A method and system for remote sensing inversion of suspended matter concentration in complex water bodies

Abstract

This invention discloses a method and system for remote sensing inversion of suspended solids concentration in complex water bodies, belonging to the technical field of ecological environment. It includes: integrating remote sensing imagery and measured data to obtain synchronous sample point pairs; analyzing the backscattering coefficient of particulate matter and determining the optimal wavelength position reflecting the total suspended solids concentration using an iterative method; analyzing the absorption coefficient of particulate matter and determining the optimal wavelength position reflecting the total suspended solids concentration using an iterative method; determining the weights of absorption and scattering coefficients in the estimation of total suspended solids concentration based on the ratio of inorganic to organic particulate matter in the water body; constructing a remote sensing estimation model for the total suspended solids concentration in complex water bodies and applying it to remote sensing imagery. This invention considers the differences in absorption and scattering characteristics between organic and inorganic particulate matter, and is applicable to water bodies dominated by algae, water bodies dominated by non-algae, and complex water environments where both are mixed, exhibiting strong adaptability and accuracy.

Description

Technical Field

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

[0002] Suspended solids concentration is a crucial parameter reflecting the condition of aquatic environments, effectively indicating water cleanliness, nutrient status, and ecosystem health. Traditional methods for monitoring suspended solids concentration rely primarily on on-site sampling and laboratory analysis, which are time-consuming, costly, and have limited spatial coverage. Remote sensing technology, with its advantages of large-scale, rapid, and dynamic monitoring, has become an important tool for water environment monitoring. However, in practical applications, complex water bodies contain various components such as inorganic particles, organic particles, and plankton. These components exhibit significant differences in optical properties, displaying different absorption and scattering behaviors, resulting in extremely complex remote sensing reflection signals. Inorganic particles (such as mineral sediments) primarily cause strong scattering, while organic particles (such as phytoplankton remains and organic debris) have higher absorption characteristics. This difference in particle composition presents several major challenges for remote sensing inversion:

[0003] 1. Insufficient model applicability: Most existing inversion models rely on the optical properties of a single type of water body for construction, making it difficult to adapt to complex water environments where multiple particulate matter coexists.

[0004] 2. Insufficient utilization of feature 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.

[0005] 3. Poor stability and generalization ability: When the water type and particle composition change, the model performance deteriorates significantly, making it difficult to guarantee the inversion accuracy in different regions or at different time scales.

[0006] 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 complex changes in aquatic environments, so as to improve the accuracy and applicability of the inversion results. Summary of the Invention

[0007] To address the technical problems existing in the background art, the present invention provides a method and system for remote sensing inversion of suspended solids concentration in complex water bodies.

[0008] This invention employs the following technical solution: a method for remotely inverting the concentration of suspended solids in complex water bodies, comprising the following steps:

[0009] Acquire and integrate remote sensing imagery and ground observation data to obtain synchronized sample point pairs;

[0010] The backscattering coefficient of particulate matter was obtained by using synchronous sample pairs based on water body analysis. and absorption coefficient And determine the backscattering coefficient and absorption coefficient at the optimal wavelength position that reflect the total suspended matter concentration, respectively.

[0011] Characterizing inorganic suspended matter components using the backscattering coefficient at the optimal wavelength position The absorption coefficient at the optimal wavelength position is used to characterize the components of organic suspended matter. ;

[0012] Calculate the ratio of inorganic to organic particulate matter in water bodies. Based on the aforementioned proportional relationship Determine the backscattering coefficient and absorption coefficient Its importance in estimating total suspended solids concentration;

[0013] In summary, the inorganic suspended matter component characterization Characterization of organic suspended matter components Based on the aforementioned importance, a remote sensing estimation model for total suspended solids concentration in complex water bodies is constructed. The total suspended solids concentration in the complex water body was estimated using a remote sensing model. Output the concentration of suspended solids in water and apply it to remote sensing images.

[0014] In a further embodiment, the ground observation data includes at least: remotely sensed reflectance and total suspended matter concentration (TSM); the process for obtaining the synchronous sampling point pair is as follows:

[0015] Pre-set the synchronization time interval between remote sensing images and ground observation data According to the synchronization time interval Collect data; perform the following steps on the collected data:

[0016] Define error function This is used to represent the difference between the spectral values ​​and remote sensing reflectance of a remote sensing image window;

[0017] By adopting a difference minimization strategy, the pixel point that best matches the remote sensing reflectance is searched within the remote sensing image window. The spatial coordinates, spectral values, remote sensing reflectance, and suspended matter concentration of the pixel point are matched to form a synchronous sample point pair containing spatiotemporal matching information.

[0018] In a further embodiment, the backscattering coefficient of the particulate matter The parsing process is as follows:

[0019] The total backscattering coefficient was calculated using remote sensing reflectance and water body characteristics. ;

[0020] Obtain pure water at wavelength Backscattering coefficient The backscattering coefficient of particulate matter is calculated using the following formula. :

[0021] .

[0022] In a further embodiment, the absorption coefficient The parsing process is as follows:

[0023] Introducing a second wavelength and the third wavelength The second wavelength Third wavelength With the first wavelength They satisfy a pre-defined length relationship;

[0024] The absorption coefficient is calculated using the following formula. :

[0025] ;

[0026] In the formula, , and The first wavelength is respectively Second wavelength and the third wavelength Remote sensing reflectance, , and Pure water at the first wavelength Second wavelength and the third wavelength The absorption coefficient.

[0027] In a further embodiment, the process for determining the backscattering coefficient at the optimal wavelength position is as follows:

[0028] Backscattering coefficient The backscattering coefficient was fitted to the total suspended solids concentration (TSM) and calculated wave-by-wave in the 400 nm–900 nm range using an iterative method. The fitting accuracy with the total suspended solids concentration (TSM) was determined, and the curve of fitting accuracy as a function of wavelength was obtained. Finally, the wavelength position with the highest fitting accuracy was determined as the optimal wavelength for the backscattering coefficient. That is, the backscattering coefficient at the optimal wavelength position;

[0029] Correspondingly, the inorganic suspended components Characterized as: , For the optimal wavelength The corresponding backscattering coefficient.

[0030] In a further embodiment, the process for determining the absorption coefficient at the optimal wavelength position is as follows:

[0031] absorption coefficient The absorption coefficient was fitted to the total suspended solids concentration (TSM) and calculated wave-by-wave in the 400 nm–900 nm range using an iterative method. The wavelength position with the highest fitting accuracy to the total suspended solids concentration (TSM) was ultimately determined as the optimal first wavelength for the absorption coefficient. Optimal second wavelength and the optimal third wavelength ;

[0032] Correspondingly, organic suspended matter components Characterized as: , The optimal first wavelength The corresponding absorption coefficient.

[0033] In a further embodiment, the importance of the backscattering coefficient is assigned to the scattering channel weights. The importance of the absorption coefficient assigns weight to the absorption channel. The process for determining its importance in the estimation of total suspended solids concentration is as follows:

[0034] The concentrations of inorganic particulate matter (IP) and organic particulate matter (OP) in the water sample were measured, and their ratio was calculated. : ;

[0035] when Then increase the weight of the scattering channel. The value, and reduce the absorption channel weight. The value; if Then reduce the weight of the scattering channel. The value, and increase the absorption channel weight. The value; For a pre-set ratio threshold;

[0036] The weights of the scattering channels are calculated using the weighting formula. and absorption channel weights Perform weight assignment:

[0037] ;in, , This is a monotonic function obtained by fitting the suspended solids concentration.

[0038] In a further embodiment, the total suspended matter concentration remote sensing estimation model The expression is as follows:

[0039] ,in, For scattering channel weights, This is for absorbing channel weights.

[0040] In a further embodiment, the optimal first wavelength Optimal second wavelength and the optimal third wavelength The determination process is as follows:

[0041] Set the first wavelength Second wavelength The initial value, the third wavelength The variation range is 400nm~900nm, and the third wavelength As a variable, the absorption coefficient is calculated. The optimal third wavelength is determined by matching the total suspended solids concentration (TSM) with the third wavelength position that yields the highest fitting accuracy. ;

[0042] Set the first wavelength and the optimal third wavelength As the initial value, the second wavelength The variation range is 400nm~900nm, and the second wavelength As a variable, the absorption coefficient is calculated. The optimal second wavelength was determined by setting the position corresponding to the third wavelength with the highest fitting accuracy to the total suspended solids (TSM) concentration. ;

[0043] Setting the optimal second wavelength and the optimal third wavelength As the initial value, the first wavelength The variation range is 400nm~900nm, with the first wavelength... As a variable, the absorption coefficient is calculated. The optimal first wavelength was determined by setting the position corresponding to the third wavelength with the highest fitting accuracy to the total suspended solids (TSM) concentration. .

[0044] A system for remotely sensing and inverting the concentration of suspended solids in complex water bodies, used to implement the method described above, includes:

[0045] The first module is configured to acquire and integrate remote sensing imagery and ground observation data to obtain synchronized sample point pairs;

[0046] The second module is configured to use synchronized sample pairs to obtain the backscattering coefficient of particulate matter based on water body analysis. and absorption coefficient And determine the backscattering coefficient and absorption coefficient at the optimal wavelength position that reflect the total suspended matter concentration, respectively.

[0047] The third module is configured to characterize the inorganic suspended matter components using the backscattering coefficient at the optimal wavelength position. The absorption coefficient at the optimal wavelength position is used to characterize the components of organic suspended matter. ;

[0048] The fourth module is configured to calculate the ratio R of inorganic and organic particulate matter in the water body, and determine the backscattering coefficient based on the ratio R. and absorption coefficient Its importance in estimating total suspended solids concentration;

[0049] The fifth module is configured to comprehensively characterize the inorganic suspended matter components. Characterization of organic suspended matter components Based on the aforementioned importance, a remote sensing estimation model for total suspended solids concentration in complex water bodies is constructed. The total suspended solids concentration in the complex water body was estimated using a remote sensing model. Output the concentration of suspended solids in water and apply it to remote sensing images.

[0050] The beneficial effects of this invention are:

[0051] This invention fully considers the differences in absorption and scattering characteristics between inorganic and organic particulate matter in water. By weighted combination of the absorption and scattering coefficients of particulate matter, a comprehensive characteristic parameter is constructed, which improves the response capability to changes in the optical properties of water and enhances the physical mechanism basis of the suspended matter concentration inversion model.

[0052] By analyzing the composition ratio of particulate matter to dynamically determine the weighting factors of absorption and scattering, the model parameters can be automatically adjusted according to different water environments, which significantly improves the model's adaptability and generalization ability under different water types and different particulate composition conditions.

[0053] In complex aquatic environments, this invention effectively reduces the bias caused by single feature information by synergistic inversion of absorption and scattering information, improves the accuracy of total suspended solids concentration inversion results and the stability of spatial distribution estimation, and has higher application reliability.

[0054] The method of this invention is adapted to satellite remote sensing data with high spatiotemporal resolution such as GOCI, enabling continuous inversion and dynamic monitoring of suspended solids concentration in complex water areas, and providing strong data support for regional water environment management and ecological protection.

[0055] In summary, this invention not only improves the accuracy and applicability of remote sensing inversion of suspended solids concentration in complex water bodies, but also expands the application scope of remote sensing technology in monitoring different water environments, possessing significant scientific value and application promotion potential. It is applicable to water bodies dominated by algae, water bodies dominated by non-algae, and complex water environments where both are present, demonstrating strong adaptability and accuracy in remote sensing inversion of suspended solids concentration in complex water bodies. Attached Figure Description

[0056] Figure 1 This is a comparison chart of the results of simultaneous ASD and GOCI spectra in Example 1.

[0057] Figure 2 This is a comparison diagram of the changes in the inherent optical properties of water with wavelength in Example 1.

[0058] Figure 3 This is a schematic diagram illustrating the principle of determining the backscattering coefficient at the optimal wavelength position in Example 1.

[0059] Figure 4 This is a schematic diagram illustrating the principle of determining the absorption coefficient at the optimal wavelength position in Example 1.

[0060] Figure 5 This is a graph showing the accuracy of the model constructed in Example 1. Detailed Implementation

[0061] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0062] Example 1

[0063] This embodiment uses Taihu Lake and Hangzhou Bay as examples to disclose a method for remote sensing inversion of suspended solids concentration in complex water bodies, including the following steps:

[0064] Acquire and integrate remote sensing imagery and ground observation data to obtain synchronized sample point pairs;

[0065] The backscattering coefficient of particulate matter was obtained by using synchronous sample pairs based on water body analysis. and absorption coefficient And determine the backscattering coefficient and absorption coefficient at the optimal wavelength position that reflect the total suspended matter concentration, respectively.

[0066] Characterizing inorganic suspended matter components using the backscattering coefficient at the optimal wavelength position The absorption coefficient at the optimal wavelength position is used to characterize the components of organic suspended matter. ;

[0067] Calculate the ratio R of inorganic and organic particulate matter in the water body, and determine the backscattering coefficient based on the ratio R. and absorption coefficient Its importance in estimating total suspended solids concentration;

[0068] In summary, the inorganic suspended matter component characterization Characterization of organic suspended matter components Based on the aforementioned importance, a remote sensing estimation model for total suspended solids concentration in complex water bodies is constructed. The total suspended solids concentration in the complex water body was estimated using a remote sensing model. Output the concentration of suspended solids in water and apply it to remote sensing images.

[0069] Furthermore, the remote sensing imagery is GOCI imagery, which boasts a high spatial resolution of 500 meters and an ultra-high temporal resolution of 1 hour, acquiring 8 images per day. GOCI is equipped with six visible light bands and two near-infrared bands, with center wavelengths of 412nm, 443nm, 490nm, 555nm, 660nm, 680nm, 745nm, and 865nm, respectively.

[0070] The ground observation data includes at least: remotely sensed reflectance and total suspended matter (TSM) concentration. In this embodiment, the remotely sensed reflectance was acquired using an ASD spectrometer, such as FieldSpec® Pro Dual VNIR, with a wavelength range of 350–1050 nm and a spectral resolution of 3.5 nm.

[0071] Combining the above data collection methods, and combining Figure 1 The process of obtaining the synchronized sample pairs in this embodiment is as follows: The synchronization time interval between remote sensing images and ground observation data is preset. According to the synchronization time interval Data collection; For example, that is, every 1 h One GOCI remote sensing image, remote sensing reflectance, and total suspended matter concentration (TSM) were acquired. Figure 1 In the figure, (A) represents the ASD reflectivity of Hangzhou Bay at different wavelengths. Figure 1 (B) represents the GOCI reflectance 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 In the figure, (D) represents the ASD reflectance of Taihu Lake at different wavelengths. Figure 1 In the figure, (E) represents the reflectance of GOCI in Taihu Lake at different wavelengths. Figure 1 (F) in the figure shows the comparison results of the synchronous ASD and GOCI spectra of Taihu Lake.

[0072] Perform the following steps on the collected data: Define the error function. This is used to represent the difference between the spectral values ​​and remote sensing reflectance of a remote sensing image window;

[0073] By adopting a difference minimization strategy, the pixel point that best matches the remote sensing reflectance is searched within the remote sensing image window. The spatial coordinates, spectral values, remote sensing reflectance, and suspended matter concentration of the pixel point are matched to form a synchronous sample point pair containing spatiotemporal matching information.

[0074] In a further embodiment, the error function The formula is expressed as:

[0075] ;in, It is the GOCI remote sensing image in window coordinates spectral values, for The maximum value, The ASD spectrometer in window coordinates The band observation values.

[0076] By minimizing the spatial error function described above, we can find the matching point that best meets the spectral consistency requirements.

[0077] In a further embodiment, the backscattering coefficient of the particulate matter The parsing process is as follows:

[0078] The total backscattering coefficient was calculated using remote sensing reflectance and water body characteristics. ,in, ,in, wavelength Water surface remote sensing reflectance, It is a dimensionless factor. wavelength The overall absorption coefficient, This is the ratio of radiation above the water surface to radiation below the water surface, in this embodiment... The value is 0.13. .

[0079] Obtain pure water at wavelength Backscattering coefficient The backscattering coefficient of particulate matter is calculated using the following formula. :

[0080] .

[0081] Based on the backscattering coefficient of the above-mentioned particulate matter The determination of the backscattering coefficient at the optimal wavelength position is as follows, combined with... Figure 3 :

[0082] Backscattering coefficient The backscattering coefficient was fitted to the total suspended solids concentration (TSM) and calculated wave-by-wave in the 400 nm–900 nm range using an iterative method. The fitting accuracy with the total suspended solids concentration (TSM) was determined, and the curve of fitting accuracy as a function of wavelength was obtained. Finally, the wavelength position with the highest fitting accuracy was determined as the optimal wavelength for the backscattering coefficient. This is the backscattering coefficient at the optimal wavelength position. The fitting accuracy can be calculated using methods such as the coefficient of determination, root mean square error, etc., based on the calculations in this embodiment. .

[0083] Correspondingly, the inorganic suspended components Characterized as: , For the optimal wavelength The corresponding backscattering coefficient is calculated using the following formula:

[0084] , wavelength Water surface remote sensing reflectance, wavelength The overall absorption coefficient, For pure water at wavelength The backscattering coefficient. Wherein Figure 3 (A) in the figure is for determining the optimal wavelength. A schematic diagram illustrating the calculated fitting accuracy. Figure 3 (B) in the figure is a characterization diagram of inorganic suspended matter components.

[0085] In another embodiment, the absorption coefficient The parsing process is as follows:

[0086] Introducing a second wavelength and the third wavelength The second wavelength Third wavelength With the first wavelength They satisfy a pre-defined length relationship with each other; it should be noted that a second wavelength is introduced. and the third wavelength The aim is to minimize the absorption coefficient of colored dissolved organic matter and the total backscattering coefficient.

[0087] Further reference Figure 2 The pre-defined length relationship is expressed as follows:

[0088] ;

[0089] in, For the first wavelength The absorption coefficient of colored dissolved organic matter, For the second wavelength The absorption coefficient of colored dissolved organic matter, For the third wavelength The absorption coefficients of colored dissolved organic matter are similar to those of [other compounds]. and The first wavelength is respectively Second wavelength absorption coefficient, The first wavelength is respectively Second wavelength and the third wavelength Total backscattering coefficient, For the third wavelength The absorption coefficient. Wherein, Figure 2 (A) in the figure is a comparison diagram of the inherent optical properties of Taihu Lake as a function of wavelength. Figure 2 (B) in the figure is a comparison of the inherent optical properties of Hangzhou Bay water as a function of wavelength.

[0090] The absorption coefficient is calculated using the following formula. :

[0091] ;

[0092] In the formula, , and The first wavelength is respectively Second wavelength and the third wavelength Remote sensing reflectance, , and Pure water at the first wavelength Second wavelength and the third wavelength The absorption coefficient.

[0093] Based on the above description, such as Figure 4 The process for determining the absorption coefficient at the optimal wavelength position is as follows:

[0094] absorption coefficient The absorption coefficient was fitted to the total suspended solids concentration (TSM) and calculated wave-by-wave in the 400 nm–900 nm range using an iterative method. The wavelength position with the highest fitting accuracy to the total suspended solids concentration (TSM) was ultimately determined as the optimal first wavelength for the absorption coefficient. Optimal second wavelength and the optimal third wavelength ;

[0095] Correspondingly, organic suspended matter components Characterized as: , The optimal first wavelength The corresponding absorption coefficient.

[0096] Therefore, ,in, The optimal first wavelength, The optimal second wavelength, The optimal third wavelength.

[0097] Therefore, in a further embodiment, the optimal first wavelength Optimal second wavelength and the optimal third wavelength The determination process is as follows:

[0098] Set the first wavelength Second wavelength The initial value, the third wavelength The variation range is 400nm~900nm, and the third wavelength As a variable, the absorption coefficient is calculated. The optimal third wavelength is determined by matching the total suspended solids concentration (TSM) with the third wavelength position that yields the highest fitting accuracy. , , Figure 4 As shown in (A), Figure 4 (A) in the equation is for determining the optimal third wavelength. A schematic diagram illustrating the calculated fitting accuracy.

[0099] Set the first wavelength and the optimal third wavelength As the initial value, the second wavelength The variation range is 400nm~900nm, and the second wavelength As a variable, the absorption coefficient is calculated. The optimal second wavelength was determined by setting the position corresponding to the third wavelength with the highest fitting accuracy to the total suspended solids (TSM) concentration. , , Figure 4 As shown in (B) in the image, Figure 4 (B) in the figure is for determining the optimal second wavelength. A schematic diagram illustrating the calculated fitting accuracy.

[0100] Setting the optimal second wavelength and the optimal third wavelength As the initial value, the first wavelength The variation range is 400nm~900nm, with the first wavelength... As a variable, the absorption coefficient is calculated. The optimal first wavelength was determined by setting the position corresponding to the third wavelength with the highest fitting accuracy to the total suspended solids (TSM) concentration. , . Figure 4 As shown in (C), Figure 4 (C) in the figure represents the determination of the optimal first wavelength. A schematic diagram illustrating the calculated fitting accuracy.

[0101] The optimal first wavelength is determined using the method described above. Optimal second wavelength and the optimal third wavelength Organic suspended matter components can be obtained at wavelengths of 550 nm, 750 nm, and 750 nm, respectively. A clear representation, refer to Figure 4 (D) Figure 4 Characterization diagram of (D) suspended solids components.

[0102] In another embodiment, the importance of the backscattering coefficient is assigned to the scattering channel weights. The importance of the absorption coefficient assigns weight to the absorption channel. The process for determining its importance in the estimation of total suspended solids concentration is as follows:

[0103] The concentrations of inorganic particulate matter (IP) and organic particulate matter (OP) in the water sample were measured, and their ratio was calculated. : ;

[0104] when Then increase the weight of the scattering channel. The value, and reduce the absorption channel weight. The value; if Then reduce the weight of the scattering channel. The value, and increase the absorption channel weight. The value; For a pre-set ratio threshold;

[0105] The weights of the scattering channels are calculated using the weighting formula. and absorption channel weights Perform weight assignment:

[0106] ;in, , This is a monotonic function obtained by fitting the suspended solids concentration.

[0107] In the context of this embodiment, such as Figure 2 Based on the water characteristics of Hangzhou Bay (large proportion of inorganic particles), its scattering channel weight Absorption channel weight .

[0108] Based on the water characteristics of Taihu Lake (large proportion of organic particulate matter), its scattering channel weight Absorption channel weight .

[0109] In summary, the remote sensing estimation model for total suspended solids concentration... The expression is as follows:

[0110] ,in, For scattering channel weights, This is for absorbing channel weights.

[0111] Applying the above model to GOCI remote sensing imagery, such as... Figure 5 As shown, the optimal wavelength for the backscattering component in the model is... Corresponding to the seventh band of the GOCI data; the optimal first wavelength for the absorption portion in the model. Optimal second wavelength and the optimal third wavelength Corresponding to the fourth, seventh, and seventh bands of GOCI data, respectively, remote sensing inversion of total suspended solids concentration and spatial distribution estimation were achieved in complex water bodies. Figure 5 (A) in the figure represents the model's accuracy performance in the ASD hyperspectral image. Figure 5 (B) in the figure shows the accuracy of the model on the GOCI satellite.

[0112] Example 2

[0113] This embodiment provides a system for remotely inverting the concentration of suspended solids in complex water bodies, used to implement the methods described in this embodiment, including:

[0114] The first module is configured to acquire and integrate remote sensing imagery and ground observation data to obtain synchronized sample point pairs;

[0115] The second module is configured to use synchronized sample pairs to obtain the backscattering coefficient of particulate matter based on water body analysis. and absorption coefficient And determine the backscattering coefficient and absorption coefficient at the optimal wavelength position that reflect the total suspended matter concentration, respectively.

[0116] The third module is configured to characterize the inorganic suspended matter components using the backscattering coefficient at the optimal wavelength position. The absorption coefficient at the optimal wavelength position is used to characterize the components of organic suspended matter. ;

[0117] The fourth module is set up to calculate the ratio of inorganic to organic particulate matter in water. Based on the aforementioned proportional relationship Determine the backscattering coefficient and absorption coefficient Its importance in estimating total suspended solids concentration;

[0118] The fifth module is configured to comprehensively characterize the inorganic suspended matter components. Characterization of organic suspended matter components Based on the aforementioned importance, a remote sensing estimation model for total suspended solids concentration in complex water bodies is constructed. The total suspended solids concentration in the complex water body was estimated using a remote sensing model. Output the concentration of suspended solids in water and apply it to remote sensing images.

Claims

1. A method for remotely inverting the concentration of suspended solids in complex water bodies, characterized in that, Includes the following steps: Acquire and integrate remote sensing imagery and ground observation data to obtain synchronized sample point pairs; The backscattering coefficient of particulate matter was obtained by using synchronous sample pairs based on water body analysis. and absorption coefficient And determine the backscattering coefficient and absorption coefficient at the optimal wavelength position that reflect the total suspended matter concentration, respectively. Characterizing inorganic suspended matter components using the backscattering coefficient at the optimal wavelength position The absorption coefficient at the optimal wavelength position is used to characterize the components of organic suspended matter. ; Calculate the ratio of inorganic to organic particulate matter in water bodies. Based on the aforementioned proportional relationship Determine the backscattering coefficient and absorption coefficient Its importance in estimating total suspended solids concentration; In summary, the inorganic suspended matter component characterization Characterization of organic suspended matter components Based on the aforementioned importance, a remote sensing estimation model for total suspended solids concentration in complex water bodies is constructed. The total suspended solids concentration in the complex water body was estimated using a remote sensing model. Output the concentration of suspended solids in water and apply it to remote sensing images; The absorption coefficient The parsing process is as follows: Introducing a second wavelength and the third wavelength The second wavelength Third wavelength With the first wavelength They satisfy a pre-defined length relationship; The absorption coefficient is calculated using the following formula. : ； In the formula, , and The first wavelength is respectively Second wavelength and the third wavelength Remote sensing reflectance, , and Pure water at the first wavelength Second wavelength and the third wavelength absorption coefficient; The process for determining the backscattering coefficient at the optimal wavelength position is as follows: Backscattering coefficient The backscattering coefficient was fitted to the total suspended solids concentration (TSM) and calculated wave-by-wave in the 400 nm–900 nm range using an iterative method. The fitting accuracy with the total suspended solids concentration (TSM) was determined, and the curve of fitting accuracy as a function of wavelength was obtained. Finally, the wavelength position with the highest fitting accuracy was determined as the optimal wavelength for the backscattering coefficient. That is, the backscattering coefficient at the optimal wavelength position; Correspondingly, the inorganic suspended components Characterized as: , For the optimal wavelength The corresponding backscattering coefficient; The process for determining the absorption coefficient at the optimal wavelength position is as follows: absorption coefficient The absorption coefficient was fitted to the total suspended solids concentration (TSM) and calculated wave-by-wave in the 400 nm–900 nm range using an iterative method. The wavelength position with the highest fitting accuracy to the total suspended solids concentration (TSM) was ultimately determined as the optimal first wavelength for the absorption coefficient. Optimal second wavelength and the optimal third wavelength ; Correspondingly, organic suspended matter components Characterized as: , The optimal first wavelength The corresponding absorption coefficient; Assigning weights to the scattering channels to reflect the importance of the backscattering coefficient. The importance of the absorption coefficient assigns weight to the absorption channel. The process for determining its importance in the estimation of total suspended solids concentration is as follows: The concentrations of inorganic particulate matter (IP) and organic particulate matter (OP) in the water sample were measured, and their ratio was calculated. : ; when Then increase the weight of the scattering channel. The value, and reduce the absorption channel weight. The value; if Then reduce the weight of the scattering channel. The value, and increase the absorption channel weight. The value; For a pre-set ratio threshold; The weights of the scattering channels are calculated using the weighting formula. and absorption channel weights Perform weight assignment: ;in, , This is a monotonic function obtained by fitting the suspended solids concentration. The remote sensing estimation model for total suspended solids concentration The expression is as follows: ,in, For scattering channel weights, This is for absorbing channel weights.

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

3. The method for remote sensing inversion of suspended solids concentration in complex water bodies according to claim 1, characterized in that, The backscattering coefficient of the particulate matter The parsing process is as follows: The total backscattering coefficient was calculated using remote sensing reflectance and water body characteristics. ; Obtain pure water at wavelength Backscattering coefficient The backscattering coefficient of particulate matter is calculated using the following formula. : 。 4. The method for remote sensing inversion of suspended solids concentration in complex water bodies according to claim 1, 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 Second wavelength The initial value, the third wavelength The variation range is 400nm~900nm, and the third wavelength As a variable, the absorption coefficient is calculated. The optimal third wavelength is determined by matching the total suspended solids concentration (TSM) with the third wavelength position that yields the highest fitting accuracy. ; Set the first wavelength and the optimal third wavelength As the initial value, the second wavelength The variation range is 400nm~900nm, and the second wavelength As a variable, the absorption coefficient is calculated. The optimal second wavelength was determined by setting the position corresponding to the third wavelength with the highest fitting accuracy to the total suspended solids (TSM) concentration. ; Setting the optimal second wavelength and the optimal third wavelength As the initial value, the first wavelength The variation range is 400nm~900nm, with the first wavelength... As a variable, the absorption coefficient is calculated. The optimal first wavelength was determined by setting the position corresponding to the third wavelength with the highest fitting accuracy to the total suspended solids (TSM) concentration. .

5. A system for remotely sensing and inverting the concentration of suspended solids in complex water bodies, used to implement the method as described in any one of claims 1 to 4, characterized in that, include: The first module is configured to acquire and integrate remote sensing imagery and ground observation data to obtain synchronized sample point pairs; The second module is configured to use synchronized sample pairs to obtain the backscattering coefficient of particulate matter based on water body analysis. and absorption coefficient And determine the backscattering coefficient and absorption coefficient at the optimal wavelength position that reflect the total suspended matter concentration, respectively. The third module is configured to characterize the inorganic suspended matter components using the backscattering coefficient at the optimal wavelength position. ; The absorption coefficient at the optimal wavelength position is used to characterize the components of organic suspended matter. ; The fourth module is configured to calculate the ratio R of inorganic and organic particulate matter in the water body, and determine the backscattering coefficient based on the ratio R. and absorption coefficient Its importance in estimating total suspended solids concentration; The fifth module is configured to comprehensively characterize the inorganic suspended matter components. Characterization of organic suspended matter components Based on the aforementioned importance, a remote sensing estimation model for total suspended solids concentration in complex water bodies is constructed. The total suspended solids concentration in the complex water body was estimated using a remote sensing model. Output the concentration of suspended solids in water and apply it to remote sensing images.

Images