River reservoir suspended sediment remote sensing recognition system and method based on water color difference

Abstract

The invention relates to a river reservoir suspended sediment remote sensing recognition system and method based on water color difference, and belongs to the technical field of water ecological environment monitoring. The method comprises the following steps: firstly, acquiring a multispectral satellite image through a GEE cloud platform, and extracting reflectivity data of main and branch flows through preprocessing such as atmospheric correction and image fusion; performing space-time registration on the in-situ turbidity measured value and the satellite reflectivity to construct a training data set; aiming at different water color characteristics of the main stream and the branch stream, respectively establishing a support vector regression machine learning model of the main stream and a multi-hydrological-period empirical model of the branch stream, and realizing high-precision zoning inversion of turbidity; and finally, identifying the suspended sediment water body through turbidity grading, and generating a spatial-temporal distribution map. The method effectively solves the problem of water color difference of the main stream and the branch stream caused by reservoir regulation, remarkably improves the precision and the reliability of remote sensing monitoring of the suspended sediment, and provides technical support for river reservoir sediment treatment and water environment management.

Description

Technical Field

[0001] This invention relates to a remote sensing identification system and method for suspended sediment in rivers and reservoirs based on water color differences, belonging to the field of water ecological environment monitoring technology, and is particularly applicable to remote sensing identification of suspended sediment in rivers and reservoirs based on water color differences. Background Technology

[0002] Suspended sediment is a core parameter determining the optical properties of river and reservoir waters, such as transparency, turbidity, and color. It is also a crucial foundation for studying water environment processes like sediment transport, geomorphological evolution, and eutrophication in rivers and reservoirs. Understanding the spatial distribution and temporal evolution of suspended sediment not only provides data support for research on the spatiotemporal evolution mechanisms of suspended sediment in rivers and reservoirs but also offers decision-making basis for flood control and emergency response, dam operation and maintenance, and navigation scheduling.

[0003] Currently, conventional methods for monitoring suspended sediment mainly rely on station-based monitoring that can continuously and accurately obtain suspended sediment content over a time scale. However, the high cost of station construction and maintenance results in a small number of stations with sparse distribution, making large-scale continuous spatial monitoring impossible. More importantly, artificial water storage and release operations significantly alter the hydrological conditions of river and reservoir main streams and tributaries. Coupled with the "storing clear water and discharging turbid water" scheduling strategy, this leads to frequent and drastic fluctuations in suspended sediment content in the backwater areas of river and reservoir main streams and tributaries, creating drastically different water color conditions in specific spatiotemporal domains. Therefore, existing monitoring methods are insufficient to reveal the spatiotemporal characteristics of suspended sediment in river and reservoir main streams and tributaries and their response to environmental changes. There is an urgent need to establish a remote sensing model for suspended sediment in river and reservoir main streams and tributaries to achieve dynamic spatiotemporal monitoring of suspended sediment, thereby providing technical support for river and reservoir sedimentation control, water environment protection, and navigation safety. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art and provide a remote sensing identification system and method for suspended sediment in rivers and reservoirs based on water color differences. By acquiring satellite multispectral image data and combining zonal modeling and multi-source data fusion, the spatiotemporal evolution process of suspended sediment can be accurately depicted.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A remote sensing identification system for suspended sediment in rivers and reservoirs based on water color differences, combined with Figure 1 Its features include a data acquisition module, a data preprocessing and fusion module, a spatiotemporal registration engine, a turbidity inversion module, a suspended sediment identification and classification module, and a report generation and visualization platform connected in sequence.

[0007] The data acquisition module includes a satellite remote sensing data acquisition unit and an in-situ turbidity data acquisition unit. The satellite remote sensing data acquisition unit interfaces with the Google Earth Engine (GEE) cloud platform API to automatically search and download multispectral satellite images according to spatiotemporal range. The in-situ turbidity data acquisition unit receives and manages in-situ observation turbidity data from monitoring stations and ship-based measurements to obtain measured turbidity values.

[0008] The data preprocessing and fusion module includes an image correction unit, an image fusion unit, and an image reflectance extraction unit. The image correction unit is connected to the satellite remote sensing data acquisition unit and automatically performs radiometric calibration, atmospheric correction, and geometric fine correction on multispectral satellite images. The image fusion unit is connected to the image correction unit and uses a pixel-level fusion algorithm to fuse high spatiotemporal resolution images, solving the problems of cloud cover and data loss. The image reflectance extraction unit is connected to the image fusion unit and automatically segments and extracts reflectance data from the main stream area and each tributary area.

[0009] The spatiotemporal registration engine is connected to the output of the in-situ turbidity data acquisition unit and the output of the image reflectance extraction unit, respectively. It automatically matches the position and time of the in-situ measurement point with the time of the multispectral satellite image to construct paired data of reflectance and turbidity under the same spatiotemporal conditions.

[0010] The turbidity inversion module includes a main stream turbidity inversion unit and a tributary turbidity inversion unit. The main stream turbidity inversion unit is connected to the output of the spatiotemporal registration engine and uses a support vector regression (SVR) machine learning model to establish a machine learning relationship model between remotely sensed reflectance and measured turbidity values, thereby realizing remote sensing inversion of turbidity in the main stream of rivers and reservoirs. The tributary turbidity inversion unit is connected to the output of the spatiotemporal registration engine and calculates the correlation coefficient between turbidity and reflectance in each band according to the reservoir operation cycle, establishing empirical formula models for different cycles, thereby realizing remote sensing inversion of turbidity in the tributaries of rivers and reservoirs.

[0011] The suspended sediment identification and classification module is a classifier that is connected to the output of the main stream turbidity inversion unit and the tributary turbidity inversion unit. According to the set turbidity threshold, the turbidity inversion results of the main stream and tributaries are uniformly divided into four levels: clear, low, medium and high, and "high turbidity" water bodies are automatically identified as "suspended sediment water bodies".

[0012] The report generation and visualization platform is connected to the output of the suspended sediment identification and classification unit, and displays the identification results of suspended sediment to the water environment management department in the form of charts, reports and interactive web maps.

[0013] A remote sensing identification method for suspended sediment in rivers and reservoirs based on water color differences, combined with Figure 2 It includes the following steps:

[0014] S1: Data Acquisition and Preprocessing: Based on the GEE platform, multispectral satellite images are acquired, and after correction, fusion, and segmentation, multispectral reflectance data of the main stream area and tributary areas of rivers and reservoirs are extracted;

[0015] S2: Dataset Construction: Spatiotemporal registration will be performed between in-situ measured turbidity data and multispectral reflectance data from multispectral satellite images; based on historical reflectance and turbidity pairing data, combined with human annotations of four levels (clear, low, medium, and high), a dataset for model training and validation will be constructed.

[0016] S3: Turbidity Inversion Model Construction and Training: Turbidity inversion models are established for the different optical characteristics of the main stream and tributaries: the turbidity inversion model of the main stream adopts the support vector regression model, and the turbidity inversion model of the tributaries is based on the empirical formula model constructed on the sensitive band; then the turbidity inversion model is trained using the dataset from step S2.

[0017] S4: Suspended sediment identification and analysis: The turbidity of the main stream and tributaries predicted by the turbidity inversion model is input into the classifier and classified into four levels of suspended sediment: clear, low, medium and high. The classifier is then trained using the dataset from step S2.

[0018] S5: Analyze the spatial distribution and temporal evolution characteristics of the suspended sediment identification results and present them to the water environment management department in the form of charts, reports and interactive web maps.

[0019] Furthermore, step S1 specifically includes:

[0020] S101: The satellite remote sensing data acquisition unit interfaces with the GEE cloud platform API to automatically acquire multispectral satellite images of the area;

[0021] S102: The image correction unit performs radiometric calibration, atmospheric correction, and geometric fine correction on the original multispectral satellite imagery in sequence to eliminate sensor and atmospheric interference;

[0022] S103: The image fusion unit uses a pixel-level fusion algorithm to fuse multispectral satellite images processed in step S102 at multiple times to solve the problems of cloud coverage and data loss, and generate high-quality seamless reflectance images.

[0023] S104: The image reflectance extraction unit performs water body masking and partitioning based on the water body index and the vector boundaries of the main stream and tributaries, and finally extracts the independent multispectral reflectance data of the main stream and each tributary.

[0024] Furthermore, the spatiotemporal registration rules described in step S2 are as follows: (a) Location registration: Match the latitude and longitude coordinates corresponding to the measured turbidity values ​​from the monitoring station and ship with the image reflectance data from the GEE platform; (b) Time registration: If there are measured turbidity values ​​on a certain day but no satellite image reflectance data, then select the closest image reflectance data within ±7 days to pair with the measured turbidity values; if there are multiple measured turbidity values ​​on the same day, then select the measured turbidity value closest to 12 noon to reduce the impact of daily variations.

[0025] Furthermore, the working principle of the main stream turbidity inversion model described in step S3 is as follows:

[0026] S3011: Extract the blue, green, red, near-infrared, and short-wave infrared frequencies from the paired data of reflectance and turbidity according to the frequency of reflectance;

[0027] S3012: Logarithmic transformation was performed on the measured turbidity values ​​in the paired data of reflectance and turbidity;

[0028] S3013: Perform One-Hot encoding on hydrological period information;

[0029] S3014: Concatenate the data from steps S3011 to S3013 and use the results as input to the support vector regression model;

[0030] S3015: Set up the hyperparameter search space for the support vector regression model, use the dataset from step S2 to find the optimal parameter combination in the parameter space through grid search, and construct a high-performance, highly generalizable main stream turbidity inversion model to realize remote sensing inversion of turbidity in river and reservoir main streams.

[0031] Furthermore, the specific working principle of the tributary turbidity inversion model described in step S3 is as follows:

[0032] S3021: Extract the blue, green, red, near-infrared, and short-wave infrared frequencies from the paired data of reflectance and turbidity according to the frequency of reflectance;

[0033] S3022: The paired data of reflectance and turbidity are divided according to the discharge period, flood period, water storage period, and high water level operation period;

[0034] S3023: Establish the measured turbidity value using a multivariate empirical formula model ( ) and blue light Green light Red light Near-infrared and shortwave infrared Quantitative relationship of reflectivity across different wavebands;

[0035] Specifically,

[0036] Discharge period: ;

[0037] Flood season: ;

[0038] Water storage period: ;

[0039] High water level operation period: ;

[0040] in, , The correlation coefficient is the coefficient to be trained.

[0041] In particular, when analyzing the optical characteristics of the main stream of rivers and reservoirs, it is necessary to consider the response characteristics of phytoplankton, suspended particulate matter, and yellow substances to remote sensing reflectance in different bands. Because the main stream environment is relatively complex, multiple spectral bands of reflectance that are sensitive to changes in turbidity are selected as independent variables, and the optimal combination is chosen as the independent variable. Five key bands, including blue light, green light, red light, near-infrared and short-wave infrared, are selected.

[0042] Furthermore, the classifier described in step S4 includes, but is not limited to, logistic regression, random forest, support vector machine, neural network, Naive Bayes classifier, and gradient boosting tree.

[0043] An electronic device includes at least one processor; and a memory communicatively connected to said at least one processor; wherein,

[0044] The memory stores a computer program that is executed by the at least one processor, which enables the at least one processor to perform the above-described remote sensing identification method for suspended sediment in rivers and reservoirs based on water color differences.

[0045] Finally, the present invention also discloses a computer-readable storage medium storing computer instructions, which are used to cause a processor to execute the above-described remote sensing identification method for suspended sediment in rivers and reservoirs based on water color differences.

[0046] The beneficial effects of this invention are as follows: It provides a remote sensing identification system and method for suspended sediment in rivers and reservoirs based on water color differences. By constructing a remote sensing identification system for suspended sediment in rivers and reservoirs based on water color differences, it achieves high-precision and automated monitoring of the spatiotemporal distribution of suspended sediment in rivers and reservoirs. This system innovatively adopts a tributary-mainstream regional inversion strategy, employing a support vector regression machine learning model for the complex optical characteristics of the mainstream and establishing multi-mode empirical models for different hydrological periods of the tributaries, effectively solving the problem of water color differences between mainstream and tributaries caused by reservoir scheduling. This invention significantly improves the accuracy and availability of turbidity inversion through multi-source remote sensing data fusion and spatiotemporal registration technology. The final generated suspended sediment classification map and spatiotemporal evolution report provide reliable decision support for river and reservoir sediment deposition control, water environment protection, and navigation safety. Attached Figure Description

[0047] To make the objectives and technical solutions of this invention clearer, the following figures are provided for illustration:

[0048] Figure 1 This is an architecture diagram of the remote sensing identification system for suspended sediment in rivers and reservoirs based on water color differences in this invention; the arrows indicate the direction of data transmission.

[0049] Figure 2 This is a flowchart of the remote sensing identification method for suspended sediment in rivers and reservoirs based on water color differences in this invention.

[0050] Figure 3 The remote sensing identification results of suspended sediment in the main stream and tributaries of the Three Gorges Reservoir in Embodiment 2 of the present invention;

[0051] Figure 4 This is a schematic diagram of the electronic device in Embodiment 3 of the present invention. Detailed Implementation

[0052] To make the objectives and technical solutions of this invention clearer, the invention will be described in detail below with reference to the accompanying drawings and embodiments.

[0053] Example 1: This invention identifies and estimates suspended sediment in typical tributaries of the Three Gorges Reservoir, including the Xiaojiang River, Caotang River, Daning River, and Xiangxi River, as well as their connecting main streams, to provide reliable decision support for river and reservoir sedimentation control, water environment protection, and navigation safety. The invention proposes a "remote sensing identification system for suspended sediment in rivers and reservoirs based on water color differences."

[0054] Combination Figure 1 The system consists of a data acquisition module 1, a data preprocessing and fusion module 2, a spatiotemporal registration engine 3, a turbidity inversion module 4, a suspended sediment identification and classification module 5, and a report generation and visualization platform 6 connected in sequence.

[0055] The data acquisition module includes a satellite remote sensing data acquisition unit 11 and an in-situ turbidity data acquisition unit 12. The satellite remote sensing data acquisition unit 11 interfaces with the GEE cloud platform API to automatically search and download multispectral satellite images such as Landsat 8 / 9 OLI and Sentinel-2 according to spatiotemporal range. The in-situ turbidity data acquisition unit 12 receives and manages in-situ observation turbidity data such as turbidity from monitoring stations and ship-based measurements to obtain measured turbidity values.

[0056] The data preprocessing and fusion module 2 includes an image correction unit 21, an image fusion unit 22, and an image reflectance extraction unit 23. The image correction unit 21 is connected to the satellite remote sensing data acquisition unit 11 and automatically performs radiometric calibration, atmospheric correction, and geometric fine correction on multispectral satellite images. The image fusion unit 22 is connected to the image correction unit 21 and uses a pixel-level fusion algorithm to fuse high spatiotemporal resolution images to solve the problems of cloud cover and data loss. The image reflectance extraction unit 23 is connected to the image fusion unit 22 and automatically segments and extracts reflectance data from the main stream area and each tributary area.

[0057] The spatiotemporal registration engine 3 is connected to the output of the in-situ turbidity data acquisition unit 12 and the output of the image reflectance extraction unit 23, respectively. It automatically matches the position and time of the in-situ measurement point with the time of the multispectral satellite image to construct paired data of reflectance and turbidity under the same spatiotemporal conditions.

[0058] The turbidity inversion module 4 includes a main stream turbidity inversion unit 41 and a tributary turbidity inversion unit 42. The main stream turbidity inversion unit 41 is connected to the output of the spatiotemporal registration engine 3. It uses a support vector regression machine learning model to establish a machine learning relationship model between remote sensing reflectance and measured turbidity values, thereby realizing remote sensing inversion of turbidity in the main stream of the river and reservoir. The tributary turbidity inversion unit 42 is connected to the output of the spatiotemporal registration engine 3. Based on the reservoir operation cycle, it calculates the correlation coefficient between turbidity and reflectance in each band, establishes an empirical formula model for different cycles, and realizes remote sensing inversion of turbidity in the tributary of the river and reservoir.

[0059] The suspended sediment identification and classification module 5 is a classifier that is connected to the output of the main stream turbidity inversion unit 41 and the tributary turbidity inversion unit 42. According to the set turbidity threshold, the turbidity inversion results of the main stream and tributaries are uniformly divided into four levels: clear, low, medium and high, and the "high turbidity" water body is automatically identified as "suspended sediment water body".

[0060] The report generation and visualization platform 6 is connected to the output of the suspended sediment identification and classification unit 5, and displays the identification results of suspended sediment to the water environment management department in the form of charts, reports and interactive web maps.

[0061] Example 2: In response to the scenario of Example 1, the present invention also proposes a "remote sensing identification method for suspended sediment in rivers and reservoirs based on water color differences".

[0062] Combination Figure 2 The method includes the following steps:

[0063] S1: Data Acquisition and Preprocessing: Satellite remote sensing data acquisition unit 11 acquires multispectral satellite images based on the GEE platform. After correction, fusion, and segmentation by data preprocessing and fusion module 2, multispectral reflectance data of the main stream area and tributary areas of the Three Gorges Reservoir are extracted.

[0064] S2: Dataset Construction: The spatiotemporal registration engine 3 will perform spatiotemporal registration between the turbidity data measured in situ and the multispectral reflectance data of the multispectral satellite image; based on the paired data of historical reflectance and turbidity, combined with the human annotations of four levels (clear, low, medium and high), a dataset for model training and validation will be constructed in a 7:3 ratio.

[0065] S3: Turbidity Inversion Model Construction and Training: Turbidity inversion models are established for the different optical characteristics of the main stream and tributaries: the main stream turbidity inversion model of the main stream turbidity inversion unit 41 adopts the support vector regression model, and the tributary turbidity inversion model of the tributary turbidity inversion unit 42 is based on the empirical formula model constructed based on the sensitive band; then the turbidity inversion model is trained using the dataset from step S2;

[0066] S4: Suspended sediment identification and analysis: The turbidity of the main stream and tributaries predicted by the turbidity inversion model is input into the classifier of the suspended sediment identification and classification module 5 and divided into four levels of suspended sediment identification results: clear, low, medium and high. The classifier is then trained using the dataset from step S2.

[0067] S5: The report generation and visualization platform analyzes the spatial distribution and temporal evolution characteristics of the suspended sediment identification results and presents them to water environment management departments in the form of charts, reports, and interactive web maps.

[0068] Furthermore, step S1 specifically includes:

[0069] S101: Satellite remote sensing data acquisition unit 11 interfaces with the GEE cloud platform API to automatically acquire multispectral satellite images of the area;

[0070] S102: Image correction unit 21 performs radiometric calibration, atmospheric correction and geometric fine correction on the original multispectral satellite image in sequence to eliminate sensor and atmospheric interference;

[0071] S103: Image fusion unit 22 uses a pixel-level fusion algorithm to fuse multispectral satellite images processed in step S102 at multiple times to solve the problems of cloud coverage and data loss, and generate high-quality seamless reflectance images;

[0072] S104: Image reflectance extraction unit 23 performs water body masking and partitioning based on water body index and tributary vector boundaries, and finally extracts independent multispectral reflectance data of the main stream and each tributary.

[0073] Furthermore, the spatiotemporal registration rules described in step S2 are as follows: (a) Location registration: Match the latitude and longitude coordinates corresponding to the measured turbidity values ​​from the monitoring station and ship with the image reflectance data from the GEE platform; (b) Time registration: If there are measured turbidity values ​​on a certain day but no satellite image reflectance data, then select the closest image reflectance data within ±7 days to pair with the measured turbidity values; if there are multiple measured turbidity values ​​on the same day, then select the measured turbidity value closest to 12:00 to reduce the impact of daily variations.

[0074] Furthermore, the working principle of the main stream turbidity inversion model described in step S3 is as follows:

[0075] S3011: Extract the blue, green, red, near-infrared, and short-wave infrared frequencies from the paired data of reflectance and turbidity according to the frequency of reflectance;

[0076] S3012: Logarithmic transformation was performed on the measured turbidity values ​​in the paired data of reflectance and turbidity;

[0077] S3013: Perform One-Hot encoding on hydrological period information;

[0078] S3014: Concatenate the data from steps S3011 to S3013 and use the results as input to the support vector regression model;

[0079] S3015: Set the hyperparameter search space (C, epsilon, kernel, gamma) for the support vector regression model. Use the dataset from step S2 to find the optimal parameter combination in the parameter space through grid search. Construct a high-performance, highly generalizable main stream turbidity inversion model to achieve remote sensing inversion of turbidity in river and reservoir main streams.

[0080] Furthermore, the specific working principle of the tributary turbidity inversion model described in step S3 is as follows:

[0081] S3021: Extract the blue, green, red, near-infrared, and short-wave infrared frequencies from the paired data of reflectance and turbidity according to the frequency of reflectance;

[0082] S3022: The paired data of reflectance and turbidity are divided according to the discharge period, flood period, water storage period, and high water level operation period;

[0083] S3023: Using a multivariate empirical formula model, establish the correlation between measured turbidity values ​​and blue light. Green light Red light Near-infrared and shortwave infrared Quantitative relationship of reflectivity across different wavebands;

[0084] Specifically,

[0085] Discharge period: ;

[0086] Flood season: ;

[0087] Water storage period: ;

[0088] High water level operation period: ;

[0089] in, , The correlation coefficient is the coefficient to be trained.

[0090] In particular, when analyzing the optical characteristics of the main stream of rivers and reservoirs, it is necessary to consider the response characteristics of phytoplankton, suspended particulate matter, and yellow substances to remote sensing reflectance in different bands. Because the main stream environment is relatively complex, multiple spectral bands of reflectance that are sensitive to changes in turbidity are selected as independent variables, and the optimal combination is chosen as the independent variable. Five key bands, including blue light, green light, red light, near-infrared and short-wave infrared, are selected.

[0091] Furthermore, the classifier described in step S4 is a support vector machine.

[0092] After conducting programming simulation experiments on the GEE platform, the following results were obtained: Figure 3 The image shows the remote sensing identification results of suspended sediment in the main stream and tributaries of the Three Gorges Reservoir.

[0093] Example 3: For the scenario in Example 1, Figure 4 A schematic diagram of an electronic device 90 that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers.

[0094] Electronic devices can also refer to various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the invention described and / or claimed herein.

[0095] like Figure 4 As shown, the electronic device 90 includes at least one processor 91 and a memory, such as a read-only memory (ROM) 92 or a random access memory (RAM) 93, communicatively connected to the at least one processor 91. The memory stores computer programs executable by the at least one processor. The processor 91 can perform various appropriate actions and processes based on the computer program stored in the ROM 92 or loaded into the RAM 93 from storage unit 98. The RAM 93 can also store various programs and data required for the operation of the electronic device 90. The processor 91, ROM 92, and RAM 93 are interconnected via a bus 94. An input / output (I / O) interface 95 is also connected to the bus 94.

[0096] Multiple components in electronic device 90 are connected to I / O interface 95, including: input unit 96, such as keyboard, mouse, etc.; output unit 97, such as various types of displays, speakers, etc.; storage unit 98, such as disk, optical disk, etc.; and communication unit 99, such as network card, modem, wireless transceiver, etc. Communication unit 99 allows electronic device 90 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0097] Processor 91 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 91 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any suitable processor, controller, microcontroller, etc. Processor 91 performs the various methods and processes described above, such as the remote sensing identification method for suspended sediment in rivers and reservoirs based on water color differences.

[0098] In some embodiments, the remote sensing identification method for suspended sediment in rivers and reservoirs based on water color differences can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 98. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 90 via ROM 92 and / or communication unit 99. When the computer program is loaded into RAM 93 and executed by processor 91, one or more steps of the remote sensing identification method for suspended sediment in rivers and reservoirs based on water color differences described above can be performed. Alternatively, in other embodiments, processor 91 can be configured to perform the remote sensing identification method for suspended sediment in rivers and reservoirs based on water color differences by any other suitable means (e.g., by means of firmware).

[0099] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0100] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0101] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0102] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0103] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0104] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0105] Finally, it should be noted that the above preferred embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail through the above preferred embodiments, those skilled in the art should understand that various changes can be made to it in form and detail without departing from the scope defined by the claims of the present invention.

Claims

1. A remote sensing identification system for suspended sediment in rivers and reservoirs based on water color differences, characterized in that, It consists of a data acquisition module (1), a data preprocessing and fusion module (2), a spatiotemporal registration engine (3), a turbidity inversion module (4), a suspended sediment identification and classification module (5), and a report generation and visualization platform (6) connected in sequence; The data acquisition module includes a satellite remote sensing data acquisition unit (11) and an in-situ turbidity data acquisition unit (12). The satellite remote sensing data acquisition unit (11) is connected to the Google Earth Engine (GEE) cloud platform API and automatically searches and downloads multispectral satellite images according to the spatiotemporal range. The in-situ turbidity data acquisition unit (12) receives and manages in-situ observation turbidity data from monitoring stations and ship surveys to obtain measured turbidity values. The data preprocessing and fusion module (2) includes an image correction unit (21), an image fusion unit (22), and an image reflectance extraction unit (23). The image correction unit (21) is connected to the satellite remote sensing data acquisition unit (11) and automatically performs radiometric calibration, atmospheric correction, and geometric fine correction on multispectral satellite images. The image fusion unit (22) is connected to the image correction unit (21) and uses a pixel-level fusion algorithm to fuse high spatiotemporal resolution images to solve the problems of cloud coverage and data loss. The image reflectance extraction unit (23) is connected to the image fusion unit (22) and automatically segments and extracts reflectance data from the main stream area and each tributary area. The spatiotemporal registration engine (3) is connected to the output of the in-situ turbidity data acquisition unit (12) and the output of the image reflectance extraction unit (23) respectively, and automatically matches the position and time of the in-situ measurement point with the time of the multispectral satellite image to construct paired data of reflectance and turbidity under the same spatiotemporal conditions. The turbidity inversion module (4) includes a main stream turbidity inversion unit (41) and a tributary turbidity inversion unit (42). The main stream turbidity inversion unit (41) is connected to the output of the spatiotemporal registration engine (3). It adopts a support vector regression (SVR) machine learning model to establish a machine learning relationship model between remote sensing reflectance and measured turbidity, thereby realizing remote sensing inversion of turbidity in the main stream of the river and reservoir. The tributary turbidity inversion unit (42) is connected to the output of the spatiotemporal registration engine (3). According to the reservoir operation cycle, it calculates the correlation coefficient between turbidity and reflectance of each band, establishes an empirical formula model for different cycles, and realizes remote sensing inversion of turbidity in the tributary of the river and reservoir. The suspended sediment identification and classification module (5) is a classifier connected to the output of the main stream turbidity inversion unit (41) and the tributary turbidity inversion unit (42). According to the set turbidity threshold, the turbidity inversion results of the main stream and tributary are uniformly divided into four levels: clear, low, medium and high, and the "high turbidity" water body is automatically identified as "suspended sediment water body". The report generation and visualization platform (6) is connected to the output of the suspended sediment identification and classification unit (5), and displays the identification results of suspended sediment to the water environment management department in the form of charts, reports and interactive web maps.

2. A remote sensing identification method for suspended sediment in rivers and reservoirs based on water color differences, characterized in that, Includes the following steps: S1: Data Acquisition and Preprocessing: Based on the GEE platform, multispectral satellite images are acquired, and after correction, fusion, and segmentation, multispectral reflectance data of the main stream area and tributary areas of rivers and reservoirs are extracted; S2: Dataset Construction: Spatiotemporal registration will be performed between in-situ measured turbidity data and multispectral reflectance data from multispectral satellite images; based on historical reflectance and turbidity pairing data, combined with human annotations of four levels (clear, low, medium, and high), a dataset for model training and validation will be constructed. S3: Turbidity Inversion Model Construction and Training: Turbidity inversion models are established for the different optical characteristics of the main stream and tributaries: the turbidity inversion model of the main stream adopts the support vector regression model, and the turbidity inversion model of the tributaries is based on the empirical formula model constructed on the sensitive band; then the turbidity inversion model is trained using the dataset from step S2. S4: Suspended sediment identification and analysis: The turbidity of the main stream and tributaries predicted by the turbidity inversion model is input into the classifier and classified into four levels of suspended sediment: clear, low, medium and high. The classifier is then trained using the dataset from step S2. S5: Analyze the spatial distribution and temporal evolution characteristics of the suspended sediment identification results and present them to the water environment management department in the form of charts, reports and interactive web maps.

3. The remote sensing identification method for suspended sediment in rivers and reservoirs based on water color differences according to claim 2, characterized in that, The specific steps of S1 are as follows: S101: The satellite remote sensing data acquisition unit (11) interfaces with the GEE cloud platform API to automatically acquire multispectral satellite images within the area; S102: The image correction unit (21) performs radiometric calibration, atmospheric correction and geometric fine correction on the original multispectral satellite image in sequence to eliminate sensor and atmospheric interference; S103: The image fusion unit (22) uses a pixel-level fusion algorithm to fuse the multispectral satellite images processed in step S102 at multiple times to solve the problems of cloud coverage and data loss, and generate a high-quality seamless reflectance image; S104: Image reflectance extraction unit (23) performs water body masking and partitioning based on water index and tributary vector boundaries, and finally extracts independent multispectral reflectance data of the main stream and each tributary.

4. The remote sensing identification method for suspended sediment in rivers and reservoirs based on water color differences according to claim 2, characterized in that, The spatiotemporal registration rules described in step S2 are as follows: (a) Location registration: Match the latitude and longitude coordinates corresponding to the measured turbidity values ​​from the monitoring station and ship with the image reflectance data from the GEE platform; (b) Time registration: If there are measured turbidity values ​​on a certain day but no satellite image reflectance data, then select the closest image reflectance data within ±7 days to match the measured turbidity values; if there are multiple measured turbidity values ​​on the same day, then select the measured turbidity value closest to 12 noon to reduce the impact of daily variations.

5. The remote sensing identification method for suspended sediment in rivers and reservoirs based on water color differences according to claim 2, characterized in that, The working principle of the main stream turbidity inversion model described in step S3 is as follows: S3011: Extract the blue, green, red, near-infrared, and short-wave infrared frequencies from the paired data of reflectance and turbidity according to the frequency of reflectance; S3012: Logarithmic transformation was performed on the measured turbidity values ​​in the paired data of reflectance and turbidity; S3013: Perform One-Hot encoding on hydrological period information; S3014: Concatenate the data from steps S3011 to S3013 and use the results as input to the support vector regression model; S3015: Set up the hyperparameter search space for the support vector regression model, use the dataset from step S2 to find the optimal parameter combination in the parameter space through grid search, and construct a high-performance, highly generalizable main stream turbidity inversion model to realize remote sensing inversion of turbidity in river and reservoir main streams.

6. The remote sensing identification method for suspended sediment in rivers and reservoirs based on water color differences according to claim 2, characterized in that, The specific working principle of the tributary turbidity inversion model described in step S3 is as follows: S3021: Extract the blue, green, red, near-infrared, and short-wave infrared frequencies from the paired data of reflectance and turbidity according to the frequency of reflectance; S3022: The paired data of reflectance and turbidity are divided according to the discharge period, flood period, water storage period, and high water level operation period; S3023: Establish measured turbidity values ​​using a multivariate empirical formula model. With blue light Green light Red light Near-infrared and shortwave infrared Quantitative relationship of reflectivity across different wavebands; Specifically, Discharge period: ; Flood season: ; Water storage period: ; High water level operation period: ; in, , The correlation coefficient is the coefficient to be trained.

7. The remote sensing identification method for suspended sediment in rivers and reservoirs based on water color differences according to claim 2, characterized in that, The classifiers mentioned in step S4 include, but are not limited to, logistic regression, random forest, support vector machine, neural network, Naive Bayes classifier, and gradient boosting tree.

8. An electronic device and a computer-readable storage medium, characterized in that, The electronic device includes at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the remote sensing identification method for suspended sediment in rivers and reservoirs based on water color differences as described in any one of claims 2 to 7; the computer-readable storage medium stores computer instructions for causing the processor to implement the remote sensing identification method for suspended sediment in rivers and reservoirs based on water color differences as described in any one of claims 2 to 7 when executed.

Images