Method and system for inverting concentration of suspended solids in complex water body through remote sensing

Abstract

The invention discloses a method and system for inverting the concentration of suspended solids in a complex water body through remote sensing, and belongs to the technical field of ecological environments. Comprising the following steps: integrating a remote sensing image and measured data to obtain a synchronous sampling point pair; analyzing the backscattering coefficient of the particulate matters, and determining the backscattering coefficient of the optimal wavelength position for reflecting the total suspended matter concentration through an iteration method; analyzing the particulate matter absorption coefficient, and determining the absorption coefficient of the optimal wavelength position for reflecting the total suspended matter concentration through an iteration method; according to the proportional relation between inorganic particles and organic particles in the water body, determining the weight of the absorption and scattering coefficient for estimating the total suspended matter concentration; and constructing a remote sensing estimation model of the total suspended matter concentration in the complex water body and applying the model to a remote sensing image. According to the method, the difference of the organic particulate matters and the inorganic particulate matters in the aspects of absorption and scattering characteristics is considered, and the method is suitable for a water body mainly containing algae, a water body mainly containing non-algae and a complex water body environment in which the water bodies and the non-algae are mixed, and has relatively high adaptability and accuracy.

Description

Technical Field

[0001] The present invention belongs to the field of ecological environment, and in particular relates to a method and system for remote sensing inversion of suspended matter concentration in complex water bodies. Background Art

[0002] Suspended matter concentration is one of the important parameters reflecting the environmental status of water bodies, and can effectively indicate the cleanliness, nutrient status and ecosystem health level of water bodies. Traditional monitoring methods for suspended matter concentration mainly rely on on-site sampling and laboratory analysis, which have limitations such as time-consuming, high cost and limited spatial coverage. Remote sensing technology has become an important means of water environment monitoring due to its advantages of large-scale, rapid and dynamic monitoring. However, in practical applications, complex water bodies contain multiple components such as inorganic particles, organic particles, and plankton. These components have significant differences in optical properties, which manifest as different absorption and scattering behaviors, resulting in extremely complex remote sensing reflection signals of water bodies. Inorganic particles (such as mineral sediments) mainly cause strong scattering, while organic particles (such as phytoplankton debris and organic debris) have higher absorption characteristics. This difference in particle composition poses the following major challenges to remote sensing inversion: 1. Insufficient model applicability: Most existing inversion models rely on the optical properties of a single type of water body and are difficult to adapt to complex water environments where multiple types of particulate matter coexist.

[0003] 2. Insufficient utilization of characteristic information: Some methods only consider the change in reflectivity and ignore the combined influence of absorption and scattering characteristics, resulting in systematic errors in the inversion results.

[0004] 3. Poor stability and generalization ability: When the water type and particle composition change, the model performance decreases significantly, and it is difficult to ensure the inversion accuracy in different regions or at different time scales.

[0005] Therefore, there is an urgent need for a remote sensing inversion method and system for suspended matter concentration that can comprehensively consider the absorption and scattering characteristics of particulate matter and adapt to changes in complex water environments, so as to improve the accuracy and applicability of the inversion results. Summary of the Invention

[0006] In order to solve the technical problems existing in the above-mentioned background technology, the present invention provides a method and system for remote sensing inversion of suspended matter concentration in complex water bodies.

[0007] The present invention is implemented by the following technical solution: a method for remote sensing inversion of suspended matter concentration in complex water bodies, comprising the following steps: Acquire and integrate remote sensing images and ground observation data to obtain synchronized sample pairs; Using synchronized sample pairs, the backscatter coefficient of particles is obtained based on water analysis. and absorption coefficient , and respectively determine the backscattering coefficient at the optimal wavelength position reflecting the total suspended matter concentration and the absorption coefficient at the optimal wavelength position; Characterizing Inorganic Suspended Matter Components Using Backscattering Coefficients at Optimal Wavelengths ; Use the absorption coefficient at the optimal wavelength position to characterize the components of organic suspended matter ; Calculate the ratio of inorganic particles to organic particles in water , based on the proportional relationship Determine the backscatter coefficient and absorption coefficient Importance in estimating total suspended matter concentrations; Comprehensive characterization of inorganic suspended matter components , Characterization of organic suspended matter components The importance of constructing a remote sensing estimation model for total suspended matter concentration in complex water bodies , using the remote sensing estimation model of total suspended solids concentration in the complex water body Output the suspended matter concentration in water bodies and apply it to remote sensing images.

[0008] In a further embodiment, the ground observation data includes at least remote sensing reflectance and total suspended matter concentration (TSM); the synchronous sample point pairs are obtained as follows: Pre-set synchronization time intervals for remote sensing images and ground observation data , according to the synchronization time interval Collect data; perform the following steps on the collected data: Define the error function , used to represent the difference between the spectral value of the remote sensing image window and the remote sensing reflectance; Using the difference minimization strategy, the pixel points that best match the remote sensing reflectance are searched in the remote sensing image window, and the spatial coordinates, spectral values, remote sensing reflectance, and suspended matter concentration of the pixel points are matched to form synchronized sample point pairs containing spatiotemporal matching information.

[0009] In a further embodiment, the backscatter coefficient of the particle is The parsing process is as follows: The total backscatter coefficient is calculated using remote sensing reflectivity and water body characteristics ; Get pure water at wavelength The backscatter coefficient , the backscattering coefficient of particles is calculated using the following formula: : .

[0010] In a further embodiment, the absorption coefficient The parsing process is as follows: Introducing a second wavelength and the third wavelength , the second wavelength , the third wavelength With the first wavelength Satisfy a pre-set length relationship between each other; The absorption coefficient is calculated using the following formula: : ; Where, 、 and The first wavelength , second wavelength and the third wavelength Remote sensing reflectivity, 、 and Pure water at the first wavelength , second wavelength and the third wavelength The absorption coefficient of .

[0011] In a further embodiment, the backscattering coefficient at the optimal wavelength position is determined as follows: The backscatter coefficient Fitting with the total suspended matter concentration TSM, the backscattering coefficient is calculated wave by wave within 400nm~900nm by an iterative method The fitting accuracy of the total suspended matter concentration TSM is calculated, and the curve of fitting accuracy versus wavelength is obtained, and the wavelength position with the highest fitting accuracy is finally determined as the optimal wavelength of the backscattering coefficient. , which is the backscattering coefficient at the optimal wavelength position; Correspondingly, the inorganic suspended matter components Characterized by: , The optimal wavelength The corresponding backscattering coefficient.

[0012] In a further embodiment, the absorption coefficient at the optimal wavelength position is determined as follows: The absorption coefficient Fitting with the total suspended matter concentration TSM, the absorption coefficient is calculated wave by wave within 400nm~900nm by an iterative method The fitting accuracy of the total suspended matter concentration TSM is finally determined, and the wavelength position with the highest fitting accuracy is taken as the optimal first wavelength of the absorption coefficient. , optimal second wavelength and the optimal third wavelength ; Correspondingly, the organic suspended matter components Characterized by: , The optimal first wavelength The corresponding absorption coefficient.

[0013] In a further embodiment, the importance of the backscatter coefficient is assigned to the scatter channel weights , the importance of the absorption coefficient gives the absorption channel a weight ; The importance determination process in the estimation of total suspended solids concentration is as follows: Determine the concentration of inorganic particles IP and organic particles OP in the water sample and calculate the proportional relationship : ; when , then increase the scattering channel weight and reduce the absorption channel weight The value of ; if , then reduce the scattering channel weight and increase the absorption channel weight The value of is a preset ratio threshold; Use the weight calculation formula to weight the scattering channel and absorption channel weights Assign weights: ;in, 、 is a monotonic changing function obtained by fitting the suspended matter concentration.

[0014] In a further embodiment, the total suspended solids concentration remote sensing estimation model The expression is as follows: ,in, is the scattering channel weight, is the absorption channel weight.

[0015] In a further embodiment, the optimal first wavelength , optimal second wavelength and the optimal third wavelength The determination process is as follows: Set the first wavelength and the second wavelength is the initial value, the third wavelength The range of change is 400nm~900nm, and the third wavelength Calculation of the absorption coefficient as a variable The fitting accuracy of the total suspended matter concentration TSM is used to set the position corresponding to the third wavelength with the highest fitting accuracy as the optimal third wavelength. ; Set the first wavelength and the optimal third wavelength is the initial value, the second wavelength The range of change is 400nm~900nm, and the second wavelength Calculation of the absorption coefficient as a variable The fitting accuracy of the total suspended matter concentration TSM is used to set the position corresponding to the third wavelength with the highest fitting accuracy as the optimal second wavelength. ; Setting the optimal second wavelength and the optimal third wavelength is the initial value, the first wavelength The range of change is 400nm~900nm, the first wavelength Calculation of the absorption coefficient as a variable The fitting accuracy of the total suspended matter concentration TSM is used to set the position corresponding to the third wavelength with the highest fitting accuracy as the optimal first wavelength. .

[0016] A system for remote sensing inversion of suspended matter concentration in complex water bodies, used to implement the above-mentioned method, includes: The first module is configured to acquire and integrate remote sensing images and ground observation data to obtain synchronized sample pairs; The second module is set up to use synchronized sample pairs to obtain the backscatter coefficient of particulate matter based on water analysis. and absorption coefficient , and respectively determine the backscattering coefficient at the optimal wavelength position reflecting the total suspended matter concentration and the absorption coefficient at the optimal wavelength position; The third module is configured to characterize the inorganic suspended matter components using the backscattering coefficient at the optimal wavelength position. ; Use the absorption coefficient at the optimal wavelength position to characterize the components of organic suspended matter ; The fourth module is configured to calculate the ratio R between inorganic particles and organic particles in the water body, and determine the backscattering coefficient based on the ratio R. and absorption coefficient Importance in estimating total suspended matter concentrations; The fifth module is set up to comprehensively characterize the inorganic suspended matter components , Characterization of organic suspended matter components The importance of constructing a remote sensing estimation model for total suspended matter concentration in complex water bodies , using the remote sensing estimation model of total suspended solids concentration in the complex water body Output the suspended matter concentration in water bodies and apply it to remote sensing images.

[0017] Beneficial effects of the present invention: The present invention fully considers the differences in absorption and scattering characteristics between inorganic and organic particles in water bodies, constructs comprehensive characteristic parameters by weighted combination of the absorption coefficient and scattering coefficient of the particles, improves the response ability to changes in the optical properties of water bodies, and strengthens the physical mechanism basis of the suspended matter concentration inversion model.

[0018] By analyzing the proportion of particulate matter composition and dynamically determining the weight factors of absorption and scattering, the model parameters can be automatically adjusted according to different water environments, significantly improving the adaptability and generalization ability of the model under different water types and different particle composition conditions.

[0019] In a complex water environment, the present invention effectively reduces the deviation caused by single characteristic information through the collaborative inversion of absorption and scattering information, improves the accuracy of the total suspended matter concentration inversion results and the stability of spatial distribution estimation, and has higher application reliability.

[0020] The method of the present invention is adapted to high-temporal and spatial resolution satellite remote sensing data such as GOCI, realizing continuous inversion and dynamic monitoring of suspended matter concentrations in complex waters, and providing strong data support for regional water environment management and ecological protection.

[0021] In summary, this invention not only improves the accuracy and applicability of remote sensing inversion of suspended matter concentration in complex water bodies, but also expands the application scope of remote sensing technology in monitoring different water environments, with great scientific significance and application promotion value. It is applicable to complex water environments dominated by algae, non-algae, and mixed algae, and has strong adaptability and accuracy in remote sensing inversion of suspended matter concentration in complex water bodies. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 This is a comparison chart of the results of the synchronized ASD and GOCI spectra in Example 1.

[0023] Figure 2 This is a comparison chart of the changes in the inherent optical properties of water bodies with wavelength in Example 1.

[0024] Figure 3 This is a diagram showing the principle of determining the backscattering coefficient at the optimal wavelength position in Example 1.

[0025] Figure 4 This is a diagram showing the principle of determining the absorption coefficient at the optimal wavelength position in Example 1.

[0026] Figure 5 This is a graph showing the accuracy of the model constructed in Example 1. DETAILED DESCRIPTION

[0027] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0028] Example 1 This embodiment uses Taihu Lake and Hangzhou Bay as examples to disclose a method for remotely sensing the concentration of suspended solids in complex water bodies, including the following steps: Acquire and integrate remote sensing images and ground observation data to obtain synchronized sample pairs; Using synchronized sample pairs, the backscatter coefficient of particles is obtained based on water analysis. and absorption coefficient , and respectively determine the backscattering coefficient at the optimal wavelength position reflecting the total suspended matter concentration and the absorption coefficient at the optimal wavelength position; Characterizing Inorganic Suspended Matter Components Using Backscattering Coefficients at Optimal Wavelengths ; Use the absorption coefficient at the optimal wavelength position to characterize the components of organic suspended matter ; Calculate the ratio R of inorganic particles to organic particles in the water body, and determine the backscattering coefficient based on the ratio R and absorption coefficient Importance in estimating total suspended matter concentrations; Comprehensive characterization of inorganic suspended matter components , Characterization of organic suspended matter components The importance of constructing a remote sensing estimation model for total suspended matter concentration in complex water bodies , using the remote sensing estimation model of total suspended solids concentration in the complex water body Output the suspended matter concentration in water bodies and apply it to remote sensing images.

[0029] Furthermore, the remote sensing images are GOCI remote sensing images, which have a high spatial resolution of 500 meters and an ultra-high temporal resolution of 1 hour, and can acquire 8 images per day. GOCI is equipped with six visible light bands and two near-infrared bands, with central wavelengths of 412nm, 443nm, 490nm, 555nm, 660nm, 680nm, 745nm, and 865nm.

[0030] The ground observation data includes at least remote sensing reflectance and total suspended matter concentration (TSM). The remote sensing reflectance in this embodiment is collected using an ASD spectrometer, such as the FieldSpec® Pro Dual VNIR, with a wavelength range of 350-1050 nm and a spectral resolution of 3.5 nm.

[0031] Combined with the above data collection methods, Figure 1 The process of obtaining the synchronous sample point pair in this embodiment is as follows: pre-set the synchronization time interval of the remote sensing image and the ground observation data , according to the synchronization time interval Collect data; For example, every 1 h Collect GOCI remote sensing images, remote sensing reflectance and total suspended matter concentration TSM once. Figure 1 (A) is the reflectivity of Hangzhou Bay ASD at different wavelengths. Figure 1 (B) is the GOCI reflectivity of Hangzhou Bay at different wavelengths. Figure 1 (C) in the figure shows the comparison results of the synchronous ASD and GOCI spectra of Hangzhou Bay. Figure 1 (D) in the figure is the reflectivity of Taihu Lake ASD at different wavelengths. Figure 1 (E) is the Taihu GOCI reflectivity at different wavelengths. Figure 1 (F) in the figure is the comparison result of the simultaneous ASD and GOCI spectra of Taihu Lake.

[0032] Perform the following steps on the collected data: Define the error function , used to represent the difference between the spectral value of the remote sensing image window and the remote sensing reflectance; Using the difference minimization strategy, the pixel points that best match the remote sensing reflectance are searched in the remote sensing image window, and the spatial coordinates, spectral values, remote sensing reflectance, and suspended matter concentration of the pixel points are matched to form synchronized sample point pairs containing spatiotemporal matching information.

[0033] In a further embodiment, the error function The formula is expressed as: ;in, is the GOCI remote sensing image in the window coordinates The spectral value of for The maximum value of is the window coordinate of the ASD spectrometer The band observation value.

[0034] By minimizing the above spatial error function, we can find the matching point that best meets the spectral consistency requirements.

[0035] In a further embodiment, the backscatter coefficient of the particle The parsing process is as follows: The total backscatter coefficient is calculated using remote sensing reflectivity and water body characteristics ,in, ,in, is the wavelength The water surface remote sensing reflectivity, is a dimensionless factor, is the wavelength The total absorption coefficient, is the ratio of the radiation above the water surface to the radiation below the water surface. In this embodiment, The value of is 0.13, .

[0036] Get pure water at wavelength The backscatter coefficient , the backscattering coefficient of particles is calculated using the following formula: : .

[0037] Based on the backscattering coefficient of the above particles The determination process of the backscattering coefficient at the optimal wavelength position is as follows, combined with Figure 3 : The backscatter coefficient Fitting with the total suspended matter concentration TSM, the backscattering coefficient is calculated wave by wave within 400nm~900nm by an iterative method The fitting accuracy of the total suspended matter concentration TSM is calculated, and the curve of fitting accuracy versus wavelength is obtained, and the wavelength position with the highest fitting accuracy is finally determined as the optimal wavelength of the backscattering coefficient. , which is the backscattering coefficient at the optimal wavelength position. The calculation method of fitting accuracy can be the determination coefficient, root mean square error, etc. .

[0038] Correspondingly, the inorganic suspended matter components Characterized by: , The optimal wavelength The corresponding backscattering coefficient is calculated as follows: , is the wavelength The water surface remote sensing reflectivity, is the wavelength The total absorption coefficient, For pure water at wavelength The backscattering coefficient of . Figure 3 (A) is to determine the optimal wavelength Schematic diagram of the calculation fitting accuracy, Figure 3 (B) in the figure is a characterization diagram of the inorganic suspended matter components.

[0039] In another embodiment, the absorption coefficient The parsing process is as follows: Introducing a second wavelength and the third wavelength , the second wavelength , the third wavelength With the first wavelength The length relationship between them is pre-set. It should be noted that the introduction of the second wavelength and the third wavelength The purpose is to minimize the absorption coefficient of colored dissolved organic matter and the total backscattering coefficient.

[0040] For further reference, Figure 2 , the predetermined length relationship is expressed as: ; in, The first wavelength The absorption coefficient of colored dissolved organic matter, The second wavelength The absorption coefficient of colored dissolved organic matter, The third wavelength The absorption coefficient of colored dissolved organic matter is similar to that of and The first wavelength and the second wavelength The absorption coefficient, The first wavelength , second wavelength and the third wavelength The total backscatter coefficient, The third wavelength The absorption coefficient of . Figure 2 (A) is a comparison chart showing the changes in the inherent optical properties of Taihu Lake water as the wavelength changes. Figure 2 (B) is a comparison chart showing the changes in the inherent optical properties of Hangzhou Bay water as a function of wavelength.

[0041] The absorption coefficient is calculated using the following formula: : ; Where, 、 and The first wavelength , second wavelength and the third wavelength Remote sensing reflectivity, 、 and Pure water at the first wavelength , second wavelength and the third wavelength The absorption coefficient of .

[0042] Based on the above description, if Figure 4 The absorption coefficient at the optimal wavelength position shown is determined as follows: The absorption coefficient Fitting with the total suspended matter concentration TSM, the absorption coefficient is calculated wave by wave within 400nm~900nm by an iterative method The fitting accuracy of the total suspended matter concentration TSM is finally determined, and the wavelength position with the highest fitting accuracy is taken as the optimal first wavelength of the absorption coefficient. , optimal second wavelength and the optimal third wavelength ; Correspondingly, the organic suspended matter components Characterized by: , The optimal first wavelength The corresponding absorption coefficient.

[0043] Therefore, ,in, is the optimal first wavelength, is the optimal second wavelength, is the optimal third wavelength.

[0044] Therefore, in a further embodiment, the optimal first wavelength , optimal second wavelength and the optimal third wavelength The determination process is as follows: Set the first wavelength and the second wavelength is the initial value, the third wavelength The range of change is 400nm~900nm, and the third wavelength Calculation of the absorption coefficient as a variable The fitting accuracy of the total suspended matter concentration TSM is used to set the position corresponding to the third wavelength with the highest fitting accuracy as the optimal third wavelength. , , Figure 4 As shown in (A), Figure 4 (A) is used to determine the optimal third wavelength Schematic diagram of the calculated fitting accuracy.

[0045] Set the first wavelength and the optimal third wavelength is the initial value, the second wavelength The range of change is 400nm~900nm, and the second wavelength Calculation of the absorption coefficient as a variable The fitting accuracy of the total suspended matter concentration TSM is used to set the position corresponding to the third wavelength with the highest fitting accuracy as the optimal second wavelength. , , Figure 4 As shown in (B), Figure 4 (B) is to determine the optimal second wavelength Schematic diagram of the calculated fitting accuracy.

[0046] Setting the optimal second wavelength and the optimal third wavelength is the initial value, the first wavelength The range of change is 400nm~900nm, the first wavelength Calculation of the absorption coefficient as a variable The fitting accuracy of the total suspended matter concentration TSM is used to set the position corresponding to the third wavelength with the highest fitting accuracy as the optimal first wavelength. , . Figure 4 As shown in (C), Figure 4 (C) is used to determine the optimal first wavelength Schematic diagram of the calculated fitting accuracy.

[0047] Determine the optimal first wavelength by the above method , optimal second wavelength and the optimal third wavelength The organic suspended matter components can be obtained at 550nm, 750nm and 750nm respectively. A clear characterization of Figure 4 (D) in the Figure 4 (D) Characterization diagram of the suspended matter components.

[0048] In another embodiment, the importance of the backscatter coefficient is assigned to the scatter channel weights , the importance of the absorption coefficient gives the absorption channel a weight ; The importance determination process in the estimation of total suspended solids concentration is as follows: Determine the concentration of inorganic particles IP and organic particles OP in the water sample and calculate the proportional relationship : ; when , then increase the scattering channel weight and reduce the absorption channel weight The value of ; if , then reduce the scattering channel weight and increase the absorption channel weight The value of is a preset ratio threshold; Use the weight calculation formula to weight the scattering channel and absorption channel weights Assign weights: ;in, 、 is a monotonic changing function obtained by fitting the suspended matter concentration.

[0049] Combined with the scenario in this embodiment, such as Figure 2 According to the water characteristics of Hangzhou Bay (large proportion of inorganic particles), the scattering channel weight , absorption channel weight .

[0050] According to the water characteristics of Taihu Lake (large proportion of organic particles), the scattering channel weight , absorption channel weight .

[0051] In summary, the remote sensing estimation model of total suspended matter concentration The expression is as follows: ,in, is the scattering channel weight, is the absorption channel weight.

[0052] Apply the above model to GOCI remote sensing images, such as Figure 5 As shown, the optimal wavelength of the backscattering part in the model is Corresponding to the seventh band of GOCI data; the first wavelength of the optimal absorption part in the model , optimal second wavelength and the optimal third wavelength They correspond to the fourth, seventh and seventh bands of GOCI data respectively, and realize the remote sensing inversion and spatial distribution estimation of total suspended matter concentration in complex water areas. Figure 5 (A) is the accuracy performance diagram of the model in ASD hyperspectral. Figure 5 (B) in the figure shows the accuracy of the model on the GOCI satellite.

[0053] Example 2 This embodiment provides a system for remotely sensing and inverting the suspended matter concentration of complex water bodies, which is used to implement the method described in the embodiment, including: The first module is configured to acquire and integrate remote sensing images and ground observation data to obtain synchronized sample pairs; The second module is set up to use synchronized sample pairs to obtain the backscatter coefficient of particulate matter based on water analysis. and absorption coefficient , and respectively determine the backscattering coefficient at the optimal wavelength position reflecting the total suspended matter concentration and the absorption coefficient at the optimal wavelength position; The third module is configured to characterize the inorganic suspended matter components using the backscattering coefficient at the optimal wavelength position. ; Use the absorption coefficient at the optimal wavelength position to characterize the components of organic suspended matter ; The fourth module is set to calculate the ratio of inorganic particles to organic particles in water , based on the proportional relationship Determine the backscatter coefficient and absorption coefficient Importance in estimating total suspended matter concentrations; The fifth module is set up to comprehensively characterize the inorganic suspended matter components , Characterization of organic suspended matter components The importance of constructing a remote sensing estimation model for total suspended matter concentration in complex water bodies , using the remote sensing estimation model of total suspended solids concentration in the complex water body Output the suspended matter concentration in water bodies and apply it to remote sensing images.

Claims

1. A method for remote sensing inversion of suspended matter concentration in complex water bodies, characterized in that: The following steps are involved: Acquire and integrate remote sensing images and ground observation data to obtain synchronized sample pairs; Using synchronized sample pairs, the backscatter coefficient of particles is obtained based on water analysis. and absorption coefficient , and respectively determine the backscattering coefficient at the optimal wavelength position reflecting the total suspended matter concentration and the absorption coefficient at the optimal wavelength position; Characterizing Inorganic Suspended Matter Components Using Backscattering Coefficients at Optimal Wavelengths ; Use the absorption coefficient at the optimal wavelength position to characterize the components of organic suspended matter ; Calculate the ratio of inorganic particles to organic particles in water , based on the proportional relationship Determine the backscatter coefficient and absorption coefficient Importance in estimating total suspended matter concentrations; Comprehensive characterization of inorganic suspended matter components , Characterization of organic suspended matter components The importance of constructing a remote sensing estimation model for total suspended matter concentration in complex water bodies , using the remote sensing estimation model of total suspended solids concentration in the complex water body Output the suspended matter concentration in water bodies and apply it to remote sensing images.

2. The method for remote sensing inversion of suspended matter concentration in complex water bodies according to claim 1, characterized in that: The ground observation data at least includes: remote sensing reflectance and total suspended matter concentration TSM; the process of obtaining the synchronous sample point pair is as follows: Pre-set synchronization time intervals for remote sensing images and ground observation data , according to the synchronization time interval Collect data; perform the following steps on the collected data: Define the error function , used to represent the difference between the spectral value of the remote sensing image window and the remote sensing reflectance; Using the difference minimization strategy, the pixel points that best match the remote sensing reflectance are searched in the remote sensing image window, and the spatial coordinates, spectral values, remote sensing reflectance, and suspended matter concentration of the pixel points are matched to form synchronized sample point pairs containing spatiotemporal matching information.

3. The method for remote sensing inversion of suspended matter concentration in complex water bodies according to claim 1, characterized in that: The backscattering coefficient of the particle The parsing process is as follows: The total backscatter coefficient is calculated using remote sensing reflectivity and water body characteristics ; Get pure water at wavelength The backscatter coefficient , the backscattering coefficient of particles is calculated using the following formula: : 。 4. The method for remote sensing inversion of suspended matter concentration in complex water bodies according to claim 1, characterized in that: The absorption coefficient The parsing process is as follows: Introducing a second wavelength and the third wavelength , the second wavelength , the third wavelength With the first wavelength Satisfy a pre-set length relationship between each other; The absorption coefficient is calculated using the following formula: : ； Where, 、 and The first wavelength , second wavelength and the third wavelength Remote sensing reflectivity, 、 and Pure water at the first wavelength , second wavelength and the third wavelength The absorption coefficient of .

5. The method for remote sensing inversion of suspended matter concentration in complex water bodies according to claim 1, characterized in that: The process of determining the backscattering coefficient at the optimal wavelength position is as follows: The backscatter coefficient Fitting with the total suspended matter concentration TSM, the backscattering coefficient is calculated wave by wave within 400nm~900nm by an iterative method The fitting accuracy of the total suspended matter concentration TSM is calculated, and the curve of fitting accuracy versus wavelength is obtained, and the wavelength position with the highest fitting accuracy is finally determined as the optimal wavelength of the backscattering coefficient. , which is the backscattering coefficient at the optimal wavelength position; Correspondingly, the inorganic suspended matter components Characterized by: , The optimal wavelength The corresponding backscattering coefficient.

6. The method for remote sensing inversion of suspended solids concentration in complex water bodies according to claim 1, characterized in that: The process of determining the absorption coefficient at the optimal wavelength position is as follows: The absorption coefficient Fitting with the total suspended matter concentration TSM, the absorption coefficient is calculated wave by wave within 400nm~900nm by an iterative method The fitting accuracy of the total suspended matter concentration TSM is finally determined, and the wavelength position with the highest fitting accuracy is taken as the optimal first wavelength of the absorption coefficient. , optimal second wavelength and the optimal third wavelength ; Correspondingly, the organic suspended matter components Characterized by: , The optimal first wavelength The corresponding absorption coefficient.

7. The method for remote sensing inversion of suspended matter concentration in complex water bodies according to claim 1, characterized in that: Assigning weights to the scattering channels based on the importance of the backscatter coefficient , the importance of the absorption coefficient gives the absorption channel a weight ; The importance determination process in the estimation of total suspended solids concentration is as follows: Determine the concentration of inorganic particles IP and organic particles OP in the water sample and calculate the proportional relationship : ; when , then increase the scattering channel weight and reduce the absorption channel weight The value of ; if , then reduce the scattering channel weight and increase the absorption channel weight The numerical value of is a preset ratio threshold; Use the weight calculation formula to weight the scattering channel and absorption channel weights Assign weights: ;in, 、 is a monotonic changing function obtained by fitting the suspended matter concentration.

8. The method for remote sensing inversion of suspended solids concentration in complex water bodies according to claim 1, characterized in that: The total suspended matter concentration remote sensing estimation model The expression is as follows: ,in, is the scattering channel weight, is the absorption channel weight.

9. The method for remote sensing inversion of suspended matter concentration in complex water bodies according to claim 4, characterized in that: The optimal first wavelength , optimal second wavelength and the optimal third wavelength The determination process is as follows: Set the first wavelength and the second wavelength is the initial value, the third wavelength The range of change is 400nm~900nm, and the third wavelength Calculation of the absorption coefficient as a variable The fitting accuracy of the total suspended matter concentration TSM is used to set the position corresponding to the third wavelength with the highest fitting accuracy as the optimal third wavelength. ; Set the first wavelength and the optimal third wavelength is the initial value, the second wavelength The range of change is 400nm~900nm, and the second wavelength Calculation of the absorption coefficient as a variable The fitting accuracy of the total suspended matter concentration TSM is used to set the position corresponding to the third wavelength with the highest fitting accuracy as the optimal second wavelength. ; Setting the optimal second wavelength and the optimal third wavelength is the initial value, the first wavelength The range of change is 400nm~900nm, the first wavelength Calculation of the absorption coefficient as a variable The fitting accuracy of the total suspended matter concentration TSM is used to set the position corresponding to the third wavelength with the highest fitting accuracy as the optimal first wavelength. .

10. A system for remote sensing inversion of suspended solids concentration in complex water bodies, used to implement the method according to any one of claims 1 to 9, characterized in that: include: The first module is configured to acquire and integrate remote sensing images and ground observation data to obtain synchronized sample pairs; The second module is set up to use synchronized sample pairs to obtain the backscatter coefficient of particles based on water analysis. and absorption coefficient , and respectively determine the backscattering coefficient at the optimal wavelength position reflecting the total suspended matter concentration and the absorption coefficient at the optimal wavelength position; The third module is configured to characterize the inorganic suspended matter components using the backscattering coefficient at the optimal wavelength position. ; Characterizing organic suspended matter components using the absorption coefficient at the optimal wavelength position ; The fourth module is configured to calculate the ratio R between inorganic particles and organic particles in the water body, and determine the backscattering coefficient based on the ratio R. and absorption coefficient Importance in estimating total suspended matter concentrations; The fifth module is set up to comprehensively characterize the inorganic suspended matter components , Characterization of organic suspended matter components The importance of constructing a remote sensing estimation model for total suspended matter concentration in complex water bodies , using the remote sensing estimation model of total suspended solids concentration in the complex water body Output the suspended matter concentration in water bodies and apply it to remote sensing images.

Images