, noise filtering, and image enhancement.

[0017] Preferably, multispectral image fusion includes:

[0018] Performing image registration on the first image and the second image respectively;

[0019] Extracting features from the first image and the second image after image registration to obtain a first feature;

[0020] Classify the first feature to obtain a classification result;

[0021] Decision-level fusion is performed based on the classification results to obtain a fused multispectral image.

[0022] Preferably, the analytical model is used to extract features from the preprocessed multispectral image to obtain current water quality data and water body data of the target water body, including:

[0023] The preprocessed multispectral image is input into a first feature extraction module to extract local spatial information and global spectral information to obtain local spatial features and global spectral features;

[0024] Input the multispectral image into the second feature extraction module to extract local spectral information and global spatial information, and obtain central area features and adjacent area features;

[0025] The local spatial features, the global spectral features, the central area features and the adjacent area features are input into a prediction and classification module to classify the multispectral image, and output a class label of the multispectral image;

[0026] The current water quality data and water body data of the target water body are obtained according to the class label.

[0027] Preferably, the water body data is input into a classification model to extract the water surface line of the target water body, including:

[0028] Extracting features from the water body data to obtain water body grid features;

[0029] Extracting the water body grid features to obtain at least two local water body grid features;

[0030] Performing feature fusion on the at least two local water body grid features to obtain a global water body grid feature;

[0031] Based on the classification and regression processing of the global water body grid features, the key points of the water surface line are obtained;

[0032] The key points are fitted to obtain the water surface system of the target water body.

[0033] Preferably, the current water quality data and the water surface line are input into the water body plankton coverage inversion model to perform water body plankton coverage inversion, including:

[0034] Obtain historical water quality monitoring data of target water bodies for multiple historical time periods;

[0035] Generate an independent variable data set and a dependent variable data set according to the historical water quality detection data;

[0036] Extracting a first independent component from the independent variable data set, and extracting a first dependent component from the dependent variable data set;

[0037] Using a partial least squares regression algorithm, performing autoregression processing on the first autovariant component and the first dependent component, and determining the model accuracy according to the autoregression result;

[0038] If the accuracy of the model does not meet the preset accuracy condition, extracting a second independent variable component from the independent variable data set, and extracting a second dependent variable component from the dependent variable data set for iterative processing until the model meets the preset stop iteration condition, thereby obtaining the water body plankton coverage inversion model;

[0039] The current water quality data and the water surface line are input into the water body plankton coverage inversion model to perform water body plankton coverage inversion to obtain the water body plankton coverage inversion result of the target water body.

[0040] Preferably, the current water quality data and the water surface line are input into the water surface eutrophication inversion model to perform water surface eutrophication inversion, including:

[0041] Based on a variety of machine learning algorithms, multiple water surface eutrophication inversion models are constructed, and the entropy weight method is used to organically combine multiple water surface eutrophication inversion models to obtain the optimal water surface eutrophication inversion model;

[0042] The current water quality data and the water surface line are input into the optimal water surface eutrophication inversion model to perform water surface eutrophication inversion to obtain the current water surface eutrophication inversion result of the target water body.

[0043] Preferably, the water quality parameter analysis model is used to fuse the water body plankton coverage inversion result, the water body transparency inversion result and the water surface eutrophication inversion result to output the final water quality analysis result, including:

[0044] Respectively performing correlation analysis on the inversion result of water body plankton coverage, the water body transparency inversion result and the water surface eutrophication inversion result to obtain correlation analysis results;

[0045] Adjusting the parameters of the water quality analysis model based on the correlation analysis results;

[0046] The water body plankton coverage inversion result, water body transparency inversion result and water surface eutrophication inversion result are input into the adjusted water quality analysis model to obtain the final water quality analysis result.

[0047] On the other hand, the present invention discloses an online water quality analysis system based on a multispectral image analysis model, which is used to implement the above-mentioned online water quality analysis method based on a multispectral image analysis model, comprising:

[0048] An acquisition module, used for acquiring multispectral images of target water bodies;

[0049] A transfer station, used for preprocessing the multispectral image;

[0050] Analysis module, used for water quality analysis based on pre-processed multispectral images; including:

[0051] An analysis unit, used for extracting features from the preprocessed multispectral image using an analysis model, and obtaining current water quality data and water body data of the target water body;

[0052] A water surface line extraction unit, used for inputting the water body data into a classification model to extract the water surface line of the target water body;

[0053] An inversion unit is used to input the current water quality data and the water surface line into the water body plankton coverage inversion model, the water body transparency inversion model and the water surface eutrophication inversion model, respectively perform water body plankton coverage inversion, water body transparency inversion and water surface eutrophication inversion, and obtain water body plankton coverage inversion result, water body transparency inversion result and water surface eutrophication inversion result;

[0054] The water quality analysis unit is used to use the water quality parameter analysis model to fuse the water body plankton coverage inversion result, the water body transparency inversion result and the water surface eutrophication inversion result, and output the final water quality analysis result.

[0055] It can be seen from the above technical solutions that, compared with the prior art, the present invention discloses an online water quality analysis method and system based on a multispectral image analysis model. Through the multispectral image analysis model, high-precision monitoring of water quality is achieved, and detailed water quality data is provided. Automated multispectral image preprocessing and feature extraction simplify the data processing process and improve analysis efficiency. By integrating the inversion of water body plankton coverage, water body transparency inversion and water surface eutrophication inversion, the mutual influence between the inversion results of water body plankton coverage, water body transparency inversion and water surface eutrophication inversion is taken into account, making the water quality analysis more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0057] Figure 1 The present invention provides a schematic flow chart of the method.

[0058] Figure 2 This is a system framework diagram provided by the present invention. DETAILED DESCRIPTION

[0059] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0060] The embodiment of the present invention discloses an online water quality analysis method based on a multispectral image analysis model, such as Figure 1 As shown, including:

[0061] Collect multispectral images of the target water body and preprocess the multispectral images; use a multispectral camera equipped with a drone to take aerial photos of the target water body to obtain high-quality multispectral images. The camera can capture spectral information in different bands, providing a rich data source for water quality analysis. The collected images are transmitted to the transfer workstation in real time via a wireless network. At the transfer workstation, the received multispectral images are quickly preprocessed and spliced ​​and fitted. The preprocessing steps include geometric correction, radiation correction, cloud and shadow removal, noise filtering, and image enhancement to ensure the accuracy and availability of image data.

[0062] Construct a multispectral image analysis model, which includes: analysis model, classification model, water body plankton coverage inversion model, water body transparency inversion model, water surface eutrophication inversion model and water quality parameter analysis model;

[0063] The analytical model is used to extract features from the preprocessed multispectral image to obtain the current water quality data and water body data of the target water body;

[0064] Input water body data into the classification model to extract the water surface line of the target water body;

[0065] Input the current water quality data and water surface line into the water body plankton coverage inversion model, water body transparency inversion model and water surface eutrophication inversion model, respectively perform water body plankton coverage inversion, water body transparency inversion and water surface eutrophication inversion, and obtain the water body plankton coverage inversion result, water body transparency inversion result and water surface eutrophication inversion result;

[0066] The water quality parameter analysis model is used to integrate the inversion results of water body plankton coverage, water body transparency and water surface eutrophication, and output the final water quality analysis results. The water quality analysis results include water temperature, water depth, algae coverage, water body transparency, total phosphorus, total nitrogen, etc.

[0067] Furthermore, the preprocessing of the multispectral image includes: geometric correction and radiation correction, multispectral image fusion, cloud and shadow removal, noise filtering, and image enhancement.

[0068] Geometric correction is used to correct geometric distortions in images, such as tilt, distortion or misalignment, to ensure that the spatial information in the image is accurate. Radiometric correction focuses on adjusting the brightness, contrast and color balance of the image to eliminate radiation distortion caused by factors such as sensor characteristics, atmospheric conditions or sun angle, making the image closer to the real scene.

[0069] Multispectral image fusion includes: integrating image information from different spectral bands (such as visible light, infrared, near infrared, short-wave infrared, etc.) to generate a single image that contains richer information and is easy to interpret, ensuring the effective fusion of information from each band while avoiding information redundancy and distortion to improve the overall information content and readability of the image. Multiple images are stitched together to obtain the overall multispectral image of the target water body.

[0070] In multispectral images, clouds and shadows may obscure the target area, affecting the image quality and subsequent analysis. Advanced algorithms can identify and remove clouds and shadows, restore the information of the obscured area, or at least reduce its interference with the overall image analysis.

[0071] Due to sensor noise, environmental factors or errors in data transmission, various noises often exist in images. Noise filtering technology aims to remove these unnecessary random changes, retain useful information in the image, and improve the clarity and signal-to-noise ratio of the image.

[0072] In order to further improve the visualization and readability of the image, image enhancement processing is usually performed. This includes contrast enhancement, sharpening, color enhancement, etc., aiming to make the features in the image more prominent and the edges clearer, which is convenient for manual observation or automatic machine recognition.

[0073] Furthermore, multispectral image fusion includes:

[0074] Performing image registration on the first image and the second image respectively includes:

[0075] First, the first image and the second image are preprocessed (the first image and the second image are from different spectral bands), including removing noise, adjusting image size and resolution, etc., to ensure that they have similar image quality and spatial resolution when being registered.

[0076] A feature point detection algorithm (such as SIFT, SURF, etc.) is used to detect key feature points in the first image and the second image. These feature points are usually significant structures or texture changes in the images.

[0077] By comparing the descriptors of the feature points, the feature points in the first image are matched with the feature points in the second image, and a correspondence between the two images is established.

[0078] Based on the matched feature point pairs, the transformation matrix between the images (such as affine transformation, perspective transformation, etc.) is calculated, and the second image is transformed so that it is completely aligned with the first image in space.

[0079] The first image and the second image after image registration are subjected to feature extraction to obtain the first feature; the first image and the second image after image registration are subjected to multi-scale analysis to extract image features at different scales, such as edge, texture, shape, etc. The features from different spectral bands are fused to obtain a comprehensive feature set containing multi-spectral information, namely the first feature.

[0080] The first feature is classified to obtain the classification result; according to the classification result, a decision-level fusion strategy is designed, such as voting mechanism, weighted average, Bayesian fusion, etc., to comprehensively consider the information of different spectral bands. The fusion strategy is applied to fuse the classification results from different spectral bands to obtain the final fused multispectral image.

[0081] Decision-level fusion is performed based on the classification results to obtain a fused multispectral image.

[0082] Furthermore, the analytical model is used to extract features from the preprocessed multispectral image to obtain the current water quality data and water body data of the target water body, including:

[0083] The preprocessed multispectral image is input into the first feature extraction module to extract local spatial information and global spectral information to obtain local spatial features and global spectral features; the first feature extraction module is specifically used to extract local spatial information and global spectral information in the image. Local spatial information helps identify small-scale features in the image, while global spectral information provides an overview of the spectral features of the entire image. Through this module, local spatial features and global spectral features can be obtained.

[0084] Input the multispectral image into the second feature extraction module to extract local spectral information and global spatial information, and obtain central area features and adjacent area features;

[0085] The local spatial features, global spectral features, central region features and adjacent region features are input into the prediction and classification module to classify the multispectral image and output the class label of the multispectral image; the local spatial features, global spectral features, central region features and adjacent region features obtained from the two feature extraction modules are input into the prediction and classification module. This module uses these features to classify the multispectral image and outputs the class label of the multispectral image. The class label is a high-level description of the image content, which helps to identify different types of water features in the image.

[0086] The current water quality data and water body data of the target water body are obtained according to the class label, including the spectral characteristics, spatial characteristics of the water body and the water quality parameters derived from these characteristics, such as chlorophyll concentration, turbidity, etc.

[0087] In another embodiment, water body data is input into a classification model to extract a water surface line of a target water body, including:

[0088] Extract features from water body data to obtain water body grid features;

[0089] Extracting the water body grid features to obtain at least two local water body grid features;

[0090] Performing feature fusion on at least two local water body grid features to obtain a global water body grid feature;

[0091] Based on the classification and regression processing of the global water body grid features, the key points of the water surface line are obtained;

[0092] The key points are fitted to obtain the water surface system of the target water body.

[0093] In another embodiment, the current water quality data and the water surface line are input into the water body plankton coverage inversion model to perform water body plankton coverage inversion, including:

[0094] Obtain historical water quality monitoring data for the target water body for multiple historical time periods, including but not limited to key water quality parameters such as dissolved oxygen, ammonia nitrogen, transparency, plankton species and quantity, as well as data collection time, collection space location information, etc.

[0095] The independent variable data set and the dependent variable data set are generated based on the historical water quality detection data; the independent variable data set contains various factors affecting water quality, while the dependent variable data set contains the corresponding water quality parameter values.

[0096] Extract the first independent component from the independent variable data set, and extract the first dependent component from the dependent variable data set;

[0097] The partial least squares regression algorithm is used to perform autoregression on the first autovariant component and the first dependent component, and the model accuracy is determined based on the autoregression results;

[0098] If the model accuracy does not meet the preset accuracy conditions, the second independent variable component is extracted from the independent variable data set, and the second dependent variable component is extracted from the dependent variable data set for iterative processing until the model meets the preset stop iteration conditions, and the water body plankton coverage inversion model is obtained;

[0099] The current water quality data and water surface line are input into the water body plankton coverage inversion model to perform water body plankton coverage inversion and obtain the water body plankton coverage inversion result of the target water body.

[0100] In another embodiment, in actual application, the water transparency inversion process is similar to the above-mentioned water body plankton coverage inversion process.

[0101] In another embodiment, the current water quality data and the water surface line are input into the water surface eutrophication inversion model to perform water surface eutrophication inversion, including:

[0102] Based on a variety of machine learning algorithms (such as support vector machine, random forest, gradient boosting tree, neural network, etc.), multiple water surface eutrophication inversion models are constructed, and the entropy weight method is used to organically combine multiple water surface eutrophication inversion models to obtain the optimal water surface eutrophication inversion model; in order to obtain the optimal water surface eutrophication inversion model, the entropy weight method is used to evaluate the prediction performance of each model, and the weights of multiple water surface eutrophication inversion models are assigned accordingly. The entropy weight method reflects the stability and reliability of the model in the prediction by calculating the entropy value (i.e., the degree of uncertainty) of the prediction results of each model. The weight allocation follows the principle of "the smaller the entropy value, the greater the weight", that is, the more stable and reliable the prediction results of the model, the greater the contribution in the final combined model. According to the calculated weights, multiple water surface eutrophication inversion models are organically combined to obtain the optimal water surface eutrophication inversion model. This combined model can combine the advantages of each model and improve the accuracy and stability of the prediction.

[0103] The current water quality data and water surface line are input into the optimal water surface eutrophication inversion model to perform water surface eutrophication inversion and obtain the current water surface eutrophication inversion result of the target water body.

[0104] In another embodiment, the water quality parameter analysis model is used to fuse the inversion results of water body plankton coverage, water body transparency and water surface eutrophication to output the final water quality analysis results, including:

[0105] The inversion results of water plankton coverage, water transparency and water surface eutrophication are respectively conducted; specifically, the correlation analysis methods in statistics (such as Pearson correlation coefficient, Spearman rank correlation coefficient, etc.) are used to quantify the correlation between water plankton coverage, water transparency and water surface eutrophication status.

[0106] Adjust the parameters of the water quality analysis model based on the correlation analysis results;

[0107] The inversion results of water body plankton coverage, water body transparency and water surface eutrophication are input into the adjusted water quality analysis model to obtain the final water quality analysis results.

[0108] On the other hand, the present invention discloses an online water quality analysis system based on a multispectral image analysis model, which is used to implement the above-mentioned online water quality analysis method based on a multispectral image analysis model, such as Figure 2 As shown, including:

[0109] An acquisition module, used for acquiring multispectral images of target water bodies;

[0110] Transfer station, used for preprocessing multispectral images;

[0111] Analysis module, used for water quality analysis based on pre-processed multispectral images; including:

[0112] An analysis unit, used for extracting features from the preprocessed multispectral image using an analysis model, and obtaining current water quality data and water body data of the target water body;

[0113] A water surface line extraction unit is used to input water body data into a classification model to extract a water surface line of a target water body;

[0114] The inversion unit is used to input the current water quality data and the water surface line into the water body plankton coverage inversion model, the water body transparency inversion model and the water surface eutrophication inversion model, respectively perform the water body plankton coverage inversion, the water body transparency inversion and the water surface eutrophication inversion, and obtain the water body plankton coverage inversion result, the water body transparency inversion result and the water surface eutrophication inversion result;

[0115] The water quality analysis unit is used to use the water quality parameter analysis model to fuse the inversion results of water body plankton coverage, water body transparency and water surface eutrophication, and output the final water quality analysis results.

[0116] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0117] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An online water quality analysis method based on a multispectral image analysis model, characterized in that: include: Collecting a multispectral image of a target water body and preprocessing the multispectral image; Constructing a multispectral image analysis model, the multispectral image analysis model includes: an analysis model, a classification model, a water body plankton coverage inversion model, a water body transparency inversion model, a water surface eutrophication inversion model and a water quality parameter analysis model; Using the analytical model to extract features from the preprocessed multispectral image, obtaining current water quality data and water body data of the target water body; Inputting the water body data into a classification model to extract the water surface line of the target water body; Inputting the current water quality data and the water surface line into the water body plankton coverage inversion model, the water body transparency inversion model and the water surface eutrophication inversion model, respectively performing water body plankton coverage inversion, water body transparency inversion and water surface eutrophication inversion to obtain water body plankton coverage inversion result, water body transparency inversion result and water surface eutrophication inversion result; The water quality parameter analysis model is used to fuse the water body plankton coverage inversion result, the water body transparency inversion result and the water surface eutrophication inversion result to output the final water quality analysis result.

2. The online water quality analysis method based on the multispectral image analysis model according to claim 1 is characterized in that: The preprocessing of the multispectral image includes: geometric correction and radiation correction, multispectral image fusion, cloud removal and shadow processing, noise filtering, and image enhancement.

3. The online water quality analysis method based on the multispectral image analysis model according to claim 2 is characterized in that: Multispectral image fusion, including: Performing image registration on the first image and the second image respectively; Extracting features from the first image and the second image after image registration to obtain a first feature; Classify the first feature to obtain a classification result; Decision-level fusion is performed based on the classification results to obtain a fused multispectral image.

4. The online water quality analysis method based on a multispectral image analysis model according to claim 1 is characterized in that: The analytical model is used to extract features from the preprocessed multispectral image to obtain current water quality data and water body data of the target water body, including: The preprocessed multispectral image is input into a first feature extraction module to extract local spatial information and global spectral information to obtain local spatial features and global spectral features; Input the multispectral image into the second feature extraction module to extract local spectral information and global spatial information, and obtain central area features and adjacent area features; The local spatial features, the global spectral features, the central area features and the adjacent area features are input into a prediction and classification module to classify the multispectral image, and output a class label of the multispectral image; The current water quality data and water body data of the target water body are obtained according to the class label.

5. The online water quality analysis method based on a multispectral image analysis model according to claim 1 is characterized in that: Inputting the water body data into a classification model to extract the water surface line of the target water body includes: Extracting features from the water body data to obtain water body grid features; Extracting the water body grid features to obtain at least two local water body grid features; Performing feature fusion on the at least two local water body grid features to obtain a global water body grid feature; Based on the classification and regression processing of the global water body grid features, the key points of the water surface line are obtained; The key points are fitted to obtain the water surface system of the target water body.

6. The online water quality analysis method based on a multispectral image analysis model according to claim 1 is characterized in that: Inputting the current water quality data and the water surface line into the water body plankton coverage inversion model to perform water body plankton coverage inversion, including: Obtain historical water quality monitoring data of target water bodies for multiple historical time periods; Generate an independent variable data set and a dependent variable data set according to the historical water quality detection data; extracting a first independent component from the independent variable data set, and extracting a first dependent component from the dependent variable data set; Using a partial least squares regression algorithm, performing autoregression processing on the first autovariant component and the first dependent component, and determining the model accuracy according to the autoregression result; If the accuracy of the model does not meet the preset accuracy condition, extracting a second independent variable component from the independent variable data set, and extracting a second dependent variable component from the dependent variable data set for iterative processing until the model meets the preset stop iteration condition, thereby obtaining the water body plankton coverage inversion model; The current water quality data and the water surface line are input into the water body plankton coverage inversion model to perform water body plankton coverage inversion to obtain the water body plankton coverage inversion result of the target water body.

7. The online water quality analysis method based on a multispectral image analysis model according to claim 2 is characterized in that: Inputting the current water quality data and the water surface line into the water surface eutrophication inversion model to perform water surface eutrophication inversion, including: Based on a variety of machine learning algorithms, multiple water surface eutrophication inversion models are constructed, and the entropy weight method is used to organically combine multiple water surface eutrophication inversion models to obtain the optimal water surface eutrophication inversion model; The current water quality data and the water surface line are input into the optimal water surface eutrophication inversion model to perform water surface eutrophication inversion to obtain the current water surface eutrophication inversion result of the target water body.

8. The online water quality analysis method based on a multispectral image analysis model according to claim 1 is characterized in that: The water quality parameter analysis model is used to fuse the water body plankton coverage inversion result, the water body transparency inversion result and the water surface eutrophication inversion result to output the final water quality analysis result, including: Respectively performing correlation analysis on the inversion result of water body plankton coverage, the water body transparency inversion result and the water surface eutrophication inversion result to obtain correlation analysis results; Adjusting the parameters of the water quality analysis model based on the correlation analysis results; The water body plankton coverage inversion result, water body transparency inversion result and water surface eutrophication inversion result are input into the adjusted water quality analysis model to obtain the final water quality analysis result.

9. An online water quality analysis system based on a multispectral image analysis model, characterized in that: The method for implementing an online water quality analysis method based on a multispectral image analysis model as claimed in any one of claims 1 to 8 comprises: An acquisition module, used for acquiring multispectral images of target water bodies; A transfer station, used for preprocessing the multispectral image; Analysis module, used for water quality analysis based on pre-processed multispectral images; including: An analysis unit, used for extracting features from the preprocessed multispectral image using an analysis model, and obtaining current water quality data and water body data of the target water body; A water surface line extraction unit, used for inputting the water body data into a classification model to extract the water surface line of the target water body; An inversion unit is used to input the current water quality data and the water surface line into the water body plankton coverage inversion model, the water body transparency inversion model and the water surface eutrophication inversion model, respectively perform water body plankton coverage inversion, water body transparency inversion and water surface eutrophication inversion, and obtain water body plankton coverage inversion result, water body transparency inversion result and water surface eutrophication inversion result; The water quality analysis unit is used to use the water quality parameter analysis model to fuse the water body plankton coverage inversion result, the water body transparency inversion result and the water surface eutrophication inversion result, and output the final water quality analysis result.

Images