Seawater quality monitoring method and system based on automatic multi-source data assimilation

Abstract

The present invention relates to the technical field of seawater quality detection, and discloses a seawater quality monitoring method based on automatic multi-source data assimilation, including: collecting satellite remote sensing images of a seawater monitoring area, dividing the seawater monitoring area into eutrophic levels, and obtaining eutrophication-divided ocean areas; collecting optical sensing data of the seawater monitoring area, determining the concentration of suspended particles in the seawater monitoring area; collecting temperature sensing data of the seawater monitoring area, analyzing the level of water quality change in the seawater monitoring area; collecting salinity data and dissolved oxygen data of the seawater monitoring area, analyzing the degree of change in the ecological structure of the seawater monitoring area; constructing a comprehensive ocean multi-attribute map based on eutrophication-divided ocean areas, suspended particle concentration, water quality change level, and ecological structure change degree, and issuing an early warning to the corresponding monitoring users in the seawater monitoring area. The present invention can improve the comprehensiveness of seawater quality detection.

Description

Technical Field

[0001] The present invention relates to the technical field of seawater quality detection, and in particular to a seawater quality monitoring method and system based on automatic multi-source data assimilation. Background Art

[0002] Seawater quality is of vital importance to the marine ecological environment, fishery development, and economic and social stability of coastal areas. Building an automatic multi-source data assimilation platform for seawater quality monitoring can achieve comprehensive, real-time monitoring and analysis of seawater quality parameters, thereby improving the accuracy and timeliness of seawater quality monitoring, and providing a scientific basis for marine environmental protection and resource management. In addition, the platform can also facilitate information sharing and communication between different monitoring agencies, between scientific researchers and managers, and between the public and relevant departments, reduce the risk of decision-making errors due to poor information flow, and improve the collaborative efficiency of marine management.

[0003] At present, traditional fixed-point sampling and laboratory analysis methods, or monitoring technologies based on a single data source, are usually used for seawater quality monitoring. However, with this method, the coverage of monitoring data is limited and it is difficult to reflect the dynamic changes in seawater quality. Due to the complex and changeable marine environment, there are a large number of different types of data sources. Traditional methods are not accurate enough in data integration and analysis when the data volume is large and multi-source, resulting in incomplete seawater quality monitoring. Summary of the invention

[0004] The present invention provides a seawater quality monitoring method and system based on automatic multi-source data assimilation, the main purpose of which is to improve the comprehensiveness of seawater quality detection.

[0005] To achieve the above object, the present invention provides a seawater quality monitoring method based on automatic multi-source data assimilation, comprising:

[0006] Collect satellite remote sensing images of the seawater monitoring area, calibrate the satellite remote sensing images to obtain calibrated images, identify image information of the calibrated images, and divide the seawater monitoring area into eutrophic levels based on the image information to obtain eutrophic divided marine areas;

[0007] Collecting optical sensor data of the seawater monitoring area, analyzing absorption characteristics of particles in the seawater monitoring area using the optical sensor data, and analyzing the concentration of suspended particles in the seawater monitoring area based on the absorption characteristics;

[0008] Collecting temperature sensor data of the seawater monitoring area, analyzing the degree of change of seawater density and the degree of change of seawater transparency in the seawater monitoring area based on the temperature sensor data, and analyzing the level of change of water quality in the seawater monitoring area based on the degree of change of seawater density and the degree of change of seawater transparency;

[0009] Collect salinity data and dissolved oxygen data of the seawater monitoring area, and construct an ecological structure model of the seawater monitoring area based on the temperature sensor data, the salinity data and the dissolved oxygen data, so as to analyze the degree of change of the ecological structure of the seawater monitoring area by using the ecological structure model;

[0010] Based on the eutrophication-divided ocean areas, the suspended particulate matter concentration, the water quality change level and the degree of ecological structure change, a comprehensive ocean multi-attribute map of the seawater monitoring area is constructed, and multi-source seawater data of the seawater monitoring area is collected in real time, so that the multi-source seawater data is used to update the comprehensive ocean multi-attribute map to obtain a real-time updated attribute map, and the real-time updated attribute map is used to analyze the water quality level of the seawater monitoring area, and when the water quality level does not meet the preset value, an early warning is issued to the corresponding monitoring users of the seawater monitoring area.

[0011] Optionally, the performing image calibration on the satellite remote sensing image to obtain a calibrated image includes:

[0012] Extracting target points in the satellite remote sensing image;

[0013] The echo energy of the target point is calculated using the following formula:

[0014] ;

[0015] in, represents the echo energy, m represents the total number of target points, Represents the pixel intensity value of the target point area image, Represents the number of pixels in the target point area image, Indicates the number of pixels of the echo area image corresponding to the target point, Indicates the total number of echo points corresponding to the target point, Indicates the pixel size of the target point area image, Indicates the pixel size of the echo area image corresponding to the target point;

[0016] Querying the relative radiometric calibration accuracy of the satellite remote sensing image;

[0017] According to the echo energy and the relative radiation calibration accuracy, the radiation calibration constant of the satellite remote sensing image is calculated using the following formula:

[0018] ;

[0019] in, represents the calibration constant, p represents the echo energy, Represents the radiation area of ​​satellite remote sensing images, Represents the radiation angle of satellite remote sensing image.

[0020] The satellite remote sensing image is radiometrically calibrated according to the radiometric calibration constant to obtain a calibrated image.

[0021] Optionally, the identifying the image information of the calibration image includes:

[0022] Performing image decomposition on the calibration image to obtain a decomposed image;

[0023] extracting wavelet coefficients of the decomposed image;

[0024] According to the wavelet coefficients, the decomposed images are fused to obtain a fused image;

[0025] Detecting image corner points of the fused image;

[0026] Performing multi-scale characterization on the image corner points to obtain multi-scale corner points;

[0027] The multi-scale corner points are used to identify image information of the calibration image.

[0028] Optionally, performing image fusion on the decomposed images according to the wavelet coefficients to obtain a fused image includes:

[0029] The mean square error of the wavelet coefficients is calculated using the following formula:

[0030] ;

[0031] in, represents the mean square error, M represents the number of decomposed images, represents the wavelet coefficient of the u-th decomposed image, represents the wavelet coefficients of the vth decomposition image,

[0032] Based on the mean square error, decomposed images in the same frequency domain in the decomposed images are fused to obtain a fused image.

[0033] Optionally, dividing the seawater monitoring area into eutrophic levels based on the image information to obtain eutrophic marine areas includes:

[0034] Querying the band information in the image information;

[0035] Calculating the chlorophyll index of the seawater monitoring area based on the band information;

[0036] The seawater monitoring area is divided into eutrophic levels based on the chlorophyll index to obtain eutrophic marine areas.

[0037] Optionally, the analyzing the absorption characteristics of particulate matter in the seawater monitoring area using the optical sensing data includes:

[0038] Performing data preprocessing on the optical sensing data to obtain preprocessed data;

[0039] Performing data calibration on the preprocessed data to obtain calibration data;

[0040] Calculating the absorption coefficient of the particulate matter based on the calibration data;

[0041] Based on the absorption coefficient, the absorption characteristics of the particulate matter are analyzed.

[0042] Optionally, analyzing the concentration of suspended particles in the seawater monitoring area based on the absorption characteristics includes:

[0043] Based on the absorption characteristics, constructing optical characteristic parameters of seawater in the seawater monitoring area;

[0044] Querying historical water optical characteristic parameters of the seawater monitoring area;

[0045] Based on the historical water optical characteristic parameters, construct a historical turbidity curve of the seawater monitoring area;

[0046] Based on the historical turbidity curve and the seawater optical characteristic parameters, the concentration of suspended particulate matter in the seawater monitoring area is analyzed.

[0047] Optionally, analyzing the water quality change level of the seawater monitoring area based on the change degree of the seawater density and the change degree of the seawater transparency includes:

[0048] Query important parameters affecting water quality in the seawater monitoring area;

[0049] Based on the important parameters affecting water quality, the degree of change in seawater density and the degree of change in seawater transparency, a water quality evaluation factor set for the seawater monitoring area is constructed;

[0050] Constructing an evaluation grade of water quality change level in the seawater monitoring area;

[0051] Constructing a membership function of each evaluation indicator in the evaluation factor set, and assigning an indicator weight to the evaluation indicator;

[0052] The water quality change level of the seawater monitoring area is analyzed based on the water quality evaluation factor set, the evaluation level, the membership function and the indicator weight.

[0053] Optionally, constructing the ecological structure model of the seawater monitoring area based on the temperature sensor data, the salinity data and the dissolved oxygen data includes:

[0054] Performing data integration on the temperature sensing data, the salinity data and the dissolved oxygen data to obtain integrated data;

[0055] Inputting the integrated data into a preconfigured machine learning model to perform model training on the machine learning model to obtain a training model;

[0056] Constructing a test sample of the training model;

[0057] Based on the test samples, the mean absolute error of the training model is calculated using the following formula:

[0058] ;

[0059] in, represents the mean absolute error, N represents the number of test samples, represents the true value of the i-th data of the test sample, Represents the predicted value of the i-th data of the test sample;

[0060] Performing a performance evaluation on the training model according to the mean absolute error to obtain a performance evaluation result;

[0061] When the performance evaluation result meets the preset result, an ecological structure model is obtained.

[0062] In order to solve the above problems, the present invention also provides a seawater quality monitoring system based on automatic multi-source data assimilation, the system comprising:

[0063] A eutrophication area division module is used to collect satellite remote sensing images of the seawater monitoring area, calibrate the satellite remote sensing images to obtain calibrated images, identify image information of the calibrated images, and divide the seawater monitoring area into eutrophic levels based on the image information to obtain eutrophication divided ocean areas;

[0064] A suspended particle concentration analysis module, used to collect optical sensor data of the seawater monitoring area, analyze the absorption characteristics of particles in the seawater monitoring area using the optical sensor data, and analyze the suspended particle concentration in the seawater monitoring area based on the absorption characteristics;

[0065] A water quality change analysis module, used to collect temperature sensor data of the seawater monitoring area, analyze the degree of change of seawater density and the degree of change of seawater transparency in the seawater monitoring area based on the temperature sensor data, and analyze the level of water quality change in the seawater monitoring area based on the degree of change of seawater density and the degree of change of seawater transparency;

[0066] An ecological structure model building module is used to collect salinity data and dissolved oxygen data of the seawater monitoring area, and build an ecological structure model of the seawater monitoring area based on the temperature sensor data, the salinity data and the dissolved oxygen data, so as to use the ecological structure model to analyze the degree of change of the ecological structure of the seawater monitoring area;

[0067] The water quality early warning module is used to construct a comprehensive ocean multi-attribute map of the seawater monitoring area based on the eutrophication division of the ocean area, the concentration of suspended particulate matter, the level of water quality change and the degree of ecological structure change, collect seawater multi-source data of the seawater monitoring area in real time, and use the seawater multi-source data to update the ocean multi-attribute comprehensive map to obtain a real-time updated attribute map, use the real-time updated attribute map to analyze the water quality level of the seawater monitoring area, and when the water quality level does not meet the preset value, issue an early warning to the corresponding monitoring user of the seawater monitoring area.

[0068] The present invention can observe the seawater monitoring area from a high altitude and over a large area by collecting satellite remote sensing images of the seawater monitoring area, and provide macroscopic ocean surface information, which is important for understanding the status of the entire monitoring area, discovering large-scale abnormal phenomena, and conducting long-term dynamic monitoring; the optical sensor data collected in the seawater monitoring area can accurately determine the concentration of suspended particles by analyzing the optical sensor data, thereby understanding the turbidity and water quality of the seawater; the present invention can understand the impact of factors such as ocean circulation, climate change and human activities on ocean water quality by collecting temperature sensor data in the seawater monitoring area; the present invention can collect salinity data in the seawater monitoring area and dissolved oxygen data can help users analyze the composition and changes of biological communities in marine ecosystems to better understand the seawater quality; finally, the present invention divides the marine areas based on the eutrophication, the suspended particulate matter concentration, the water quality change level and the degree of change in the ecological structure, and constructs a comprehensive marine multi-attribute map of the seawater monitoring area, which can integrate multiple important marine attribute parameters, comprehensively reflect the overall state of the seawater monitoring area from different angles, and can help users quickly understand the marine conditions in different areas to formulate targeted management measures, such as taking measures to reduce the input of nutrients in eutrophic areas, and strengthening pollution control in areas with poor water quality. Therefore, the present invention can improve the comprehensiveness of seawater quality detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 A schematic flow chart of a method for monitoring seawater quality based on automatic multi-source data assimilation provided by one embodiment of the present invention;

[0070] Figure 2 A functional module diagram of a seawater quality monitoring system based on automatic multi-source data assimilation provided by one embodiment of the present invention;

[0071] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0072] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0073] The embodiment of the present application provides a method for monitoring seawater quality based on automatic multi-source data assimilation. The execution subject of the method for monitoring seawater quality based on automatic multi-source data assimilation includes but is not limited to at least one of the electronic devices such as a server, a terminal, etc. that can be configured to execute the method provided by the embodiment of the present application. In other words, the method for monitoring seawater quality based on automatic multi-source data assimilation can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc. The server can be an independent server, or it can be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and big data and artificial intelligence platforms.

[0074] Reference Figure 1 FIG. 1 is a flow chart of a method for monitoring seawater quality based on automatic multi-source data assimilation provided by an embodiment of the present invention. In this embodiment, the method for monitoring seawater quality based on automatic multi-source data assimilation includes:

[0075] S1. Collect satellite remote sensing images of the seawater monitoring area, calibrate the satellite remote sensing images to obtain calibrated images, identify image information of the calibrated images, and divide the seawater monitoring area into eutrophic levels based on the image information to obtain eutrophication-divided marine areas.

[0076] The embodiment of the present invention can observe the seawater monitoring area from a high altitude over a large area by collecting satellite remote sensing images of the seawater monitoring area, and provide macroscopic ocean surface information, which is important for understanding the status of the entire monitoring area, discovering large-scale abnormal phenomena, and conducting long-term dynamic monitoring.

[0077] The satellite remote sensing image refers to an image obtained by photographing the earth's surface using various sensors (such as optical sensors, infrared sensors, etc.) carried on satellites.

[0078] Furthermore, the embodiment of the present invention can convert the original image data acquired by the satellite sensor into a numerical value with actual physical meaning by performing image calibration on the satellite remote sensing image to obtain a calibrated image, so that subsequent analysis and interpretation of the image are more accurate and reliable.

[0079] As an embodiment of the present invention, the step of performing image calibration on the satellite remote sensing image to obtain a calibrated image includes: extracting a target point in the satellite remote sensing image, and calculating the echo energy of the target point using the following formula:

[0080] ;

[0081] in, represents the echo energy, m represents the total number of target points, Represents the pixel intensity value of the target point area image, Represents the number of pixels in the target point area image, Indicates the number of pixels of the echo area image corresponding to the target point, Indicates the total number of echo points corresponding to the target point, Indicates the pixel size of the target point area image, Indicates the pixel size of the echo area image corresponding to the target point;

[0082] The relative radiometric calibration accuracy of the satellite remote sensing image is queried, and the radiometric calibration constant of the satellite remote sensing image is calculated using the following formula according to the echo energy and the relative radiometric calibration accuracy:

[0083] ;

[0084] in, represents the calibration constant, p represents the echo energy, Represents the radiation area of ​​satellite remote sensing images, Represents the radiation angle of satellite remote sensing image.

[0085] The satellite remote sensing image is radiometrically calibrated according to the radiometric calibration constant to obtain a calibrated image.

[0086] Among them, the target point refers to the image observation area corresponding to the seawater monitoring area in the satellite remote sensing image, the echo energy refers to the energy reflected by the seawater to the probe when the probe sends a signal to the seawater, and the radiation calibration constant refers to the constant of the relationship between the seawater reflection signal area and the probe signal strength.

[0087] Optionally, the target point in the satellite remote sensing image is obtained by querying a preset observation area (the point to be observed in the image, such as setting the area with higher color depth as the observation point and setting the preset value of color depth to 0.3 , the color depth value corresponding to the color depth is greater than 0.3 The relative radiation calibration accuracy of the satellite remote sensing image is obtained by querying the historical radiation calibration accuracy of the water body to be detected.

[0088] The embodiment of the present invention can extract features related to seawater eutrophication, such as chlorophyll content, by identifying the image information of the calibration image, thereby providing data support for eutrophication level division.

[0089] As an embodiment of the present invention, the identifying image information of the calibration image includes: performing image decomposition on the calibration image to obtain a decomposed image, extracting wavelet coefficients of the decomposed image, performing image fusion on the decomposed image according to the wavelet coefficients to obtain a fused image, detecting image corner points of the fused image, performing multi-scale characterization on the image corner points to obtain multi-scale corner points, and using the multi-scale corner points to identify the image information of the calibration image.

[0090] The wavelet coefficients refer to a set of coefficients obtained by wavelet transform, which are used to represent the components of the signal at different frequencies and time scales. The image corner point refers to a pixel point in the image or a small area around a pixel point, which shows obvious edge changes in grayscale or color images.

[0091] Optionally, the decomposed image can be obtained by decomposing the calibration image through a wavelet decomposition function, and the image corner points can be detected by Harris corner detection technology. The process of detecting the image corner points of the fused image and performing multi-scale characterization on the image corner points is: using a gradient function to calculate the gradient value of the fused image, and calculating the corner point energy of the fused image, using the gradient value and the corner point energy as the corner point information of the fused image, determining the multi-scale corner points of the corner point image according to the corner point information, and using the multi-scale corner points to identify the image information of the calibration image can use a preconfigured deep learning model to use the multi-scale corner points to identify the image features of the calibration image, and identify the image information of the calibration image based on the image features.

[0092] Further, as an optional embodiment of the present invention, performing image fusion on the decomposed images according to the wavelet coefficients to obtain a fused image includes: calculating the mean square error of the wavelet coefficients using the following formula:

[0093] ;

[0094] in, represents the mean square error, M represents the number of decomposed images, represents the wavelet coefficient of the u-th decomposed image, represents the wavelet coefficients of the vth decomposition image,

[0095] Based on the mean square error, decomposed images in the same frequency domain in the decomposed images are fused to obtain a fused image.

[0096] Optionally, based on the mean square error, performing image fusion on the decomposed images in the same frequency domain in the decomposed images to obtain the fused image means fusing the decomposed images in the same frequency domain according to a preset mean square error range such as 0.5-0.6.

[0097] Furthermore, the embodiment of the present invention divides the seawater monitoring area into eutrophic levels based on the image information, and obtains eutrophication-divided ocean areas, which can help users to carry out targeted management and governance of eutrophication areas of different degrees, thereby improving the efficiency and effectiveness of marine environmental protection.

[0098] As an embodiment of the present invention, the eutrophication level division of the seawater monitoring area based on the image information to obtain the eutrophication-divided ocean area includes: querying the band information in the image information, calculating the chlorophyll index of the seawater monitoring area based on the band information, and eutrophication level division of the seawater monitoring area based on the chlorophyll index to obtain the eutrophication-divided ocean area. Wherein, the band information refers to electromagnetic wave information of different wavelength ranges.

[0099] Optionally, the band information can be obtained by querying the band list and corresponding wavelength range information in the file list corresponding to the satellite remote sensing image. The chlorophyll index of the seawater monitoring area calculated based on the band information can be calculated by the marine normalized vegetation index method. The eutrophication level of the seawater monitoring area is divided based on the chlorophyll index to obtain the eutrophication division of the marine area. For example, in some sea areas, when the chlorophyll a concentration (chlorophyll index) is greater than 10 micrograms / liter, it can be regarded as the beginning of eutrophication. However, the specific standards will vary from sea area to sea area. For example, in some nearshore sea areas, due to the greater impact of human activities, it may be necessary to pay attention to eutrophication when the chlorophyll a concentration reaches 5 micrograms / liter; in some remote sea areas that are less affected by human activities, eutrophication is considered to occur only when the chlorophyll a concentration reaches 15 micrograms / liter.

[0100] S2. Collect optical sensor data of the seawater monitoring area, analyze absorption characteristics of particulate matter in the seawater monitoring area using the optical sensor data, and analyze the concentration of suspended particulate matter in the seawater monitoring area based on the absorption characteristics.

[0101] The embodiment of the present invention can collect the optical sensor data of the seawater monitoring area and analyze the optical sensor data to accurately determine the concentration of suspended particles, thereby understanding the turbidity and water quality of the seawater.

[0102] The optical sensing data refers to data collected by optical sensors in the seawater monitoring area, such as light scattering, absorption and reflection data.

[0103] Furthermore, the embodiment of the present invention can analyze the degree of light scattering by suspended particles in seawater by analyzing the absorption characteristics of particles in the seawater monitoring area using the optical sensing data, thereby determining the turbidity of seawater. The absorption characteristics refer to the absorption capacity and characteristics of suspended particles in seawater to light of different wavelengths.

[0104] Optionally, the use of the optical sensing data to analyze the absorption characteristics of the particulate matter in the seawater monitoring area can be performed by querying the multi-angle scattering measurement data about the seawater monitoring area in the optical sensing data to identify the multi-angle scattered light distribution of the suspended particulate matter.

[0105] As an embodiment of the present invention, the use of the optical sensing data to analyze the absorption characteristics of particulate matter in the seawater monitoring area includes: performing data preprocessing on the optical sensing data to obtain preprocessed data, performing data calibration on the preprocessed data to obtain calibration data, calculating the absorption coefficient of the particulate matter based on the calibration data, and analyzing the absorption characteristics of the particulate matter based on the absorption coefficient.

[0106] Optionally, the preprocessed data can be obtained by digitally filtering the optical sensor data to remove data noise, the calibration data can be achieved by querying the standard material corresponding to the data calibration or the sample of known concentration for comparative measurement, the absorption coefficient of the particulate matter can be calculated based on the calibration data by querying the quantitative filter membrane experiment (filtering the seawater sample, intercepting the particulate matter on the filter membrane, and then using a spectrophotometer or other instrument to measure the absorbance of the filter membrane) data in the calibration data to calculate the absorption coefficient of the particulate matter, and the absorption characteristics of the particulate matter can be analyzed based on the absorption coefficient. The absorption spectrum curve of the particulate matter can be identified by the absorption coefficient, and the spatial and temporal variation of the absorption characteristics of the particulate matter can be analyzed based on the absorption spectrum curve to determine the absorption characteristics of the particulate matter in the seawater monitoring area, such as chlorophyll and other pigments in phytoplankton have strong absorption of blue and red light, but weak absorption of green light. Different types of phytoplankton contain different types and concentrations of pigments, so their absorption characteristics will also be different. Mineral particles may absorb infrared light of a specific wavelength, while the absorption characteristics of organic debris depend on their chemical composition and structure.

[0107] The embodiment of the present invention can understand the turbidity and pollution status of seawater by analyzing the concentration of suspended particles in the seawater monitoring area based on the absorption characteristics. For example, if the concentration of suspended particles is too high, it may indicate that there is a large amount of sediment, organic matter or pollutants in the seawater, which will affect the transparency and optical properties of the seawater and have a negative impact on the marine ecosystem.

[0108] The suspended particle concentration refers to the mass or quantity of suspended particles in a unit volume of seawater. Usually, the mass concentration is expressed in units such as milligrams per liter (mg / L) and micrograms per liter (μg / L), and the number concentration is expressed in units such as pieces per liter (pieces / L).

[0109] As an embodiment of the present invention, the analysis of the suspended particle concentration in the seawater monitoring area based on the absorption characteristics includes: constructing seawater optical characteristic parameters of the seawater monitoring area based on the absorption characteristics, querying historical water optical characteristic parameters of the seawater monitoring area, constructing a historical turbidity curve of the seawater monitoring area based on the historical water optical characteristic parameters, and analyzing the suspended particle concentration in the seawater monitoring area based on the historical turbidity curve and the seawater optical characteristic parameters.

[0110] Optionally, the seawater optical characteristic parameters can be used to calculate characteristic parameters of the absorption characteristics of a series of particulate matter through the absorption characteristics, such as characteristic parameters such as absorption value, absorption peak position and width at a specific wavelength, and the characteristic parameters can be digitally encoded. The historical water optical characteristic parameters can be obtained by querying the corresponding characteristic parameter data in the historical database of the seawater monitoring area, and the historical turbidity curve can be constructed through an Excel spreadsheet.

[0111] The suspended particle concentration in the seawater monitoring area can be analyzed based on the historical turbidity curve and the seawater optical characteristic parameters by utilizing a preconfigured neural network model to take the historical water optical characteristic parameters as input and the historical turbidity curve as output to analyze the historical concentration relationship, and the seawater optical characteristic parameters can be inverted and analyzed based on the historical concentration relationship.

[0112] S3. Collect temperature sensor data of the seawater monitoring area, analyze the degree of change of seawater density and the degree of change of seawater transparency in the seawater monitoring area based on the temperature sensor data, and analyze the level of water quality change in the seawater monitoring area based on the degree of change of seawater density and the degree of change of seawater transparency.

[0113] The embodiment of the present invention can understand the influence of factors such as ocean circulation, climate change and human activities on ocean water quality by collecting the temperature sensing data of the seawater monitoring area. The temperature sensing data can be monitored and acquired using temperature sensors, such as thermistors, thermocouples or infrared sensors, etc., deployed in the seawater monitoring area. Among them, the temperature sensing data refers to the seawater temperature data and distribution characteristics. In the ocean, the seawater temperature usually shows obvious vertical stratification, which can generally be divided into the surface layer, the middle layer and the deep layer. The surface seawater temperature is greatly affected by solar radiation, atmospheric temperature and sea-air interaction. The temperature is relatively high and changes more dramatically. The surface seawater temperature has obvious seasonal changes. In summer, the solar radiation is strong and the surface seawater temperature rises; in winter, on the contrary, the surface seawater temperature decreases. This seasonal change is manifested to different degrees in sea areas at different latitudes. The temperature change in low-latitude sea areas is relatively small, and the temperature change in high-latitude sea areas is large.

[0114] Furthermore, the embodiment of the present invention can understand the health status of the marine ecosystem and the trend of water quality changes by analyzing the degree of change in seawater density and seawater transparency in the seawater monitoring area based on the temperature sensor data.

[0115] The degree of change in seawater density refers to the magnitude of change in seawater density within a certain period of time or under specific conditions, and the degree of change in seawater transparency refers to the magnitude of change in seawater transparency within a certain period of time or under specific conditions.

[0116] Optionally, the analysis of the degree of change in seawater density and seawater transparency in the seawater monitoring area based on the temperature sensor data can be obtained by querying the continuous monitoring data on seawater density and seawater transparency monitoring data in the seawater monitoring area in the temperature sensor data.

[0117] Furthermore, the embodiments of the present invention can more comprehensively evaluate the level of water quality changes in the seawater monitoring area by analyzing the level of water quality changes in the seawater monitoring area based on the degree of change in the seawater density and the degree of change in the seawater transparency, thereby providing a scientific basis for marine environmental protection, resource management and ecological restoration.

[0118] As an embodiment of the present invention, the analysis of the water quality change level of the seawater monitoring area based on the degree of change of the seawater density and the degree of change of the seawater transparency includes: querying the important parameters affecting the water quality of the seawater monitoring area, constructing a water quality evaluation factor set for the seawater monitoring area based on the important parameters affecting water quality, the degree of change of the seawater density and the degree of change of the seawater transparency, constructing an evaluation grade for the water quality change level of the seawater monitoring area, constructing a membership function for each evaluation indicator in the evaluation factor set, and assigning indicator weights of the evaluation indicators, and analyzing the water quality change level of the seawater monitoring area based on the water quality evaluation factor set, the evaluation grade, the membership function and the indicator weight.

[0119] The important parameters affecting water quality refer to parameters that have a greater impact on seawater quality, such as dissolved oxygen content, nutrient concentration, pH value, etc., which can be obtained by querying seawater monitoring sample data in the seawater monitoring area.

[0120] Optionally, the water quality assessment factor set can be constructed in the following manner: Changes in seawater density, seawater transparency, dissolved oxygen content, nutrient concentration, value , It represents a water quality evaluation factor set. It should be noted that the water quality evaluation factor set needs to be constructed in combination with the data required for actual analysis. The evaluation level can be expressed as: Excellent, Good, Fair, Poor , Represents the evaluation level. The membership function can be determined according to the actual situation and experience to determine the membership function corresponding to different evaluation levels. For example, for the degree of change of seawater density, it is assumed that its value range is The membership function can be defined as follows: when the density changes in In the interval, the membership of "excellent" gradually increases from 0 to 1; interval, the membership of "good" gradually increases from 0 to 1, and the membership of "excellent" gradually decreases from 1 to 0; and so on, similarly, for indicators such as the degree of change in seawater transparency, dissolved oxygen content, nutrient concentration and pH value, the corresponding membership function is determined, and the indicator weight can be allocated by the hierarchical analysis method. The water quality change level of the seawater monitoring area based on the water quality evaluation factor set, the evaluation level, the membership function and the indicator weight analysis can be analyzed by the fuzzy comprehensive evaluation method. For example, for the seawater density change degree indicator, according to its actual value and the membership function, its evaluation level set is obtained. The membership vector ,in Indicates the index's membership to "excellent", Indicates the membership to "good", and so on. Similarly, we get the membership vectors of other indicators .

[0121] Then the membership vector of each indicator is multiplied by the corresponding weight and then synthesized to obtain the fuzzy comprehensive evaluation matrix , the final fuzzy comprehensive evaluation result ,in They respectively represent the degree of membership of water quality to the four evaluation levels of "excellent", "good", "medium" and "poor". The evaluation results of the water quality change level are determined based on the principle of maximum membership.

[0122] S4. Collect salinity data and dissolved oxygen data of the seawater monitoring area, and construct an ecological structure model of the seawater monitoring area based on the temperature sensor data, the salinity data and the dissolved oxygen data, so as to analyze the degree of change of the ecological structure of the seawater monitoring area by using the ecological structure model.

[0123] The embodiment of the present invention can help users analyze the composition and changes of biological communities in marine ecosystems by collecting salinity data and dissolved oxygen data in the seawater monitoring area, so as to better understand the seawater quality.

[0124] Among them, the salinity data refers to the total amount of dissolved solid matter contained in each kilogram of seawater in a specific seawater monitoring area, usually expressed in grams per kilogram; the dissolved oxygen data refers to the amount of oxygen dissolved in a unit volume of seawater in a specific seawater monitoring area, usually expressed in milligrams per liter.

[0125] The salinity data may be acquired through a salinity sensor, and the dissolved oxygen data may be acquired through a dissolved oxygen sensor.

[0126] Furthermore, the embodiment of the present invention can help users understand the impact mechanism of different environmental factors on marine biological communities, predict the changing trend of the ecosystem, and provide a scientific basis for marine ecological protection and management by constructing the ecological structure model of the seawater monitoring area based on the temperature sensor data, the salinity data, and the dissolved oxygen data. Among them, the ecological structure model refers to the use of mathematical methods and related theories to integrate key environmental data such as temperature, salinity, dissolved oxygen, etc. in the seawater monitoring area to establish a model that can describe the composition, distribution, and mutual relationship of the biological community in the marine ecosystem of the area.

[0127] As an embodiment of the present invention, the ecological structure model of the seawater monitoring area is constructed based on the temperature sensor data, the salinity data and the dissolved oxygen data, including: integrating the temperature sensor data, the salinity data and the dissolved oxygen data to obtain integrated data, inputting the integrated data into a preconfigured machine learning model to perform model training on the machine learning model to obtain a training model, constructing a test sample of the training model, and calculating the mean absolute error of the training model based on the test sample using the following formula:

[0128] ;

[0129] in, represents the mean absolute error, N represents the number of test samples, represents the true value of the i-th data of the test sample, Represents the predicted value of the i-th data of the test sample;

[0130] According to the mean absolute error, the training model is performance evaluated to obtain a performance evaluation result, and when the performance evaluation result meets a preset result, an ecological structure model is obtained.

[0131] Optionally, the integrated data can be obtained by assimilating the temperature sensor data, the salinity data and the dissolved oxygen data using data assimilation techniques, such as Kalman filtering, ensemble Kalman filtering, etc., and the machine learning model is trained to obtain a training model by collecting sensor data, salinity data and dissolved oxygen data. The test sample can be constructed by integrating the data. It should be noted that when the performance evaluation result meets the preset result, the ecological structure model is obtained, which means that the calculated result of the mean absolute error is not greater than 0.1, and the specific setting can be combined with the actual application scenario.

[0132] S5. Based on the eutrophication division of the ocean area, the concentration of suspended particulate matter, the level of water quality change and the degree of ecological structure change, a comprehensive ocean multi-attribute map of the seawater monitoring area is constructed, and multi-source seawater data of the seawater monitoring area is collected in real time, so as to update the comprehensive ocean multi-attribute map using the multi-source seawater data to obtain a real-time updated attribute map, and the water quality level of the seawater monitoring area is analyzed using the real-time updated attribute map, and when the water quality level does not meet the preset value, an early warning is issued to the corresponding monitoring users of the seawater monitoring area.

[0133] The embodiment of the present invention constructs a comprehensive ocean multi-attribute map of the seawater monitoring area by dividing the ocean areas based on the eutrophication, the suspended particulate matter concentration, the water quality change level and the degree of ecological structure change. It can integrate multiple important ocean attribute parameters, comprehensively reflect the overall status of the seawater monitoring area from different angles, and can help users quickly understand the ocean conditions in different areas to formulate targeted management measures, such as taking measures to reduce the input of nutrients in eutrophic areas and strengthening pollution control in areas with poor water quality.

[0134] The multi-attribute comprehensive graph refers to a visual graph formed by integrating multiple important attributes of the seawater monitoring area into one graph.

[0135] As an embodiment of the present invention, the multi-attribute comprehensive ocean map of the seawater monitoring area is constructed based on the eutrophication division of the ocean areas, the suspended particulate matter concentration, the water quality change level and the degree of change of the ecological structure, including: based on the division rules of the eutrophication division of the ocean areas, the suspended particulate matter concentration, the water quality change level and the degree of change of the ecological structure are divided into regions to obtain multi-layer division areas, attribute labeling and color labeling of the multi-layer division areas to obtain labeled areas, constructing an ocean view of the seawater monitoring area, and using the labeled areas to fill the ocean view to obtain a multi-attribute comprehensive map.

[0136] It should be explained that the division rule of the eutrophication marine area is based on the chlorophyll concentration value, that is, it is divided according to the change in the numerical value. For example, the suspended particulate matter concentration is divided according to the particle concentration size. The multi-layer divided areas are attribute labeled and color labeled to obtain the labeled area, that is, the labeled data and the division color. For example, an area is labeled as the suspended particulate matter concentration area A, with a concentration of 0.6H, a low concentration. By labeling in this way, the ocean view can be carefully drawn through GIS.

[0137] Furthermore, the embodiment of the present invention collects the multi-source seawater data of the seawater monitoring area in real time, and uses the multi-source seawater data to update the ocean multi-attribute comprehensive map, so that the real-time updated attribute map can provide continuous information support for users, so that they can adjust management strategies and measures according to real-time conditions. For example, if it is found that the water quality in a certain area suddenly deteriorates, emergency measures can be quickly taken to deal with it.

[0138] The real-time updated attribute map can be obtained by continuously collecting multi-source seawater data, including satellite remote sensing data, sensor data, field sampling data, etc., and using these data to update the ocean multi-attribute comprehensive map in real time.

[0139] Furthermore, the embodiment of the present invention uses the real-time updated attribute map to analyze the water quality level of the seawater monitoring area, and when the water quality level does not meet the preset value, an early warning is issued to the corresponding monitoring user in the seawater monitoring area, so as to quickly discover abnormal water quality and provide a basis for taking corresponding treatment measures. For example, if the water quality level drops to a certain level, it may affect the survival and reproduction of marine organisms, and timely measures need to be taken to improve it.

[0140] The water quality level can be determined by setting a series of water quality assessment indicators and standards, such as different eutrophication levels and suspended particulate matter concentration ranges corresponding to different water quality levels, and by comparing the relevant attribute values ​​in the real-time updated attribute map, the water quality level of the seawater monitoring area can be determined. Data analysis algorithms and models can be used to conduct a comprehensive analysis of multiple attributes to obtain an assessment result of the water quality level. Early warning can be issued through the alarm system of the seawater monitoring area.

[0141] It should be noted that in actual marine water quality monitoring, the setting of preset thresholds needs to consider many factors. For example, for a sea area dominated by fisheries, the preset thresholds may be relatively strict, because lower water quality levels may directly affect the survival and reproduction of fish, thereby causing significant losses to fishery production. In this case, preset thresholds for indicators such as dissolved oxygen, nutrient content, and suspended particle concentration may be set based on the fish's requirements for water quality, combined with historical water quality data and relevant standards; and for a sea area mainly used for tourism, water quality is crucial to the experience and health of tourists. The preset thresholds may focus on ensuring the transparency and cleanliness of seawater. For example, lower preset thresholds for suspended particle concentrations and bacterial content are set to ensure that tourists can safely engage in water activities such as swimming and diving;

[0142] In addition, the preset thresholds also need to consider the stability of the marine ecosystem. If a sea area is an important marine ecological protection area, the setting of the preset thresholds should pay more attention to protecting rare marine life and maintaining ecological balance. For example, for areas sensitive to eutrophication, lower preset thresholds for the content of nutrients such as nitrogen and phosphorus should be set to prevent eutrophication from damaging the ecosystem; in short, the setting of preset thresholds should be based on different actual application scenarios, taking into account the use of the ocean, ecological characteristics, economic value, and relevant laws, regulations and standards, so as to ensure that water quality problems can be warned and handled in a timely and effective manner.

[0143] like Figure 2 , which is a functional module diagram of a seawater quality monitoring system based on automatic multi-source data assimilation provided by an embodiment of the present invention.

[0144] The seawater quality monitoring system 200 based on automatic multi-source data assimilation of the present invention can be installed in an electronic device. According to the functions implemented, the seawater quality monitoring system 200 based on automatic multi-source data assimilation can include a eutrophication area division module 201, a suspended particle concentration analysis module 202, a water quality change analysis module 203, an ecological structure model construction module 204 and a water quality early warning module 205. The module of the present invention can also be called a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0145] In this embodiment, the functions of each module / unit are as follows:

[0146] The eutrophication area division module 201 is used to collect satellite remote sensing images of the seawater monitoring area, calibrate the satellite remote sensing images to obtain calibrated images, identify image information of the calibrated images, and divide the seawater monitoring area into eutrophic levels based on the image information to obtain eutrophication divided marine areas;

[0147] The suspended particle concentration analysis module 202 is used to collect optical sensor data of the seawater monitoring area, analyze the absorption characteristics of particles in the seawater monitoring area using the optical sensor data, and analyze the suspended particle concentration in the seawater monitoring area based on the absorption characteristics;

[0148] The water quality change analysis module 203 is used to collect temperature sensor data of the seawater monitoring area, analyze the degree of change of seawater density and the degree of change of seawater transparency in the seawater monitoring area based on the temperature sensor data, and analyze the level of water quality change in the seawater monitoring area based on the degree of change of seawater density and the degree of change of seawater transparency;

[0149] The ecological structure model building module 204 is used to collect salinity data and dissolved oxygen data of the seawater monitoring area, and build an ecological structure model of the seawater monitoring area based on the temperature sensor data, the salinity data and the dissolved oxygen data, so as to analyze the degree of change of the ecological structure of the seawater monitoring area by using the ecological structure model;

[0150] The water quality warning module 205 is used to construct a comprehensive ocean multi-attribute map of the seawater monitoring area based on the eutrophication division of the ocean area, the concentration of suspended particulate matter, the level of water quality change and the degree of ecological structure change, and to collect multi-source seawater data of the seawater monitoring area in real time, so as to update the ocean multi-attribute comprehensive map using the multi-source seawater data to obtain a real-time updated attribute map, analyze the water quality level of the seawater monitoring area using the real-time updated attribute map, and issue a warning to the corresponding monitoring users of the seawater monitoring area when the water quality level does not meet the preset value.

[0151] In detail, the modules described in the seawater quality monitoring system 200 based on automatic multi-source data assimilation described in the embodiment of the present invention adopt the same technical means as the seawater quality monitoring method based on automatic multi-source data assimilation described in the accompanying drawings when used, and can produce the same technical effects, which will not be repeated here.

[0152] In the several embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are only illustrative, for example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.

[0153] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0154] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0155] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0156] Therefore, no matter from which point of view, the embodiments should be regarded as illustrative and non-restrictive, and the scope of the present invention is limited by the appended claims rather than the above description, so it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any attached figure mark in the claims should not be regarded as limiting the claims involved.

[0157] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Artificial Intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0158] In addition, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or systems stated in a system claim can also be implemented by one unit or system through software or hardware. The words first, second, etc. are used to indicate names, and do not indicate any particular order.

[0159] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for monitoring seawater quality based on automatic multi-source data assimilation, characterized in that: The method comprises: Collect satellite remote sensing images of the seawater monitoring area, perform image calibration on the satellite remote sensing images to obtain calibration images, identify image information of the calibration images, and divide the seawater monitoring area into eutrophic levels based on the image information to obtain eutrophic marine areas; wherein, performing image calibration on the satellite remote sensing images to obtain calibration images includes: extracting target points in the satellite remote sensing images, and calculating the echo energy of the target points using the following formula: ； in, represents the echo energy, m represents the total number of target points, Represents the pixel intensity value of the target point area image, Represents the number of pixels in the target point area image, Indicates the number of pixels of the echo area image corresponding to the target point, Indicates the total number of echo points corresponding to the target point, Indicates the pixel size of the target point area image, Indicates the pixel size of the echo area image corresponding to the target point, The relative radiometric calibration accuracy of the satellite remote sensing image is queried, and the radiometric calibration constant of the satellite remote sensing image is calculated using the following formula according to the echo energy and the relative radiometric calibration accuracy: ； in, represents the calibration constant, p represents the echo energy, Represents the radiation area of ​​satellite remote sensing images, Indicates the radiation angle of satellite remote sensing images; According to the radiation calibration constant, the satellite remote sensing image is calibrated radiometrically to obtain a calibrated image; The identifying the image information of the calibration image comprises: performing image decomposition on the calibration image to obtain a decomposed image, extracting wavelet coefficients of the decomposed image, performing image fusion on the decomposed image according to the wavelet coefficients to obtain a fused image, detecting image corner points of the fused image, performing multi-scale characterization on the image corner points to obtain multi-scale corner points, and using the multi-scale corner points to identify the image information of the calibration image; The step of performing image fusion on the decomposed images according to the wavelet coefficients to obtain a fused image includes: calculating the mean square error of the wavelet coefficients using the following formula: ； in, represents the mean square error, M represents the number of decomposed images, represents the wavelet coefficient of the u-th decomposed image, represents the wavelet coefficients of the vth decomposition image, Based on the mean square error, performing image fusion on the decomposed images in the same frequency domain in the decomposed images to obtain a fused image; Furthermore, the performing image fusion on the decomposed images according to the wavelet coefficients to obtain a fused image includes: The mean square error of the wavelet coefficients is calculated using the following formula: ； in, represents the mean square error, M represents the number of decomposed images, represents the wavelet coefficient of the u-th decomposed image, represents the wavelet coefficients of the vth decomposition image, Based on the mean square error, performing image fusion on the decomposed images in the same frequency domain in the decomposed images to obtain a fused image; The eutrophication level division of the seawater monitoring area to obtain the eutrophication-divided ocean area includes: querying the band information in the image information, calculating the chlorophyll index of the seawater monitoring area based on the band information, and dividing the seawater monitoring area into eutrophication levels based on the chlorophyll index to obtain the eutrophication-divided ocean area; Collecting optical sensor data of the seawater monitoring area, analyzing the absorption characteristics of particulate matter in the seawater monitoring area using the optical sensor data, and analyzing the concentration of suspended particulate matter in the seawater monitoring area based on the absorption characteristics; wherein, analyzing the absorption characteristics of particulate matter in the seawater monitoring area using the optical sensor data comprises: performing data preprocessing on the optical sensor data to obtain preprocessed data, performing data calibration on the preprocessed data to obtain calibration data, calculating the absorption coefficient of the particulate matter based on the calibration data, and analyzing the absorption characteristics of the particulate matter based on the absorption coefficient; The analyzing the suspended particle concentration in the seawater monitoring area based on the absorption characteristics includes: constructing seawater optical characteristic parameters in the seawater monitoring area based on the absorption characteristics, querying historical water optical characteristic parameters in the seawater monitoring area, constructing a historical turbidity curve in the seawater monitoring area based on the historical water optical characteristic parameters, and analyzing the suspended particle concentration in the seawater monitoring area based on the historical turbidity curve and the seawater optical characteristic parameters; Collect temperature sensor data of the seawater monitoring area, analyze the degree of change of seawater density and the degree of change of seawater transparency in the seawater monitoring area based on the temperature sensor data, and analyze the level of change of water quality in the seawater monitoring area based on the degree of change of seawater density and the degree of change of seawater transparency; wherein, analyzing the level of change of water quality in the seawater monitoring area based on the degree of change of seawater density and the degree of change of seawater transparency comprises: querying important parameters affecting water quality in the seawater monitoring area, constructing a water quality evaluation factor set of the seawater monitoring area based on the important parameters affecting water quality, the degree of change of seawater density and the degree of change of seawater transparency, constructing an evaluation grade of the water quality change level in the seawater monitoring area, constructing a membership function of each evaluation index in the evaluation factor set, and assigning an index weight of the evaluation index, and analyzing the level of change of water quality in the seawater monitoring area based on the water quality evaluation factor set, the evaluation grade, the membership function and the index weight; Collect salinity data and dissolved oxygen data of the seawater monitoring area, and construct an ecological structure model of the seawater monitoring area based on the temperature sensor data, the salinity data and the dissolved oxygen data, so as to analyze the degree of change of the ecological structure of the seawater monitoring area by using the ecological structure model; Based on the eutrophication-divided ocean areas, the suspended particulate matter concentration, the water quality change level and the degree of ecological structure change, a comprehensive ocean multi-attribute map of the seawater monitoring area is constructed, and multi-source seawater data of the seawater monitoring area is collected in real time, so that the multi-source seawater data is used to update the comprehensive ocean multi-attribute map to obtain a real-time updated attribute map, and the real-time updated attribute map is used to analyze the water quality level of the seawater monitoring area, and when the water quality level does not meet the preset value, an early warning is issued to the corresponding monitoring users of the seawater monitoring area.

2. A method for monitoring seawater quality based on automatic multi-source data assimilation as claimed in claim 1, characterized in that: The step of constructing an ecological structure model of the seawater monitoring area based on the temperature sensor data, the salinity data and the dissolved oxygen data comprises: Performing data integration on the temperature sensing data, the salinity data and the dissolved oxygen data to obtain integrated data; Inputting the integrated data into a preconfigured machine learning model to perform model training on the machine learning model to obtain a training model; Constructing a test sample of the training model; Based on the test samples, the mean absolute error of the training model is calculated using the following formula: ； in, represents the mean absolute error, N represents the number of test samples, represents the true value of the i-th data of the test sample, Represents the predicted value of the i-th data of the test sample; Performing a performance evaluation on the training model according to the mean absolute error to obtain a performance evaluation result; When the performance evaluation result meets the preset result, an ecological structure model is obtained.

3. A seawater quality monitoring system based on automatic multi-source data assimilation, characterized in that: The system is used to implement a method for monitoring seawater quality based on automatic multi-source data assimilation as claimed in any one of claims 1 to 2, the system comprising: The eutrophication area division module is used to collect satellite remote sensing images of the seawater monitoring area, perform image calibration on the satellite remote sensing images to obtain calibration images, identify image information of the calibration images, and perform eutrophication level division on the seawater monitoring area based on the image information to obtain eutrophication divided ocean areas; wherein, the image calibration of the satellite remote sensing images to obtain the calibration images includes: extracting target points in the satellite remote sensing images, and calculating the echo energy of the target points using the following formula: ； in, represents the echo energy, m represents the total number of target points, Represents the pixel intensity value of the target point area image, Represents the number of pixels in the target point area image, Indicates the number of pixels of the echo area image corresponding to the target point, Indicates the total number of echo points corresponding to the target point, Indicates the pixel size of the target point area image, Indicates the pixel size of the echo area image corresponding to the target point, The relative radiometric calibration accuracy of the satellite remote sensing image is queried, and the radiometric calibration constant of the satellite remote sensing image is calculated using the following formula according to the echo energy and the relative radiometric calibration accuracy: ； in, represents the calibration constant, p represents the echo energy, Represents the radiation area of ​​satellite remote sensing images, Indicates the radiation angle of satellite remote sensing images; According to the radiation calibration constant, the satellite remote sensing image is calibrated radiometrically to obtain a calibrated image; The identifying the image information of the calibration image comprises: performing image decomposition on the calibration image to obtain a decomposed image, extracting wavelet coefficients of the decomposed image, performing image fusion on the decomposed image according to the wavelet coefficients to obtain a fused image, detecting image corner points of the fused image, performing multi-scale characterization on the image corner points to obtain multi-scale corner points, and using the multi-scale corner points to identify the image information of the calibration image; The step of performing image fusion on the decomposed images according to the wavelet coefficients to obtain a fused image includes: calculating the mean square error of the wavelet coefficients using the following formula: ； in, represents the mean square error, M represents the number of decomposed images, represents the wavelet coefficient of the u-th decomposed image, represents the wavelet coefficients of the vth decomposition image, Based on the mean square error, performing image fusion on the decomposed images in the same frequency domain in the decomposed images to obtain a fused image; Furthermore, the performing image fusion on the decomposed images according to the wavelet coefficients to obtain a fused image includes: The mean square error of the wavelet coefficients is calculated using the following formula: ； in, represents the mean square error, M represents the number of decomposed images, represents the wavelet coefficient of the u-th decomposed image, represents the wavelet coefficients of the vth decomposition image, Based on the mean square error, performing image fusion on the decomposed images in the same frequency domain in the decomposed images to obtain a fused image; The eutrophication level division of the seawater monitoring area to obtain the eutrophication-divided ocean area includes: querying the band information in the image information, calculating the chlorophyll index of the seawater monitoring area based on the band information, and dividing the seawater monitoring area into eutrophication levels based on the chlorophyll index to obtain the eutrophication-divided ocean area; A suspended particle concentration analysis module is used to collect optical sensor data of the seawater monitoring area, analyze the absorption characteristics of particles in the seawater monitoring area using the optical sensor data, and analyze the suspended particle concentration in the seawater monitoring area based on the absorption characteristics; wherein, the use of the optical sensor data to analyze the absorption characteristics of particles in the seawater monitoring area includes: preprocessing the optical sensor data to obtain preprocessed data, calibrating the preprocessed data to obtain calibration data, calculating the absorption coefficient of the particles based on the calibration data, and analyzing the absorption characteristics of the particles based on the absorption coefficient; the analysis of the suspended particle concentration in the seawater monitoring area based on the absorption characteristics includes: constructing seawater optical characteristic parameters in the seawater monitoring area based on the absorption characteristics, querying historical water optical characteristic parameters in the seawater monitoring area, constructing a historical turbidity curve in the seawater monitoring area based on the historical water optical characteristic parameters, and analyzing the suspended particle concentration in the seawater monitoring area based on the historical turbidity curve and the seawater optical characteristic parameters; A water quality change analysis module is used for the temperature sensing data of the seawater monitoring area, and based on the temperature sensing data, analyzes the degree of change of seawater density and the degree of change of seawater transparency in the seawater monitoring area, and analyzes the level of water quality change in the seawater monitoring area based on the degree of change of seawater density and the degree of change of seawater transparency; wherein, the analysis of the level of water quality change in the seawater monitoring area based on the degree of change of seawater density and the degree of change of seawater transparency comprises: querying important parameters affecting water quality in the seawater monitoring area, constructing a water quality evaluation factor set for the seawater monitoring area based on the important parameters affecting water quality, the degree of change of seawater density and the degree of change of seawater transparency, constructing an evaluation grade for the level of water quality change in the seawater monitoring area, constructing a membership function of each evaluation index in the evaluation factor set, and assigning an index weight of the evaluation index, and analyzing the level of water quality change in the seawater monitoring area based on the water quality evaluation factor set, the evaluation grade, the membership function and the index weight; An ecological structure model building module is used to collect salinity data and dissolved oxygen data of the seawater monitoring area, and build an ecological structure model of the seawater monitoring area based on the temperature sensor data, the salinity data and the dissolved oxygen data, so as to use the ecological structure model to analyze the degree of change of the ecological structure of the seawater monitoring area; The water quality early warning module is used to construct a comprehensive ocean multi-attribute map of the seawater monitoring area based on the eutrophication division of the ocean area, the concentration of suspended particulate matter, the level of water quality change and the degree of ecological structure change, collect seawater multi-source data of the seawater monitoring area in real time, and use the seawater multi-source data to update the ocean multi-attribute comprehensive map to obtain a real-time updated attribute map, use the real-time updated attribute map to analyze the water quality level of the seawater monitoring area, and when the water quality level does not meet the preset value, issue an early warning to the corresponding monitoring user of the seawater monitoring area.

Images