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

Abstract

The application is a river reservoir suspended sediment remote sensing recognition system and method based on water color difference, belonging to the technical field of aquatic ecological environment monitoring. First, multispectral satellite images are obtained through the GEE cloud platform, and dry and branch stream reflectivity data are extracted after preprocessing such as atmospheric correction and image fusion. Then, in-situ turbidity measurement values are spatiotemporally matched with satellite reflectivity to construct a training data set. According to the different water color characteristics of the dry and branch streams, a support vector regression machine learning model for the dry stream and a multi-hydrological period empirical model for the branch stream are respectively established to realize high-precision partition inversion of turbidity. Finally, suspended sediment water bodies are identified through turbidity grading, and a spatiotemporal distribution atlas is generated. The application effectively solves the problem of water color difference between dry and branch streams caused by reservoir regulation, significantly improves the precision and reliability of suspended sediment remote sensing monitoring, and provides technical support for river reservoir sediment control and water environment management.

Description

TECHNICAL FIELD

[0001] The present application relates to a river reservoir suspended sediment remote sensing identification system and method based on water color difference, belonging to the technical field of aquatic ecological environment monitoring, and is especially suitable for river reservoir suspended sediment remote sensing identification based on water color difference. BACKGROUND

[0002] The suspended sediment is the core parameter of determining the water body optical characteristics such as the transparency, turbidity and color of the river reservoir, and is also the important basis for studying the water environment processes such as the sediment transport, geomorphic evolution and eutrophication of the river reservoir. Mastering the spatial distribution and time evolution law of the suspended sediment can not only provide data support for the mechanism research on the spatial and temporal evolution of the suspended sediment of the river reservoir, but also provide decision basis for flood control emergency, dam operation and maintenance, and navigation scheduling.

[0003] At present, the conventional monitoring method of the suspended sediment mainly focuses on the site monitoring which can continuously and accurately obtain the suspended sediment content on the time scale. However, the site construction and operation cost is high, which leads to a small number of sites and a sparse distribution, and cannot perform spatial continuous monitoring on a large scale. More importantly, the manual water storage and release operation causes the hydrological regime of the main stream and tributaries of the river reservoir to change significantly. In addition, the "storing clean water and discharging turbid water" scheduling strategy causes the suspended sediment content in the backwater area of the main stream and tributaries of the river reservoir to fluctuate sharply, and forms completely different water color conditions in the specific space-time domain. Therefore, the existing monitoring method cannot reveal the spatial and temporal characteristics of the suspended sediment of the main stream and tributaries of the river reservoir and the response law of the suspended sediment to the environmental change, and it is urgent to establish a suspended sediment remote sensing model for the main stream and tributaries of the river reservoir to realize the spatial and temporal dynamic monitoring of the suspended sediment, so as to provide technical support for the sediment deposition control, water environment protection and navigation safety of the river reservoir. SUMMARY

[0004] The purpose of the present application is to overcome the shortcomings of the prior art, and to provide a river reservoir suspended sediment remote sensing identification system and method based on water color difference, which realizes the accurate description of the spatial and temporal evolution process of the suspended sediment by acquiring satellite multispectral image data, combining partition modeling and multi-source data fusion.

[0005] In order to achieve the above purpose, the present application provides the following technical scheme:

[0006] The river reservoir suspended sediment remote sensing identification system based on water color difference is connected with Figure 1 characterized in that it is composed of a collection module, a data preprocessing and fusion module, a space-time registration engine, a turbidity inversion module, a suspended sediment identification and grading 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 is connected with the Google Earth Engine (GEE) cloud platform API, automatically searches and downloads multi-spectral satellite images according to the space-time range; the in-situ turbidity data acquisition unit receives and manages in-situ observation turbidity data such as turbidity from monitoring stations and ship measurements to obtain turbidity measured values.

[0008] The data preprocessing and fusion module includes an image correction unit, an image fusion unit and an image reflectivity extraction unit; the image correction unit is connected with the satellite remote sensing data acquisition unit, automatically performs radiation calibration, atmospheric correction and geometric precision correction on the multi-spectral satellite images; the image fusion unit is connected with the image correction unit, adopts a pixel-level fusion algorithm, fuses high spatio-temporal resolution images, and solves the problems of cloud coverage and data loss; the image reflectivity extraction unit is connected with the image fusion unit, automatically segments and extracts reflectivity data of the dry river area and each tributary area.

[0009] The spatio-temporal registration engine is connected with the output end of the in-situ turbidity data acquisition unit and the output end of the image reflectivity extraction unit, respectively, automatically matches the position and time of the in-situ measurement point with the multi-spectral satellite image time, and constructs the paired data of reflectivity and turbidity under the same space-time.

[0010] The turbidity inversion module includes a dry river turbidity inversion unit and a tributary turbidity inversion unit; the dry river turbidity inversion unit is connected with the output end of the spatio-temporal registration engine, adopts a support vector regression (SVR) machine learning model, establishes a machine learning relationship model between the remote sensing reflectivity and the turbidity measured value, and realizes the remote sensing inversion of the turbidity of the dry river and reservoir; the tributary turbidity inversion unit is connected with the output end of the spatio-temporal registration engine, calculates the correlation coefficient of turbidity and reflectivity of each band according to the operation period of the reservoir, establishes an empirical formula model of different periods, and realizes the remote sensing inversion of the turbidity of the tributary of the river and reservoir.

[0011] The suspended sediment identification and classification module is a classifier connected with the output ends of the dry river turbidity inversion unit and the tributary turbidity inversion unit, sets turbidity thresholds, uniformly classifies the turbidity inversion results of the dry river and tributary into four levels of clear, low, medium and high, and automatically identifies the "high turbidity" water body as a "suspended sediment water body".

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

[0013] The river and reservoir suspended sediment remote sensing identification method based on water color difference combines Figure 2 , and includes the following steps:

[0014] S1: Data collection and preprocessing: Based on the GEE platform, collect multispectral satellite images, and after correction, fusion, and segmentation, extract the multispectral reflectance data of the main stream and tributary areas of rivers and reservoirs;

[0015] S2: Data set construction: Temporally and spatially register the in-situ measured turbidity data with the multispectral reflectance data of the multispectral satellite images; based on the paired data of historical reflectance and turbidity, combined with the artificial labeling of clear, low, medium, and high four levels, construct a data set for model training and verification;

[0016] S3: Turbidity inversion model construction and training: For different water optical properties of the main stream and tributaries, turbidity inversion models are established respectively: the main stream turbidity inversion model uses a support vector regression model, and the tributary turbidity inversion model is based on an empirical formula model constructed by sensitive bands; then the data set of step S2 is used to train the turbidity inversion model;

[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 to divide the identification results of suspended sediment into clear, low, medium, and high four levels, and the data set of step S2 is used to train the classifier;

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

[0019] Further, the step S1 is specifically:

[0020] S101: The satellite remote sensing data acquisition unit is connected with the GEE cloud platform API, and automatically acquires the multispectral satellite images in the region;

[0021] S102: The image correction unit sequentially performs radiation calibration, atmospheric correction, and geometric precision correction on the original multispectral satellite images to eliminate sensor and atmospheric interference;

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

[0023] S104: The image reflectance extraction unit performs water mask and partition cutting based on the water body index and the main stream and tributary vector boundary, and finally extracts the independent multispectral reflectance data of the main stream and each tributary.

[0024] Further, the spatio-temporal registration rule in step S2 is: (a) position registration: matching the latitude and longitude coordinates corresponding to the turbidity measured value of the monitoring station and the ship measurement with the image reflectivity data of the GEE platform; (b) time registration: if there is a turbidity measured value on a certain day but no satellite image reflectivity data, the closest image reflectivity data within ±7 days is selected to be paired with the turbidity measured value; if there are multiple turbidity measured values on the same day, the turbidity measured value closest to 12 o'clock noon is selected to reduce the influence of diurnal variation.

[0025] Further, the dry flow turbidity inversion model in step S3 has the following specific working principle:

[0026] S3011: According to the frequency of reflectivity, the blue light, green light, red light, near-infrared and short-wave infrared frequencies in the paired data of reflectivity and turbidity are extracted;

[0027] S3012: The turbidity measured value in the paired data of reflectivity and turbidity is logarithmically transformed;

[0028] S3013: The hydrological period information is One-Hot encoded;

[0029] S3014: The data of steps S3011-S3013 are concatenated as the input of the support vector regression model;

[0030] S3015: The hyperparameter search space of the support vector regression model is set, and the data set in step S2 is used to optimize the parameter space through grid search to determine the optimal parameter combination, thereby constructing a dry flow turbidity inversion model with excellent performance and strong generalization ability, and realizing remote sensing inversion of the turbidity of the dry flow of the river and reservoir.

[0031] Further, the tributary turbidity inversion model in step S3 has the following specific working principle:

[0032] S3021: According to the frequency of reflectivity, the blue light, green light, red light, near-infrared and short-wave infrared frequencies in the paired data of reflectivity and turbidity are extracted;

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

[0034] S3023: A quantitative relationship between the turbidity measured value (NT) and the reflectivity of each waveband of blue light (Rb), green light (Rg), red light (Rr), near-infrared (Rnir) and short-wave infrared (Rswir) is established by using a multivariate empirical formula model;

[0035] ​​​​​​Specifically,

[0036] Flood period: ;

[0037] Flood period: ;

[0038] Flood period: ;

[0039] Flood period: ;

[0040] Wherein, 、 The correlation coefficient to be trained.

[0041] In particular, the analysis of the optical characteristics of the river reservoir main stream water body needs to consider the response characteristics of phytoplankton, suspended particles and yellow substances to different waveband remote sensing reflectivity; Because the main stream environment is relatively complex, multiple spectral waveband reflectivity sensitive to turbidity change is selected as the independent variable, and the optimal combination is selected as the independent variable, and then five key wavebands including blue light, green light, red light, near infrared and short wave infrared are selected.

[0042] Further, the classifier of 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, comprising at least one processor; and a memory connected with the at least one processor in communication; wherein,

[0044] The memory stores a computer program executed by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the above-mentioned river reservoir suspended sediment remote sensing identification method based on water color difference.

[0045] Finally, the present application also discloses a computer readable storage medium, the computer readable storage medium stores computer instructions, the computer instructions are used to make the processor execute to realize the above-mentioned river reservoir suspended sediment remote sensing identification method based on water color difference.

[0046] The beneficial effects of the present application are: a river reservoir suspended sediment remote sensing identification system and method based on water color difference are provided, by constructing a river reservoir suspended sediment remote sensing identification system based on water color difference, high-precision and automatic monitoring of the spatial and temporal distribution of river reservoir suspended sediment is realized; the system innovatively adopts a dry and branch stream partition inversion strategy, a support vector regression machine learning model is used for the dry stream complex optical characteristics, and a multi-mode empirical model is established for the branch stream in different hydrological periods, effectively solving the problem of water color difference between the dry and branch streams caused by reservoir regulation. Through multi-source remote sensing data fusion and spatio-temporal registration technology, the turbidity inversion precision and data availability are significantly improved, and finally the generated suspended sediment classification atlas and spatio-temporal evolution report provide reliable decision support for river reservoir sediment deposition control, water environment protection and navigation safety. BRIEF DESCRIPTION OF DRAWINGS

[0047] In order to make the purpose, technical scheme of the present application clearer, the present application provides the following drawings for description:

[0048] Figure 1 It is the architecture diagram of the river reservoir suspended sediment remote sensing identification system based on water color difference in the present application; the arrow is the data transmission direction;

[0049] Figure 2 It is the flow chart of the river reservoir suspended sediment remote sensing identification method based on water color difference in the present application;

[0050] Figure 3 It is the remote sensing identification result of the suspended sediment of the dry and branch streams of the Three Gorges Reservoir in the embodiment 2 of the present application;

[0051] Figure 4 It is the structural schematic diagram of the electronic equipment in the embodiment 3 of the present application. DETAILED DESCRIPTION

[0052] In order to make the purpose and technical scheme of the present application clearer, the present application is described in detail below in combination with the drawings and embodiments.

[0053] Embodiment 1: The suspended sediment of the typical branch streams of the Three Gorges Reservoir, i.e. Xiaojiang River, Caotang River, Daning River, Xiangxi River and the suspended sediment connected with the dry stream, is identified and estimated, so as to provide reliable decision support for river reservoir sediment deposition control, water environment protection and navigation safety. The present application proposes a "river reservoir suspended sediment remote sensing identification system based on water color difference".

[0054] In combination with Figure 1 , the system is connected in sequence by a collection module 1, a data preprocessing and fusion module 2, a spatio-temporal registration engine 3, a turbidity inversion module 4, a suspended sediment identification and classification module 5, and a report generation and visualization platform 6.

[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 is connected with the GEE cloud platform API, automatically searches and downloads multi-spectral satellite images such as Landsat8 / 9 OLI and Sentinel-2 according to the space-time 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 measurements to obtain turbidity measured values.

[0056] The data preprocessing and fusion module 2 includes an image correction unit 21, an image fusion unit 22 and an image reflectivity extraction unit 23; the image correction unit 21 is connected with the satellite remote sensing data acquisition unit 11, automatically performs radiation calibration, atmospheric correction and geometric precision correction on the multi-spectral satellite images; the image fusion unit 22 is connected with the image correction unit 21, adopts a pixel-level fusion algorithm to fuse high spatio-temporal resolution images and solve the problems of cloud coverage and data loss; the image reflectivity extraction unit 23 is connected with the image fusion unit 22, automatically segments and extracts reflectivity data of the dry river area and each tributary area.

[0057] The spatio-temporal registration engine 3 is connected with the output end of the in-situ turbidity data acquisition unit 12 and the output end of the image reflectivity extraction unit 23, automatically matches the position and time of the in-situ measurement point with the multi-spectral satellite image time, and constructs paired data of reflectivity and turbidity under the same space-time.

[0058] The turbidity inversion module 4 includes a dry river turbidity inversion unit 41 and a tributary turbidity inversion unit 42; the dry river turbidity inversion unit 41 is connected with the output end of the spatio-temporal registration engine 3, adopts a support vector regression machine learning model, establishes a machine learning relationship model between the remote sensing reflectivity and the turbidity measured value, and realizes remote sensing inversion of the turbidity of the dry river and reservoir; the tributary turbidity inversion unit 42 is connected with the output end of the spatio-temporal registration engine 3, calculates the correlation coefficient of turbidity and reflectivity of each band according to the operation period of the reservoir, establishes an empirical formula model of different periods, and realizes remote sensing inversion of the turbidity of the tributary of the river and reservoir.

[0059] The suspended sediment identification and classification module 5 is a classifier connected with the output ends of the dry river turbidity inversion unit 41 and the tributary turbidity inversion unit 42, sets turbidity thresholds, uniformly divides the turbidity inversion results of the dry river and tributary into four levels of clear, low, medium and high, and automatically identifies the “high turbidity” water body as a “suspended sediment water body”.

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

[0061] For the scenario of embodiment 1, the application also proposes a "river reservoir suspended sediment remote sensing identification method based on water color difference".

[0062] In combination Figure 2 , the method comprises the following steps:

[0063] S1: Data acquisition and preprocessing: the satellite remote sensing data acquisition unit 11 acquires multispectral satellite images based on the GEE platform, and after correction, fusion and segmentation by the data preprocessing and fusion module 2, extracts the multispectral reflectance data of the main stream area and each tributary area of the Three Gorges Reservoir;

[0064] S2: Data set construction: the spatio-temporal registration engine 3 performs spatio-temporal registration on the turbidity data measured in situ and the multispectral reflectance data of the multispectral satellite images; based on the paired data of historical reflectance and turbidity, combined with the artificial labeling of clear, low, medium and high four levels, a data set for model training and verification is constructed according to 7:3;

[0065] S3: Turbidity inversion model construction and training: turbidity inversion models are established for the different water optical properties of the main stream and tributaries: the main stream turbidity inversion model of the main stream turbidity inversion unit 41 adopts a support vector regression model, and the tributary turbidity inversion model of the tributary turbidity inversion unit 42 is based on an empirical formula model constructed by sensitive bands; then the turbidity inversion model is trained using the data set of 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 to divide the identification results of the suspended sediment into clear, low, medium and high four levels, and the classifier is trained using the data set of step S2;

[0067] S5: The report generation and visualization platform 6 analyzes the spatial distribution and temporal evolution characteristics of the identification results of the suspended sediment, and displays them in the form of charts, reports and interactive Web maps to the water environment management department.

[0068] Further, the step S1 is specifically:

[0069] S101: The satellite remote sensing data acquisition unit 11 is connected with the GEE cloud platform API, and automatically acquires multispectral satellite images in the region;

[0070] S102: The image correction unit 21 sequentially performs radiation calibration, atmospheric correction and geometric precision correction on the original multispectral satellite images to eliminate sensor and atmospheric interference;

[0071] S103: The image fusion unit 22 adopts a pixel-level fusion algorithm to fuse the multi-spectral satellite images at multiple moments processed in step S102 to solve the cloud coverage and data missing problem, and generate high-quality seamless reflectivity images;

[0072] S104: The image reflectivity extraction unit 23 performs water mask and partition cutting based on the water body index and the dry tributary vector boundary, and finally extracts the multi-spectral reflectivity data of the dry stream and each tributary.

[0073] Further, the spatio-temporal registration rule in step S2 is: (a) position registration: matching the latitude and longitude coordinates corresponding to the turbidity measured value of the monitoring station and the ship measurement with the image reflectivity data of the GEE platform; (b) time registration: if there is a turbidity measured value on a certain day but no satellite image reflectivity data, select the closest image reflectivity data within ±7 days to pair with the turbidity measured value; if there are multiple turbidity measured values on the same day, select the turbidity measured value closest to 12:00 to reduce the influence of diurnal variation.

[0074] Further, the dry stream turbidity inversion model described in step S3 has the following specific working principle:

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

[0076] S3012: Perform logarithmic transformation on the turbidity measured value in the paired data of reflectivity and turbidity;

[0077] S3013: One-Hot encode the hydrological period information;

[0078] S3014: Concatenate the data of steps S3011-S3013 as the input of the support vector regression model;

[0079] S3015: Set the hyperparameter search space (C, epsilon, kernel, gamma) of the support vector regression model, use the data set of step S2 to optimize in the parameter space through grid search, determine the optimal parameter combination, and build a dry stream turbidity inversion model with excellent performance and strong generalization ability, realizing remote sensing inversion of river reservoir dry stream turbidity.

[0080] Further, the tributary turbidity inversion model described in step S3 has the following specific working principle:

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

[0082] S3022: According to the period of water release, flood period, water storage period, high water level operation period, the paired data of reflectivity and turbidity are divided;

[0083] S3023: Using a multivariate empirical formula model, the quantitative relationship between the measured value of turbidity and the reflectivity of blue light , green light , red light , near infrared and short wave infrared of each wave band is established.

[0084] Specifically,

[0085] Water release period: ;

[0086] Flood period: ;

[0087] Water storage period: ;

[0088] High water level operation period: ;

[0089] Among them, , are the correlation coefficients to be trained.

[0090] In particular, in order to analyze the optical characteristics of the river reservoir main stream water body, the response characteristics of phytoplankton, suspended particles and yellow substances to different wave band remote sensing reflectivity need to be considered; because the main stream environment is relatively complex, the multiple spectral wave band reflectivity sensitive to turbidity change is selected as the independent variable, and the optimal combination is selected as the independent variable, and then five key wave bands including blue light, green light, red light, near infrared and short wave infrared are selected.

[0091] Further, the classifier of step S4 is a support vector machine.

[0092] After programming simulation experiment on GEE platform, the remote sensing recognition result of suspended sediment in the main stream and branch stream of the Three Gorges Reservoir is obtained as shown in Figure 3 .

[0093] Embodiment 3: For the scene of embodiment 1, Figure 4 A structural schematic diagram of an electronic device 90 that can be used to implement the embodiments of the present application is shown. The electronic device is intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate 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 river reservoir suspended sediment remote sensing identification method based on water color difference can be implemented as a computer program tangibly embodied in a computer readable storage medium, e.g., storage unit 98. In some embodiments, parts or all of the computer program can be loaded and / or installed onto electronic device 90 via, e.g., ROM 92 and / or communication unit 99. When the computer program is loaded onto RAM 93 and executed by processor 91, one or more steps of the river reservoir suspended sediment remote sensing identification method based on water color difference described above can be performed. Alternatively, in other embodiments, processor 91 can be configured to perform the river reservoir suspended sediment remote sensing identification method based on water color difference by way of other means (e.g., by way of firmware).

[0099] Various implementations of the systems and techniques described above can be realized in digital electronic circuitry, integrated circuitry, specially designed application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0100] Computer programs used to implement the methods of the application can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the computer program, when executed by the processor of the machine, implements the functions / acts specified in the flowcharts and / or block diagrams. The computer program can be executed entirely on a machine, partially on a machine, partially on a machine and partially on a remote machine or entirely on a remote machine or server.

[0101] In the context of the present application, a computer-readable storage medium can be a tangible medium that can contain or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. A computer-readable storage medium can include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium can be a machine-readable signal medium. More specific examples of a machine-readable storage medium will include one or more lines of a program of instructions in a transitory signal, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0102] To provide for interaction with a user, the systems and techniques described here 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 a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the electronic device. Other kinds of devices can be used to provide for interaction with a user as well; 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 acoustic, speech, or tactile input.

[0103] The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a user computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0104] The computing system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system, to solve the defects of large management difficulty and weak business scalability in traditional physical host and VPS service.

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

Claims

1. A river reservoir suspended sediment remote sensing identification system based on water color difference, characterized in that, The collecting module (1), the data preprocessing and fusion module (2), the space-time registration engine (3), the turbidity inversion module (4), the suspended sediment identification and classification module (5) and the report generation and visualization platform (6) are sequentially connected to constitute; The collecting module (1) includes a satellite remote sensing data collecting unit (11) and an in-situ turbidity data collecting unit (12); the satellite remote sensing data collecting unit (11) is connected with a Google Earth Engine (GEE) cloud platform API, automatically searches and downloads multispectral satellite images according to a space-time range; the in-situ turbidity data collecting unit (12) receives and manages in-situ observation turbidity data such as turbidity from monitoring stations and ship measurements to obtain turbidity measured values; The data preprocessing and fusion module (2) includes an image correction unit (21), an image fusion unit (22) and an image reflectivity extraction unit (23); the image correction unit (21) is connected with the satellite remote sensing data collecting unit (11) and automatically performs radiation calibration, atmospheric correction and geometric precision correction on the multispectral satellite images; the image fusion unit (22) is connected with the image correction unit (21) and adopts a pixel-level fusion algorithm to fuse high space-time resolution images and solve cloud coverage and data missing problems; the image reflectivity extraction unit (23) is connected with the image fusion unit (22) and automatically extracts reflectivity data of dry river areas and tributary areas; The space-time registration engine (3) is connected with output ends of the in-situ turbidity data collecting unit (12) and the image reflectivity extraction unit (23), automatically matches positions and times of in-situ measurement points with a multispectral satellite image time to construct paired data of reflectivity and turbidity under the same space-time; The turbidity inversion module (4) includes a dry river turbidity inversion unit (41) and a tributary turbidity inversion unit (42); the dry river turbidity inversion unit (41) is connected with an output end of the space-time registration engine (3), adopts a support vector regression (SVR) machine learning model, establishes a machine learning relationship model between remote sensing reflectivity and turbidity measured values, and realizes remote sensing inversion of dry river turbidity; the tributary turbidity inversion unit (42) is connected with the output end of the space-time registration engine (3), calculates correlation coefficients of turbidity and reflectivity of each wave band according to a reservoir operation period, establishes an empirical formula model of different periods, and realizes remote sensing inversion of tributary turbidity of a river reservoir; The suspended sediment identification and classification module (5) is a classifier connected with output ends of the dry river turbidity inversion unit (41) and the tributary turbidity inversion unit (42), sets turbidity threshold values, uniformly divides turbidity inversion results of the dry river and the tributary into four levels of clear, low, medium and high, and automatically identifies "high turbidity" water bodies as "suspended sediment water bodies"; The report generation and visualization platform (6) is connected with an output end of the suspended sediment identification and classification module (5), displays identification results of the suspended sediment in the form of charts, reports and interactive Web maps to a water environment management department.

2. The method for remote sensing identification of river reservoir suspended sediment based on water color difference, characterized in that, The method comprises the following steps: S1: Data collection and preprocessing: Based on the GEE platform, multispectral satellite images are collected, and after correction, fusion, segmentation, the multispectral reflectance data of the main stream area and each tributary area of the river reservoir are extracted; S2: Data set construction: The in-situ measured turbidity data are spatio-temporally registered with the multispectral reflectance data of the multispectral satellite images; based on the paired data of historical reflectance and turbidity, combined with the human annotation of clear, low, medium and high four levels, a data set for model training and verification is constructed; S3: Turbidity inversion model construction and training: According to the different water optical properties of the main stream and the tributary, turbidity inversion models are established respectively: the main stream turbidity inversion model adopts the support vector regression model, and the tributary turbidity inversion model is based on the sensitive band to construct the empirical formula model; then the turbidity inversion model is trained by using the data set of step S2; S4: Suspended sediment identification and analysis: the turbidity of the main stream and the tributary predicted by the turbidity inversion model is input into the classifier to divide the identification results of the suspended sediment into clear, low, medium and high four levels, and the classifier is trained by using the data set of step S2; S5: Analyze the spatial distribution and temporal evolution characteristics of the identification results of the suspended sediment, and display them to the water environment management department in the form of charts, reports and interactive Web maps.

3. The water color difference based river reservoir suspended sediment remote sensing identification method according to claim 2, characterized in that, The step S1 is specifically: S101: The satellite remote sensing data collection unit (11) is connected with the GEE cloud platform API, and the multispectral satellite images in the region are automatically obtained; S102: The image correction unit (21) sequentially performs radiation calibration, atmospheric correction and geometric precision correction on the original multispectral satellite images to eliminate sensor and atmospheric interference; S103: The image fusion unit (22) uses a pixel-level fusion algorithm to fuse the multispectral satellite images at multiple times processed in step S102 to solve the problems of cloud coverage and data loss, and generates high-quality seamless reflectance images; S104: The image reflectance extraction unit (23) performs water mask and partition cutting based on the water body index and the main stream and tributary vector boundary, and finally extracts the independent multispectral reflectance data of the main stream and each tributary.

4. The water color difference based river reservoir suspended sediment remote sensing identification method according to claim 2, characterized in that, The spatio-temporal registration rule of step S2 is: (a) position registration: according to the latitude and longitude coordinates corresponding to the turbidity measured value of the monitoring station and the ship measurement, the image reflectance data of the GEE platform are matched; (b) time registration: if there is a turbidity measured value on a certain day but no satellite image reflectance data, the closest image reflectance data within ±7 days is selected to be paired with the turbidity measured value; if there are multiple turbidity measured values on the same day, the turbidity measured value closest to 12 o'clock noon is selected to reduce the influence of diurnal variation.

5. The water color difference based river reservoir suspended sediment remote sensing identification method according to claim 2, characterized in that, The working principle of the main stream turbidity inversion model in step S3 is specifically: S3011: The blue light, green light, red light, near-infrared and short-wave infrared frequencies in the paired data of reflectance and turbidity are extracted according to the frequency of reflectance; S3012: The turbidity measured value in the paired data of reflectance and turbidity is logarithmically transformed; S3013: The hydrological period information is One-Hot encoded; S3014: Concatenate the data of steps S3011-S3013 as the input of the support vector regression model; S3015: Set the hyperparameter search space of the support vector regression model, and determine the optimal parameter combination by grid search in the parameter space using the data set of step S2, so as to construct a dry flow turbidity inversion model with excellent performance and strong generalization ability, and realize remote sensing inversion of the turbidity of the dry flow of the river reservoir.

6. The water color difference based river reservoir suspended sediment remote sensing identification method according to claim 2, characterized in that, The branch flow turbidity inversion model described in step S3 has the following specific working principle: S3021: Extract the blue light, green light, red light, near-infrared and short-wave infrared frequencies in the paired data of reflectivity and turbidity according to the frequency of reflectivity; S3022: Divide the paired data of reflectivity and turbidity according to the drainage period, flood period, storage period and high water level operation period; S3023: Establishing turbidity measured value by using multivariate empirical formula model with blue light , green light , red light , near infrared and short wave infrared reflectance of each wave band; Specifically, spill period: ​ Flood period: ; Filling period: ; High water level operation period: ; wherein , is the correlation coefficient to be trained.

7. The water color difference based river reservoir suspended sediment remote sensing identification method according to claim 2, characterized in that, 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.

8. An electronic device, comprising: The electronic device includes at least one processor; and a memory connected in communication with the at least one processor; wherein the memory stores a computer program executed by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the river reservoir suspended sediment remote sensing identification method based on water color difference according to any one of claims 2-7.

9. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer instructions for enabling the processor to execute the river reservoir suspended sediment remote sensing identification method based on water color difference according to any one of claims 2-7.

Images