Remote Sensing Estimation Method, System, Medium and Device for CDOM Absorption Spectrum

Abstract

The present invention discloses a remote sensing estimation method, system, medium and device for the CDOM absorption spectrum. The method includes: obtaining the measured remote sensing reflectance spectra of several points on the water surface of the estuary during the flood season, and the CDOM absorption spectra of the water samples at the corresponding points; calculating the maximum correlation coefficient between the measured remote sensing reflectance spectra in the first band range and the CDOM absorption spectra in the second band range, calculating the spectral gradient of the measured remote sensing reflectance spectra in the first band range as the reflectance short-wave spectral gradient, calculating the spectral slope of the CDOM absorption spectra in the second band range as the CDOM short-wave spectral slope, calculating the spectral slope of the CDOM absorption spectra in the third band range as the CDOM long-wave spectral slope, and establishing an inversion model for the CDOM absorption spectrum; obtaining the spectral satellite remote sensing reflectance image of the estuary during the flood season, and inputting each pixel into the model to obtain the spatial distribution of the CDOM absorption spectrum. Using the present invention can significantly improve the inversion accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of water quality remote sensing inversion, and more specifically, to a remote sensing estimation method, system, medium and device for the absorption spectrum of CDOM. Background Art

[0002] Chromophoric Dissolved Organic Matter (CDOM) is the light-absorbing part of Dissolved Organic Matter (DOM) in water, which has a significant light-absorbing effect in the ultraviolet region (UV, 250 nm - 400 nm). It can affect the light field in the upper water layer, cause light attenuation in the ultraviolet and blue light ranges in water, trigger various photochemical reactions and the release of inorganic carbon, and change the microbial utilization of organic matter. Due to its strong chemical, biological and optical activities at the same time, its absorption spectrum is currently widely used as a tracer for the composition, source and activity of DOM in water, which has important enlightenment for understanding various processes of the marine biogeochemical system and the global carbon cycle.

[0003] In water quality remote sensing inversion, due to its light-absorbing ability in the ultraviolet-visible range, CDOM is considered to be one of the main light-active components in water, and together with chlorophyll and suspended particulate matter, it determines the water surface reflectance. The optical properties of open ocean waters (type I waters) are mainly dominated by plankton. CDOM and suspended particulate matter are both products of plankton, and their absorption coefficients co-vary with the chlorophyll concentration. For coastal and inland waters (type II waters), the main sources of CDOM and suspended solids are terrigenous inputs, and their concentrations vary relatively independently of the chlorophyll concentration. In this case, remote sensing inversion algorithms need to consider more unknown variables, which greatly increases the uncertainty of CDOM remote sensing inversion.

[0004] During the flood season, the runoff in the estuary increases due to frequent heavy rainfall. On the one hand, a large amount of terrigenous CDOM and suspended particulate matter are transported, and on the other hand, it is difficult for chlorophyll to accumulate in the estuary. CDOM and suspended particulate matter together become the dominant factors of the optical characteristics of estuarine waters. Since the absorption spectra of CDOM and suspended particulate matter are similar, both decay exponentially with the increase of wavelength, and their absorption intensities are usually comparable. To accurately estimate the CDOM absorption spectrum from the remote sensing reflectance, it is particularly crucial to eliminate the interference of suspended particulate matter.

[0005] In the existing inversion algorithms for type-II water bodies, whether it is a semi-analytical model based on physical mechanisms or a statistical model based on empirical relationships, the absorption coefficient of CDOM at 412 nm or 443 nm is used as the inversion target, and the remote sensing reflectance below 500 nm is used as the input parameter. The selection of these two wavelength ranges can neither effectively distinguish the absorption effects of CDOM and suspended particulate matter, nor can the inversion algorithm achieve a high inversion accuracy due to the low absorption intensity of CDOM above 400 nm and the noise interference introduced by the low signal-to-noise ratio of the remote sensing reflectance below 500 nm.

[0006] The above information disclosed in the background art section is only used to enhance the understanding of the background of the present invention, and thus may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0007] In view of this, the present invention provides a remote sensing estimation method, system, medium and device for the absorption spectrum of CDOM that solves at least the above partial technical problems, which can truly invert the spatial distribution of CDOM in the estuary during the flood season and help improve the inversion accuracy.

[0008] To achieve the above object, the technical solution adopted by the present invention is as follows:

[0009] In the first aspect, the present invention provides a remote sensing estimation method for the absorption spectrum of CDOM, and the method includes the following steps:

[0010] Step 1: Obtain the measured remote sensing reflectance spectra of several points on the water surface of the estuary during the flood season, and obtain the absorption spectra of CDOM of the water samples at the corresponding points;

[0011] Step 2: Calculate the maximum correlation coefficient between the measured remote sensing reflectance spectrum in the first wavelength range and the CDOM absorption spectrum in the second wavelength range, and record the CDOM absorption wavelength and the remote sensing reflectance wavelength under the maximum correlation coefficient, which are respectively used as the optimal band for the CDOM absorption coefficient and the optimal band for the reflectance;

[0012] Step 3: Calculate the spectral gradient of the measured remote sensing reflectance spectrum in the first wavelength range, and the spectral slopes of the CDOM absorption spectrum in the second wavelength range and the third wavelength range, which are respectively used as the short-wave spectral gradient of the reflectance, the short-wave spectral slope of CDOM and the long-wave spectral slope of CDOM;

[0013] Step 4: Based on the reflectance at the optimal band, the CDOM absorption coefficient at the optimal band, the short-wave spectral gradient of the reflectance, the short-wave spectral slope of CDOM and the long-wave spectral slope of CDOM, construct an inversion model for the CDOM absorption spectrum;

[0014] Step 5: Obtain the spectral satellite remote sensing reflectance image of the estuary during the flood season, select the satellite remote sensing reflectance corresponding to the optimal reflectance band, and input it into the CDOM absorption spectrum inversion model to obtain the spatial distribution map of the CDOM absorption spectrum.

[0015] Further, the first band range is 400nm - 600nm, the second band range is 250nm - 400nm, and the third band range is 250nm - 700nm.

[0016] Further, in the said Step 2, the calculation formula of the correlation coefficient is:

[0017]

[0018] Among them, represents the correlation coefficient, X represents the measured remote sensing reflectance at each wavelength within the first band range, and Y represents the CDOM absorption coefficient at each wavelength within the second band range; represents the average value of the measured remote sensing reflectance at each wavelength within the first band range at all sampling points; represents the average value of the CDOM absorption coefficient at each wavelength within the second band range at all sampling points; calculate the correlation coefficient one by one for the measured remote sensing reflectance at each wavelength within the first band range and the CDOM absorption coefficient at each wavelength within the second band range to obtain a correlation coefficient matrix, and calculate the maximum value and its row and column numbers of the elements of the correlation coefficient matrix.

[0019] Further, in the said Step 3, the short - wave spectral gradient of the reflectance is the growth rate of the measured remote sensing reflectance spectrum from the trough to the peak within the first band range, and the calculation formula is as follows:

[0020]

[0021] Among them, represents the short - wave spectral gradient of the reflectance, and respectively represent the maximum value and the minimum value of the measured remote sensing reflectance spectrum within the first band range, and respectively represent and the corresponding wavelengths.

[0022] Further, in the said Step 3, the calculation formula of the CDOM spectral slope is:

[0023]

[0024] Among them, λ represents any wavelength within the second band range or the third band range, represents the absorption spectrum of CDOM at the wavelength λ, Represents the absorption spectrum of CDOM at the optimal absorption coefficient band λ0, exp represents the exponential relationship, S g represents the spectral slope of CDOM;

[0025] Respectively, fit the CDOM absorption spectra in the second band range and the third band range according to the calculation formula to obtain the short-wave spectral slope of CDOM and the long-wave spectral slope of CDOM .

[0026] Furthermore, in the step 4, the constructed CDOM absorption spectrum inversion model includes:

[0027]

[0028]

[0029]

[0030]

[0031] Among them, represents the remote sensing reflectance at the optimal band, λ1 represents the optimal reflectance band, log is the logarithmic relationship, and A, B, C, D, E, F are model parameters obtained through data fitting.

[0032] Furthermore, in the step 5, the spectral satellite remote sensing reflectance is the water body reflectance obtained by a multispectral or hyperspectral remote sensing satellite, and has at least 3 bands within the first band range.

[0033] In a second aspect, the present invention also provides a remote sensing estimation system for the CDOM absorption spectrum, which is applied to the above-mentioned remote sensing estimation method for the CDOM absorption spectrum to perform CDOM absorption spectrum inversion. The system includes:

[0034] A spectral acquisition module, configured to acquire the measured remote sensing reflectance spectra of several points on the water surface of the estuary during the flood season, and acquire the CDOM absorption spectra of the water body samples at the corresponding points;

[0035] A calculation module, configured to calculate the maximum correlation coefficient between the measured remote sensing reflectance spectrum in the first band range and the CDOM absorption spectrum in the second band range, and record the CDOM absorption wavelength and the remote sensing reflectance wavelength under the maximum correlation coefficient, respectively, as the optimal CDOM absorption coefficient band and the optimal reflectance band;

[0036] And calculate the spectral gradient of the measured remote sensing reflectance spectrum in the first band range, and the spectral slopes of the CDOM absorption spectrum in the second band range and the third band range, as the short-wave spectral gradient of reflectance, the short-wave spectral slope of CDOM, and the long-wave spectral slope of CDOM, respectively;

[0037] A model construction module for constructing an inversion model of the CDOM absorption spectrum based on the optimal band reflectance, the optimal band CDOM absorption coefficient, the short-wave spectral gradient of reflectance, the short-wave spectral slope of CDOM, and the long-wave spectral slope of CDOM;

[0038] A spatial distribution generation module for taking the spectral satellite remote sensing reflectance image of the estuary during the flood season, selecting the satellite remote sensing reflectance corresponding to the optimal band of reflectance, and inputting it into the CDOM absorption spectrum inversion model to obtain the spatial distribution map of the CDOM absorption spectrum.

[0039] In a third aspect, the present invention also provides a storage medium, on which one or more programs readable by a computing device are stored. The one or more programs include instructions that, when executed by the computing device, cause the computing device to execute a remote sensing estimation method of a CDOM absorption spectrum as described above.

[0040] In a fourth aspect, an embodiment of the present invention also provides an electronic device, including a processor and a memory. The memory stores machine-executable instructions that can be executed by the processor, and the processor executes the machine-executable instructions to implement a remote sensing estimation method of a CDOM absorption spectrum that can execute the above.

[0041] Compared with the prior art, the present invention has at least the following beneficial technical effects:

[0042] 1. The present invention solves the problems of confusion interference caused by similar absorption characteristics of suspended particulate matter in the existing CDOM remote sensing inversion algorithm, and noise interference caused by low signal-to-noise ratio of the CDOM absorption coefficient and remote sensing reflectance, and improves the inversion accuracy.

[0043] 2. The present invention adopts an inversion strategy from the short-wave range to the long-wave range. Based on the remote sensing reflectance above 500 nm, the absorption coefficient and spectral slope of CDOM in the short-wave range can be inverted. Then, using the autocorrelation relationship of the CDOM absorption spectrum, the spectral slope of CDOM is extrapolated from the short-wave range to the long-wave range, and thus the absorption spectrum of CDOM in the full wavelength range is obtained. Since suspended particulate matter has no absorption characteristics in the short-wave band, this strategy can effectively avoid the confusion interference of suspended particulate matter. This strategy significantly reduces the use of short-wave band remote sensing reflectance and long-wave band CDOM absorption coefficient during modeling, effectively avoiding the noise interference caused by the low absorption of CDOM in the long-wave band and the low signal-to-noise ratio of remote sensing reflectance in the short-wave band, and can significantly improve the remote sensing inversion accuracy of the CDOM absorption spectrum.

[0044] Other features and advantages of the present invention will be described in the following specification, and in part will be obvious from the specification, or will be understood by implementing the present invention. The objectives and other advantages of the present invention can be achieved and obtained through the structures specifically pointed out in the written specification and the drawings.

[0045] The technical solutions of the present invention will be further described in detail below through the drawings and embodiments. Description of the Drawings

[0046] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0047] The drawings are used to provide a further understanding of the present invention, and constitute a part of the specification. They are used together with the embodiments of the present invention to explain the present invention, and do not constitute a limitation to the present invention.

[0048] Figure 1 It is a schematic flow chart of a remote sensing estimation method for the CDOM absorption spectrum provided by an embodiment of the present invention.

[0049] Figure 2 It is a schematic diagram of the location distribution of sampling points for the water surface remote sensing experiment provided by an embodiment of the present invention.

[0050] Figure 3 It is a schematic diagram of the spatial distribution map of the CDOM absorption spectrum obtained by satellite remote sensing inversion provided by an embodiment of the present invention.

[0051] Figure 4 It is a schematic diagram of the spatial distribution map of the CDOM absorption spectrum measured at the estuary during the flood season provided by an embodiment of the present invention.

[0052] Figure 5 Schematic diagram of the mean absolute percentage error (MAPD) and root mean square error (RMSE) of the inversion results of the per-wavelength CDOM absorption coefficient (a g ) provided by the embodiments of the present invention.

[0053] Figure 6 Schematic diagram of the comparison between the measured value and the inversion value of the full-band CDOM absorption coefficient of the water body sample provided by the embodiments of the present invention.

[0054] Figure 7 Schematic diagram of the structure of the electronic device provided by the embodiments of the present invention. Detailed implementation manners

[0055] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some but not all of the embodiments of the present invention.

[0056] In the description of the present invention, it should be noted that in some processes described in the specification and drawings of this application, there are multiple operations that appear in a specific order, but it should be clearly understood that these operations may not be executed in the order in which they appear herein or may be executed in parallel. In addition, various serial numbers, etc. are only for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0057] Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed present invention, but merely represents selected embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0058] Embodiment 1:

[0059] The embodiments of the present invention provide a remote sensing estimation method for CDOM absorption spectra, which can solve the confusion interference caused by the similar absorption characteristics of suspended particulate matter in the existing CDOM remote sensing inversion algorithm and the noise interference caused by the low signal-to-noise ratio of the CDOM absorption coefficient and the remote sensing reflectance in the range of 400-500 nm, and help to significantly improve the inversion accuracy. The general process of this method is as Figure 1 shown, and mainly includes the following steps:

[0060] Step 1, obtain the measured remote sensing reflectance spectra of several points on the estuary water surface during the flood season, and obtain the CDOM absorption spectra of the water body samples at the corresponding points;

[0061] Step 2, calculate the maximum correlation coefficient between the measured remote sensing reflectance spectrum in the first band range and the CDOM absorption spectrum in the second band range, record the CDOM absorption wavelength corresponding to the maximum correlation coefficient as the optimal band λ0 of the CDOM absorption coefficient, and record the remote sensing reflectance wavelength corresponding to the maximum correlation coefficient as the optimal band λ1 of the reflectance;

[0062] Step 3, calculate the spectral gradient of the measured remote sensing reflectance spectrum in the first band range as the short-wave spectral gradient R rs _Gradient (first band range), calculate the spectral slope of the CDOM absorption spectrum in the second band range as the short-wave spectral slope S g (second band range), calculate the spectral slope of the CDOM absorption spectrum in the third band range as the long-wave spectral slope S g (third band range), the third band range is greater than the second band range;

[0063] Step 4, based on the CDOM absorption coefficient of the optimal band 、the remote sensing reflectance of the optimal band 、the short-wave spectral gradient R rs _Gradient (first band range), the short-wave spectral slope S g (second band range) and the long-wave spectral slope S g (third band range), use the data fitting method to establish an inversion model of the CDOM absorption spectrum. Among them, based on the remote sensing reflectance of the optimal band establish a fitting model of the CDOM absorption coefficient of the optimal band , based on the short-wave spectral gradient R rs _Gradient (first band range) to establish a fitting model of the short-wave spectral slope S g (second band range), based on the short-wave spectral slope S g (second band range) to establish a fitting model of the long-wave spectral slope S g (third band range), based on the CDOM absorption coefficient of the optimal band and the long-wave spectral slope S g (third band range) to restore the absorption coefficient of CDOM in the third band range using an exponential relationship;

[0064] Step 5, obtain the multi-spectral or hyperspectral satellite remote sensing reflectance image of the estuary during the flood season, select the satellite remote sensing reflectance image corresponding to the optimal band λ1 of the reflectance, and input the remote sensing reflectance of all pixels in the satellite remote sensing reflectance image into the CDOM absorption spectrum inversion model one by one to obtain the spatial distribution map of the CDOM absorption spectrum of the estuary during the flood season.

[0065] In a specific embodiment, the first wavelength band range is 400 nm - 600 nm, the second wavelength band range is 250 nm - 400 nm, and the third wavelength band range is 250 nm - 700 nm. The first wavelength band range of 400 nm - 600 nm is the optimal wavelength band range for the remote sensing reflectance of the water surface used for the remote sensing inversion of the CDOM absorption spectrum. The second wavelength band range of 250 nm - 400 nm is the optimal wavelength band range for the CDOM absorption spectrum that can be inverted through the remote sensing reflectance of the water surface. The reason for the above selection is that through the correlation analysis of experimental data and the analysis of the physical and chemical properties of CDOM, the correlation between the remote sensing reflectance of the water surface and the CDOM absorption spectrum is the strongest within these two wavelength band ranges, and the data signal-to-noise ratio is high, the spatial difference degree is significant, and the inversion accuracy is high. The third wavelength band range of 250 nm - 700 nm is the entire wavelength band range where the CDOM has spectral characteristics presented, and it is also the general wavelength band range for the conventional measurement experiment of the CDOM spectral characteristics.

[0066] In a specific embodiment, the short-wave spectral gradient R rs _Gradient (first wavelength band range) is the growth rate of the measured remote sensing reflectance spectrum from the trough to the peak within the first wavelength band range, and the calculation formula is as follows:

[0067]

[0068] Among them, R rs _Gradient represents the short-wave spectral gradient of the reflectance, and respectively represent the maximum value and the minimum value of the measured remote sensing reflectance spectrum within the first wavelength band range, and respectively represent and the corresponding wavelengths.

[0069] In a specific embodiment, the calculation formula for the CDOM spectral slope is:

[0070]

[0071] Among them, λ is any wavelength within the second wavelength band range or the third wavelength band range, is the absorption spectrum of CDOM at the wavelength λ, is the absorption spectrum of CDOM at the optimal absorption coefficient wavelength λ0, exp is the exponential relationship, and S g is the CDOM spectral slope. The CDOM absorption spectra of the second wavelength band range and the third wavelength band range are respectively fitted with data according to the calculation formula to obtain the CDOM short-wave spectral slope S g (second wavelength band range) and the CDOM long-wave spectral slope Sg (Third band range).

[0072] In a specific embodiment, the constructed CDOM absorption spectrum inversion model is:

[0073]

[0074]

[0075]

[0076]

[0077] Wherein, log represents the logarithmic relationship, and A, B, C, D, E, and F are model parameters obtained by data fitting in step 3.

[0078] In a specific embodiment, the spectral satellite remote sensing reflectance is the water body reflectance obtained by a multispectral or hyperspectral remote sensing satellite, and has at least 3 bands within the first band range.

[0079] The following combines Figures 1 to 6 As shown, the embodiments and working principles of the method of the present invention will be introduced in detail:

[0080] This embodiment proposes a remote sensing estimation method for the absorption spectrum of colored dissolved organic matter in the river mouth water body during the flood season. It adopts an inversion strategy from ultraviolet to visible light, that is, first, based on the remote sensing reflectance above 500 nm, the absorption coefficient and spectral slope of CDOM in the ultraviolet band (250 nm - 400 nm) are inverted, and then the spectral slope of CDOM is extrapolated from the ultraviolet band (250 nm - 400 nm) to the visible light band (250 nm - 700 nm) by using the autocorrelation relationship of the CDOM absorption spectrum, so as to obtain the absorption spectrum of CDOM in the full band range. Since suspended particulate matter has no absorption characteristics in the ultraviolet band, this strategy can effectively avoid the interference of suspended particulate matter. This strategy significantly reduces the use of remote sensing reflectance below 500 nm and the CDOM absorption coefficient in the visible light band (400 nm - 700 nm) during modeling, effectively avoiding the noise interference caused by the low absorption effect of CDOM in the visible light band and the low signal-to-noise ratio of remote sensing reflectance below 500 nm, and can significantly improve the remote sensing inversion accuracy of the CDOM absorption spectrum. As Figure 1 shown, the method includes the following steps S1 to S5:

[0081] S1, Obtain the measured remote sensing reflectance spectra of several points on the water surface of the river mouth during the flood season, and obtain the CDOM absorption spectra of the water body samples at the corresponding points.

[0082] In this embodiment, as Figure 2As shown in the figure, 56 points were selected at a certain river estuary, and a water surface remote sensing experiment was carried out during the flood season (May). The remote sensing reflectance spectrum above the water surface in the range of 400 - 900 nm was obtained through waterborne radiation measurement; and at the same time, water samples were collected and filtered, and the CDOM absorption spectrum in the range of 250 nm - 700 nm was measured in the laboratory.

[0083] S2, calculate the maximum correlation coefficient between the measured remote sensing reflectance spectrum in the range of 400 nm - 600 nm and the CDOM absorption spectrum in the range of 250 nm - 400 nm, and record the CDOM absorption wavelength and the remote sensing reflectance wavelength under the maximum correlation coefficient, which are respectively denoted as the optimal CDOM absorption band λ0 and the optimal reflectance band λ1.

[0084] In this embodiment, for the measured remote sensing reflectance spectrum in the range of 400 nm - 600 nm and the CDOM absorption spectrum in the range of 250 nm - 400 nm, the Pearson correlation coefficient was calculated band by band, and a correlation coefficient matrix with a dimension of 201 * 151 was generated. The maximum value of the correlation coefficient matrix was obtained, and the maximum correlation coefficient was 0.86. The row and column numbers corresponding to 0.86 were 197 and 41 respectively. Then, the CDOM absorption wavelength λ0 at this correlation coefficient was 290 nm, and the remote sensing reflectance wavelength λ1 was 596 nm.

[0085] S3, calculate the spectral gradient of the measured remote sensing reflectance spectrum in the range of 400 nm - 600 nm, and the spectral slopes of the CDOM absorption spectrum in the ranges of 250 nm - 400 nm and 250 nm - 700 nm, which are respectively defined as the short - wave spectral gradient of reflectance, the short - wave spectral slope of CDOM, and the long - wave spectral slope of CDOM.

[0086] The short - wave spectral gradient of reflectance is the growth rate of the measured remote sensing reflectance spectrum from any wavelength in the range of 400 - 450 nm to the peak in the range of 400 nm - 600 nm, and is calculated according to the following formula:

[0087] (1)

[0088] Among them, R rs _Gradient represents the short - wave spectral gradient of reflectance, and respectively represent the maximum value and the minimum value of the measured remote sensing reflectance spectrum in the first band range, and respectively represent and the corresponding wavelengths.

[0089] In this embodiment, since there is relatively large signal noise in the measured remote sensing reflectance at 380 nm - 420 nm, the lower edge wavelength is set to 420 nm, that is, λ(min) = 420 nm, R rs(min) = R rs (420). Then search for R within the range of 420 - 600 nm rs(max) , and the search results show that the actual peak positions of 56 points are basically concentrated around 580 nm.

[0090] The CDOM spectral slope is obtained by fitting the CDOM absorption spectra in the ranges of 250 nm - 400 nm and 250 nm - 700 nm to the following formula respectively:

[0091] (2)

[0092] Among them, λ0 is set as the optimal band of the CDOM absorption coefficient, that is, 290 nm.

[0093] S4. Based on the optimal band of reflectance, the optimal band of CDOM absorption coefficient, the short - wave spectral gradient of reflectance, the short - wave spectral slope of CDOM, and the long - wave spectral slope of CDOM, establish an inversion model for the CDOM absorption spectrum.

[0094] First, based on the remotely sensed reflectance at the optimal band (Rrs(596)), perform data fitting on . The fitting results show that the linear fitting has the highest fitting accuracy (its determination coefficient can reach 0.74). Therefore, the remotely sensed inversion relationship for the optimal band of the CDOM absorption coefficient is established as:

[0095] (3)

[0096] Secondly, based on the short - wave spectral gradient of reflectance (R rs _Gradient), perform data fitting on . The fitting results show that the power - law relationship has the highest fitting accuracy (its determination coefficient is 0.53). Therefore, the remotely sensed inversion relationship for the short - wave spectral slope of CDOM is established as:

[0097] (4)

[0098] Thirdly, based on the short - wave spectral slope of CDOM ( ), perform data fitting on . The fitting results show that the logarithmic relationship has the highest fitting accuracy (its determination coefficient is 0.97). Therefore, the extrapolation relationship for the long - wave spectral slope of CDOM is established as:

[0099] (5)

[0100] Finally, restore the CDOM absorption spectrum in the range of 250 - 700 nm through the CDOM exponential relationship:

[0101] (6)

[0102] In this embodiment, the absorption coefficient of CDOM in the ultraviolet band shows a positive correlation with the reflectance in the visible light band, while the short-wave spectral slope of CDOM shows a negative correlation with the short-wave spectral gradient of the reflectance. This is because CDOM has fluorescence characteristics and can be excited under the irradiation of solar ultraviolet radiation and emit fluorescence in the visible light range. As the excitation wavelength increases, the maximum radiation of CDOM fluorescence will gradually increase. The increase in the excitation wavelength means that the band range of the strong absorption of CDOM increases, and the spectral slope within this band range decreases accordingly. At the same time, the increase in CDOM fluorescence radiation will cause the remote sensing reflectance of the water body to increase in the visible light band, and the spectral gradient increases accordingly.

[0103] S5. Obtain the multi-spectral or hyperspectral satellite remote sensing reflectance images of the estuary during the flood season, select the satellite remote sensing reflectance image corresponding to the optimal reflectance band, and input all the pixels of the image into the CDOM absorption spectrum inversion model one by one to obtain the spatial distribution map of the CDOM absorption spectrum of the estuary.

[0104] In this embodiment, four types of multi-spectral or hyperspectral satellite data are obtained and the inversion model is used for spatial distribution remote sensing generation. The satellite remote sensing images include Suomi NPP VIIRS, Sentinel-3A OLCI, ISS HICO, and Landsat 8 OLI data products (satellite and payload parameters are shown in Table 1). The spatial resolution of all satellite data is better than 1 km, and for estuaries with an area less than 2000 km 2 it is a better choice than the older generation satellite data MODIS and SeaWiFS.

[0105] Table 1 Multi-spectral or hyperspectral satellite payload parameters used in this embodiment

[0106]

[0107] Select images with a cloud cover of less than 20% in the flood season (April and May) from 2012 to 2018 and in the research area of 22°-23°N, 113.5°-114.5°E for download to match the seasons with the measured data. For the Landsat 8 OLI data with a 16-day revisit cycle, the time range is extended to March to June.

[0108] The atmospheric correction and normalized surface reflectance calculation of OLCI 1B data were performed using the Class II Water Region / Coastal Color Processing Module (C2RCC) of SNAP. The atmospheric correction and surface reflectance calculation of HICO 1B data were performed using the ENVI 5.2 FLAASH module. All remote sensing reflectances were normalized. VIIRS, OLCI, and HICO data were geometrically corrected using ENVI 5.2. The near-infrared reflectance threshold method was used to extract the water area, and the threshold was set to 10%, 15%, or 20% depending on the specific situation. The NIR Black algorithm was used to further eliminate the influence of the atmosphere and the water surface. Finally, pixels with negative reflectance in the visible light band were removed.

[0109] After preprocessing, the CDOM absorption coefficient and spectral slope are remotely sensed and inverted using formulas (1), (3)-(5). rs (596) According to the strategies listed in Table 2, the standard band values ​​of the four sensors are respectively corresponded. The remote sensing reflectance of all pixels of the standard band images of the four sensors or their algebraic weighted average images is input into formula (3) pixel by pixel to calculate the corresponding reflectance of each image pixel. .

[0110] Table 2 R rs (596) Correspondence strategy with the standard band values ​​of four sensors

[0111]

[0112] In the embodiment of the present invention, in the inversion S g (250-400), taking into account the available spectral range of each satellite data and the atmospheric correction effect, for VIIRS, HICO, OLCI and OLI data, the calculation of R rs _Gradient lower edge λ min Set to 445nm, 415.5nm, 400nm and 443nm respectively. The upper edge of the band of all sensors λ max The reflectance spectrum slope of each pixel in each remote sensing image is calculated according to formula (1), and on this basis, the S of each pixel in each remote sensing image is calculated according to formulas (4)-(5). g (250-400) and S g (250-700), and then calculate the total absorption coefficient of CDOM in the range of 250nm-700nm for each pixel in each remote sensing image according to formula (6). Each pixel value is arranged according to longitude and latitude, and the spatial distribution map of the CDOM absorption coefficient of the estuary area is obtained.

[0113] Figure 3Shows a obtained by inverting four types of multispectral or hyperspectral satellite remote sensing data g (290) and S g (250 - 400), and each figure represents a obtained by inverting images from different sensors at different times g (290) and S g (250 - 400), with different colors representing the magnitudes of a g (290) and S g (250 - 400), with the units being m -1 and nm -1 , where blue represents the minimum value and red represents the maximum value. It can be seen that the inversion results of all satellite images well reflect the main distribution patterns of a g (290) and S g (250 - 400) in the estuary during the flood season (as Figure 4 shown): The upper reaches and the west side of the estuary have higher a g (290) and lower S g (250 - 400), while the opposite is true in the eastern estuary and its adjacent offshore waters. Since this method adopts an inversion strategy from ultraviolet to visible light, that is, first based on the remote sensing reflectance above 500 nm, the absorption coefficient and spectral slope of CDOM in the ultraviolet band (250 nm - 400 nm) are inverted, and then using the autocorrelation relationship of the CDOM absorption spectrum, the CDOM spectral slope is extrapolated from the ultraviolet band (250 nm - 400 nm) to the visible light band (250 nm - 700 nm), thereby obtaining the absorption spectrum of CDOM in the full wavelength range. Since suspended particulate matter has no absorption characteristics in the ultraviolet band, this strategy can effectively avoid the interference of suspended particulate matter. This strategy significantly reduces the use of remote sensing reflectance below 500 nm and the CDOM absorption coefficient in the visible light band (400 nm - 700 nm) during modeling, effectively avoiding the noise interference caused by the low absorption of CDOM in the visible light band and the low signal-to-noise ratio of remote sensing reflectance below 500 nm, and significantly improving the ability of satellite remote sensing to depict the spatial distribution characteristics of CDOM in the estuary.

[0114] In this embodiment, taking the water surface remote sensing experiment and satellite remote sensing inversion application in a certain estuary area as an example, it transports more than 87 million tons of suspended particulate matter (mainly mineral particles) and 30 million tons of dissolved substances to the northern South China Sea every year. The optical properties of the water body in this estuary are significantly affected by the river runoff, especially during the flood season. Multiple bio-optical measurements of the water body in this estuary found that its annual average chlorophyll concentration is generally below 5 mg / m 3 , while the concentration of suspended mineral particles during the flood season can reach 50 - 100 mg / L. Based on this estimation, the average level of the absorption coefficient a ph (440) of phytoplankton at 440 nm is usually below 0.1 m-1 and the absorption coefficient a of suspended mineral particles at 440 nm d (440) is greater than 0.5 m -1 and can reach 1 m during the flood season -1 ; meanwhile, the absorption coefficient a of CDOM at 440 nm g (440) is also much greater than a ph (440), with an average level of 0.3 - 0.5 m -1 . In this case, chlorophyll is no longer the main component determining the optical characteristics of the water body in this area, and suspended particulate matter and CDOM from land sources become the dominant factors affecting the water color of this estuary.

[0115] Application and verification of technical effects:

[0116] The application and verification of technical effects are implemented in two ways: model accuracy evaluation and satellite remote sensing inversion application verification. Specifically as follows:

[0117] Inversion accuracy evaluation: It is mainly estimated through three statistics, including mean absolute percentage error (MAPD), root mean square error (RMSE), and coefficient of determination (R 2 ), and the calculation formulas are listed as follows:

[0118] (7)

[0119] (8)

[0120] (9)

[0121] Among them, C alg and C insitu are the inversion value and the measured value respectively; N represents the number of sampling points.

[0122] Due to the limited measured values, the K-fold cross-validation method is used to evaluate the accuracy of the algorithm respectively, and the mean value of the statistics obtained from K-fold cross-validation is used to measure the performance of the algorithm. Using 54 measured values and setting the k value to 6, so each group has 9 observation values. The cross-validation results (Table 3) show that the average values of the mean absolute percentage error (MAPD) of a g (290), S g (250 - 400) and S g (250 - 700) are ~31%, ~5.3% and ~6% respectively, and the 6 groups of cross-validation results are similar, reflecting good model generalization ability.

[0123] Table 3 k-fold cross-validation results

[0124]

[0125] According to the inversion of a g (290) and S g (250-700) Reconstruction of the CDOM absorption spectrum in the range of 250-700 nm. Figure 5 The wavelength-by-wavelength CDOM absorption coefficient (a g ), the horizontal axis represents the wavelength in nm, the left vertical axis and the blue line represent the mean absolute percentage error (MAPD) ​​in percentage, and the right vertical axis and the green line represent the root mean square error (RMSE) in m -1 The mean absolute percentage error (MAPD) ​​is stable at about 30% below 450nm, and increases with increasing wavelength above 450nm, mainly affected by the low measurement signal-to-noise ratio caused by the reduced absorption of CDOM in the long-wavelength band; the root mean square error (RMSE) is consistent with the trend of the CDOM absorption coefficient spectral curve, and decreases exponentially with increasing wavelength. Figure 6 The comparison between the measured and inverted values ​​of the full-band CDOM absorption coefficient of all water samples is shown. The horizontal axis is all sampling points, the vertical axis is the wavelength, the unit is nm, and the different colors represent the size of the CDOM absorption coefficient, the unit is m -1 , blue represents the minimum absorption coefficient, and red represents the maximum absorption coefficient. It can be seen that the distribution of the true value and the inversion value is highly consistent. Figure 5 and Figure 6 The accuracy analysis results show that the inversion accuracy of this inversion model for the CDOM absorption coefficient in the strong absorption band (250nm-500nm) can be stabilized at around 70%, demonstrating a relatively high inversion accuracy.

[0126] Satellite remote sensing inversion verification: Due to the lack of matching data between field observations and satellite observations on the same day, images collected by different sensors on the same day were selected for cross-comparison: HICO and VIIRS (13 / 04 / 20), OLI and VIIRS (2016 / 03 / 26), OLCI and VIIRS (2017 / 04 / 02). Figure 3 It can be seen that the a obtained by inversion of different sensors in the same phase g (290) and S g (250-400) has good consistency in both numerical range and distribution pattern, and is consistent with the actual distribution of CDOM in estuaries reflected by measured data (e.g. Figure 4 This shows that the inversion model is not affected by different sensor types and can truly invert the spatial distribution of CDOM in the estuary during the flood season.

[0127] From the description of the above embodiments, those skilled in the art can know that: The embodiments of the present invention provide a remote sensing estimation method for the absorption spectrum of colored soluble organic matter, which solves the problems of confusion interference caused by the similar absorption characteristics of suspended particulate matter in the existing CDOM remote sensing inversion algorithm, and the noise interference caused by the low signal-to-noise ratio of the CDOM absorption coefficient and remote sensing reflectance in the range of 400nm - 500nm, and significantly improves the inversion accuracy.

[0128] Embodiment 2:

[0129] The embodiments of the present invention also provide a remote sensing estimation system for the CDOM absorption spectrum, which is applied to a remote sensing estimation method for the CDOM absorption spectrum described in Embodiment 1 above to perform remote sensing estimation of the CDOM absorption spectrum. The system includes:

[0130] A spectrum acquisition module, configured to acquire the measured remote sensing reflectance spectra of several points on the estuary water surface during the flood season, and acquire the CDOM absorption spectra of the water samples at the corresponding points;

[0131] A calculation module, configured to calculate the maximum correlation coefficient between the measured remote sensing reflectance spectrum in the first wavelength range and the CDOM absorption spectrum in the second wavelength range, and record the CDOM absorption wavelength and the remote sensing reflectance wavelength under the maximum correlation coefficient, respectively, as the optimal band of the CDOM absorption coefficient and the optimal band of the reflectance;

[0132] And calculate the spectral gradient of the measured remote sensing reflectance spectrum in the first wavelength range, and the spectral slopes of the CDOM absorption spectrum in the second wavelength range and the third wavelength range, respectively, as the short-wave spectral gradient of the reflectance, the short-wave spectral slope of the CDOM, and the long-wave spectral slope of the CDOM;

[0133] A model construction module, configured to construct a CDOM absorption spectrum inversion model based on the optimal band reflectance, the optimal band CDOM absorption coefficient, the short-wave spectral gradient of the reflectance, the short-wave spectral slope of the CDOM, and the long-wave spectral slope of the CDOM;

[0134] A spatial distribution generation module, configured to take the spectral satellite remote sensing reflectance image of the estuary during the flood season, select the satellite remote sensing reflectance corresponding to the optimal band of the reflectance, and input it into the CDOM absorption spectrum inversion model to obtain the spatial distribution map of the CDOM absorption spectrum.

[0135] For a remote sensing estimation system for the CDOM absorption spectrum provided by the embodiments of the present invention, its implementation principle and the technical effects generated are the same as those of the foregoing method embodiments. For the sake of brief description, for the parts not mentioned in this embodiment, reference may be made to the corresponding contents in the foregoing method embodiments, and details will not be described herein again.

[0136] Embodiment 3:

[0137] An embodiment of the present invention also provides a storage medium, on which one or more programs readable by a computing device are stored. The one or more programs include instructions that, when executed by the computing device, cause the computing device to execute a remote sensing estimation method for CDOM absorption spectrum in Embodiment 1.

[0138] Examples of such computer-readable storage media include: read-only memory (ROM), programmable read-only memory (PROM), electrically erasable programmable read-only memory (EEPROM), random access memory (RAM), dynamic random access memory (DRAM), static random access memory (SRAM), flash memory, non-volatile memory, CD-ROM, DVD-ROM, Blu-ray or optical disc memory, hard disk drive (HDD), solid state drive (SSD), cartridge memory (such as multimedia card, secure digital (SD) card or extreme digital (XD) card), magnetic tape, floppy disk, magneto-optical data storage device, optical data storage device, hard disk, solid state disk, and any other device configured to store a computer program and any associated data, data files, and data structures in a non-transitory manner and provide the computer program and any associated data, data files, and data structures to a processor or computer such that the processor or computer can execute the computer program. The computer program in the above computer-readable storage medium can run in an environment deployed in computer devices such as clients, hosts, proxy devices, servers, etc. In addition, in one example, the computer program and any associated data, data files, and data structures are distributed on a networked computer system such that the computer program and any associated data, data files, and data structures are stored, accessed, and executed in a distributed manner by one or more processors or computers.

[0139] Embodiment 4:

[0140] Referring to Figure 7 As shown, an embodiment of the present invention also provides an electronic device for remote sensing estimation of CDOM absorption spectrum. The electronic device may include a processor 10, a memory 11, a communication bus 12, and a communication interface 13, and may also include a computer program stored in the memory 11 and executable on the processor 10.

[0141] Among them, in some embodiments, the processor 10 may be composed of integrated circuits. For example, it may be composed of a single packaged integrated circuit, or may be composed of multiple packaged integrated circuits with the same or different functions, including the combination of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips, etc. The processor 10 is the control core (Control Unit) of the electronic device, connecting various components of the entire electronic device through various interfaces and circuits, and by running or executing programs or modules stored in the memory 11, and calling the data stored in the memory 11, to execute various functions of the electronic device and process data.

[0142] Those skilled in the art should understand that the embodiments of the present invention can be provided as methods, systems, devices, or computer program products, etc. Therefore, the present invention can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program codes.

[0143] It should be noted that the word "comprising" does not exclude the presence of components or steps not listed in the claims. The word "a" or "an" preceding a component does not exclude the presence of a plurality of such components. The present invention can be implemented by means of hardware including several different components and by means of a suitably programmed computer.

[0144] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The same or similar parts among the various embodiments can be referred to each other.

[0145] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A remote sensing estimation method for the absorption spectrum of CDOM, characterized in that, The method includes the following steps: Step 1: Obtain the measured remote sensing reflectance spectra of several points on the water surface of the estuary during the flood season, and obtain the CDOM absorption spectra of the water samples at the corresponding points; Step 2: Calculate the maximum correlation coefficient between the measured remote sensing reflectance spectrum in the first band range and the CDOM absorption spectrum in the second band range, and record the CDOM absorption wavelength and the remote sensing reflectance wavelength at the maximum correlation coefficient, which are used as the optimal band of the CDOM absorption coefficient and the optimal band of the reflectance respectively; Step 3: Calculate the spectral gradient of the measured remote sensing reflectance spectrum in the first band range, and the spectral slopes of the CDOM absorption spectrum in the second band range and the third band range, which are used as the short-wave spectral gradient of the reflectance, the short-wave spectral slope of the CDOM, and the long-wave spectral slope of the CDOM respectively; Step 4: Based on the reflectance of the optimal band, the CDOM absorption coefficient of the optimal band, the short-wave spectral gradient of the reflectance, the short-wave spectral slope of the CDOM, and the long-wave spectral slope of the CDOM, construct an inversion model for the CDOM absorption spectrum; Step 5: Obtain the spectral satellite remote sensing reflectance image of the estuary during the flood season, select the satellite remote sensing reflectance corresponding to the optimal band of the reflectance, input it into the inversion model of the CDOM absorption spectrum, and obtain the spatial distribution map of the CDOM absorption spectrum; In the said Step 3, the calculation formula of the CDOM spectral slope is: a g ψ(λ) = a g ψ(λ0) exp(-S g (λ - λ0)) where λ represents any wavelength within the second or third wavelength band range, and a g (λ) represents the absorption spectrum of CDOM at wavelength λ, and a g (λ0) represents the absorption spectrum of CDOM at the optimal absorption coefficient wavelength band λ0, exp represents the exponential relationship, and S g represents the CDOM spectral slope; The absorption spectra of CDOM in the second wavelength range and the third wavelength range are respectively fitted according to the calculation formula to obtain the short - wavelength spectral slope S g (for the second wavelength range) and the long - wavelength spectral slope S g (for the third wavelength range).

2. The remote sensing estimation method of a CDOM absorption spectrum according to claim 1, characterized in that, The first band range is 400nm - 600nm, the second band range is 250nm - 400nm, and the third band range is 250nm - 700nm.

3. The remote sensing estimation method of CDOM absorption spectrum according to claim 2, characterized in that, In the said Step 2, the calculation formula of the correlation coefficient is: Among them, ρ X，Y represents the correlation coefficient, X represents the measured remote sensing reflectance at each wavelength within the first band range, and Y represents the CDOM absorption coefficient at each wavelength within the second band range; represents the average value of the measured remote sensing reflectance at each wavelength within the first band range at all sampling points; represents the average value of the CDOM absorption coefficient at each wavelength within the second band range at all sampling points.

4. A remote sensing estimation method for CDOM absorption spectrum according to claim 3, characterized in that In the said Step 3, the short-wave spectral gradient of the reflectance is the growth rate of the measured remote sensing reflectance spectrum from the trough to the peak in the first band range, and the calculation formula is as follows: R rs _Gradient = [R rs(max) -R rs(min) / [λ (max) -λ (min) ​ where, R rs _Gradient represents the short-wave spectral gradient of reflectance, R rs(max) and R rs(min) respectively represent the maximum and minimum values of the measured remote sensing reflectance spectrum in the first band range, λ (max) and λ (min) respectively represent the wavelengths corresponding to R rs(max) and R rs(min) respectively.

5. The remote sensing estimation method of a CDOM absorption spectrum according to claim 4, characterized in that In the said Step 4, the constructed inversion model of the CDOM absorption spectrum includes: a g (λ0) = A * R rs (λ1) + B S g (Second band range) = C * R rs _Gradient(First band range) -D S g (Third band range) = E * log(S g (Second band range)) + F a g ρ(λ) = a g (λ0) exp(-S g (in the third wavelength range) (λ - λ0)) Among them, R rs (λ1) represents the optimal band remote sensing reflectance, λ1 represents the optimal band of reflectance, log represents the logarithmic relationship, and A, B, C, D, E, and F are model parameters obtained by data fitting.

6. A remote sensing estimation method for CDOM absorption spectrum according to claim 1, characterized in that, In the said Step 5, the spectral satellite remote sensing reflectance is the water body reflectance obtained by a multi-spectral or hyperspectral remote sensing satellite, and has at least 3 bands in the first band range.

7. A remote sensing estimation system for CDOM absorption spectra, characterized in that, When applied, execute a remote sensing estimation method for a CDOM absorption spectrum as described in any one of claims 1 - 6 to perform inversion of the CDOM absorption spectrum. The system includes: A spectrum acquisition module, which is used to obtain the measured remote sensing reflectance spectra of several points on the water surface of the estuary during the flood season, and obtain the CDOM absorption spectra of the water samples at the corresponding points; A calculation module, which is used to calculate the maximum correlation coefficient between the measured remote sensing reflectance spectrum in the first band range and the CDOM absorption spectrum in the second band range, and record the CDOM absorption wavelength and the remote sensing reflectance wavelength at the maximum correlation coefficient, which are used as the optimal band of the CDOM absorption coefficient and the optimal band of the reflectance respectively; And calculate the spectral gradient of the measured remote sensing reflectance spectrum in the first band range, and the spectral slopes of the CDOM absorption spectrum in the second band range and the third band range, which are used as the short-wave spectral gradient of the reflectance, the short-wave spectral slope of the CDOM, and the long-wave spectral slope of the CDOM respectively; A model construction module, configured to construct an inversion model for CDOM absorption spectrum based on the optimal band reflectance, the optimal band CDOM absorption coefficient, the short-wave spectral gradient of reflectance, the short-wave spectral slope of CDOM, and the long-wave spectral slope of CDOM; A spatial distribution generation module, configured to obtain a spectral satellite remote sensing reflectance image of the estuary during the flood season, select the satellite remote sensing reflectance corresponding to the optimal band of reflectance, input the reflectance into the CDOM absorption spectrum inversion model, and obtain a spatial distribution map of the CDOM absorption spectrum.

8. A storage medium having one or more programs readable by a computing device stored thereon, characterized in that: The one or more programs include instructions that, when executed by a computing device, cause the computing device to execute a remote sensing estimation method for a CDOM absorption spectrum according to any one of claims 1 to 6.

9. An electronic device, characterized in that, Comprising a processor and a memory, the memory stores machine-executable instructions capable of being executed by the processor, and the processor executes the machine-executable instructions to implement a remote sensing estimation method for a CDOM absorption spectrum according to any one of claims 1-6.

Images