River flow estimation method for deducing river channel open area by using height measurement satellite

Abstract

The invention provides a river flow estimation method for deducing a river channel open area by using a height measurement satellite, which comprises the following steps of: extracting and preprocessing an optical image river width, preprocessing a radar image and extracting the river width; extracting a height measurement satellite preprocessing water level; inverting the river topography by the elevation boundary; inverting a riverway terrain by a remote sensing image; constructing a riverway terrain; and carrying out river flow inversion. According to the river flow estimation method for deducing the river channel open area by using the height measurement satellite, river flow prediction is realized by extracting river width and water level inversion river channel topography based on multi-source remote sensing satellite data and establishing a cross section-flow empirical formula.

Description

Technical Field

[0001] The invention relates to the technical field of remote sensing inversion of river flow, in particular to a river flow estimation method using an altimetry satellite to deduce a river flow area. Background Art

[0002] Existing studies have extracted river width and water level through multi-source remote sensing data, and established empirical formulas of river width-flow and water level-flow to predict river flow.

[0003] At present, river flow inversion is only carried out based on single factors such as river width and water level. The scope of application is small and it is greatly affected by river morphology. In order to improve the accuracy of river flow prediction, some studies have considered multiple factors such as river width and water level to invert river flow. Compared with single-factor flow inversion, the accuracy and applicability of multi-factor flow inversion have been improved.

[0004] These studies use empirical formulas to invert flow through river width and water level for prediction, which requires a large amount of measured flow data support. When the terrain is complex or the riverbed changes, the inverted flow has large errors.

[0005] The Chinese invention patent application with application number CN114037905A discloses a method and system for remote sensing inversion of water surface salinity in estuary areas based on satellite-ground collaboration. On the one hand, the method utilizes the ground-measured data of the estuary area with high accuracy, and uses the measured data of the synchronized ground-measured stations as the output data to build a model, thereby improving the inversion accuracy of the remote sensing inversion model. On the other hand, it improves the traditional remote sensing inversion model that only inverts water surface salinity through spectral information or indirect water quality parameters, adds geographic grid information as input parameters, and makes full use of the latitude and longitude information of the ground-measured stations and remote sensing images, as well as the spatially continuous data characteristics of remote sensing images. It is particularly suitable for complex estuary areas to achieve high-precision inversion of water surface salinity in estuary areas. This method relies on the measured data of the ground-measured stations. If the data of the ground stations is inaccurate or incomplete, it will directly affect the accuracy of the inversion results.

[0006] The Chinese invention patent application with application number CN110133655A discloses a river runoff monitoring inversion method based on multi-source radar remote sensing technology, which includes the following steps: Step 1, multi-source radar remote sensing data preprocessing: SAR image preprocessing, radar altimeter data preprocessing; Step 2, extracting river width based on SAR image; Step 3, extracting river relative water depth based on radar altimeter; Step 4, constructing a runoff calculation model based on Manning's formula; Step 5, monitoring river runoff changes in time series. This method relies on the measured minimum water level data. If the historical minimum water level data is missing, the water level data obtained only by the radar altimeter will directly affect the flow inversion accuracy. Summary of the invention

[0007] In order to solve the above technical problems, the present invention proposes a river flow estimation method using altimetry satellites to deduce the flow area of ​​the river channel. The river width and water level are extracted based on multi-source remote sensing satellite data to invert the river topography, and an empirical cross-section-flow formula is established to realize river flow prediction.

[0008] The purpose of the present invention is to provide a method for estimating river flow by using altimetry satellites to deduce river flow area, including optical image river width extraction and preprocessing, and also including the following steps: Step 1: Radar image preprocessing and river width extraction; Step 2: Altimetry satellite preprocessing water level extraction; Step 3: Invert river terrain using elevation boundaries; Step 4: Invert river topography from remote sensing images; Step 5: Construct river terrain; Step 6: River flow inversion.

[0009] Preferably, the optical image river width extraction and preprocessing comprises the following sub-steps: Step 01: Filter and preprocess the surface reflectance product data to obtain the river optical image dataset; Step 02: Calculate the normalized water index NDWI for the data set, extract the water body threshold by the maximum inter-class variance method, and extract the river water body range by masking. The formula is: NDWI= (Green-NIR) / (Green+NIR); Step 03: Use morphology to remove small holes and smooth water contours in the water body, and delete areas with larger slopes in the water body; Step 04: Apply the skeletonization algorithm to extract the center line of the river, calculate the cross-sectional angle of the river to obtain the median orthogonal angle, and extract the river width; Among them, Green is the green band and NIR is the near infrared band.

[0010] In any of the above schemes, preferably, the method of removing small holes and smooth water body contours in the water body by morphology comprises: performing image opening and image closing operations on the image, The image opening operation smoothes the contours of the object, disconnects narrow necks and eliminates thin protrusions; The image closing operation bridges narrow discontinuities and elongated gullies, eliminates small holes, and fills in breaks in contour lines; The image opening operation first erodes and then dilates; the image closing operation first dilates and then erodes.

[0011] In any of the above schemes, preferably, the method for deleting the area with a larger slope in the water body range includes: performing slope calculation on the DEM obtained in the GEE platform, and performing masking on the water body to extract the area with a smaller slope.

[0012] In any of the above schemes, preferably, step 1 includes screening Sentinel-1 GRD image data with an interference wide-band mode of IW, and performing Refined Lee filtering denoising, geographic correction, and radiometric normalization data processing to obtain a river radar image data set.

[0013] In any of the above schemes, preferably, step 2 includes downloading ICEsat-2 ATL03 data, denoising the image using an adaptive DBSCAN algorithm, performing Gaussian fitting on the water level to obtain the expectation, calculating the elevation conversion coefficient for elevation conversion, and obtaining the river water level in a unified coordinate system.

[0014] In any of the above schemes, preferably, the method for calculating the elevation conversion coefficient for elevation conversion includes converting the coordinates between different geodetic coordinate systems into corresponding coordinates through seven parameters, and the formula is: X' = ​​X + dx-Y*rz + Z*ry + S Y' = Y + X*rZ +Z*rx + S Z'=ZX*ry + Y*rx + S Among them, (X, Y, Z) are the coordinates of the original coordinate system, (X', Y', Z') are the coordinates of the target coordinate system, dx, dy and dz are translation parameters, rx, ry and rz are rotation parameters, and s is the scale parameter.

[0015] In any of the above schemes, it is preferred that step 3 includes obtaining multi-period river boundary elevations by combining water levels extracted from multi-period altimetry satellite data with river boundary information, interpolating the multi-period boundary elevations using the Kriging interpolation method to obtain river boundary elevation points above the lowest water level, and then applying the Kriging interpolation method to extrapolate the entire river elevation boundary to invert the underwater elevation, and obtaining the complete river topography by combining the inverted underwater topography.

[0016] In any of the above schemes, preferably, step 4 includes obtaining the river topography by interpolating the elevation boundary and calculating the water depth value of the corresponding point by using the multi-period water level data acquired by the altimetry satellite, using the blue and green band radiation brightness and water depth of the optical remote sensing image Sentinel-2 as prior data of the logarithmic ratio model, calculating relevant parameters, establishing a logarithmic ratio water depth inversion model, and inverting the multi-period underwater topography.

[0017] In any of the above schemes, it is preferred that step 5 includes retaining the overlapping areas of the elevation boundary and the river topography inverted by remote sensing images, and the non-overlapping areas are mainly based on the river topography results inverted by the elevation boundary, and the multi-period river topography inverted by remote sensing images is used as a reference to fuse the river topography to obtain the river topography.

[0018] In any of the above schemes, preferably, step 5 further includes taking the blue and green band radiation brightness and water depth of the optical remote sensing image Sentinel-2 as prior data of the logarithmic ratio model, calculating relevant parameters, and establishing the construction of the logarithmic ratio water depth inversion model:

[0019] In the formula, n , m 1. m 0 is the empirical parameter, and is the radiance value of the two bands, and is the water depth value.

[0020] In any of the above schemes, preferably, step 6 includes calculating the cross-sectional area of ​​the multi-period river channel based on the inverted river channel topography and the multi-period water level obtained by the altimetry satellite, establishing a cross-sectional area-flow empirical formula based on the cross-sectional area and the measured flow, and performing accuracy evaluation, and the cross-sectional area-flow empirical formula is: Q ≈ k A+ b in, Q is the river flow, A is the cross-sectional area of ​​the water flow, k and b is the empirical relationship coefficient.

[0021] In any of the above schemes, preferably, the accuracy evaluation method includes the root mean square error RMSE and correlation coefficient R For accuracy assessment, the calculation formula of RMSE is:

[0022] Correlation coefficient R The calculation formula is

[0023] in, Q p To measure river flow, q i To predict river flow, N is the total number of samples.

[0024] The present invention proposes a river flow estimation method using altimetry satellites to deduce the flow area of ​​a river channel, thereby achieving accurate inversion of river channel topography. The river flow inversion based on multi-source remote sensing satellite data has high accuracy, thereby achieving accurate prediction of river flow.

[0025] GEE platform: Google Earth Engine is a cloud platform provided by Google for online visualization, calculation and analysis of large amounts of global-scale earth science data (especially satellite data).

[0026] DEM: Digital Elevation Model, is a digital simulation of ground terrain through limited terrain elevation data.

[0027] DBSCAN: A density-based spatial clustering algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 The present invention is a flow chart of a preferred embodiment of a method for estimating river flow by using altimetry satellites to deduce the flow area of ​​a river channel.

[0029] Figure 2 The present invention is a flow chart of another preferred embodiment of a method for estimating river flow by using altimetry satellites to deduce the flow area of ​​a river channel. DETAILED DESCRIPTION

[0030] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments.

[0031] Example 1 like Figure 1 As shown, executing step 100, optical image river width extraction and preprocessing, includes the following sub-steps: Execute step 101 to filter and preprocess the surface reflectance product data to obtain a river optical image dataset.

[0032] Execute step 102 to calculate the normalized water index NDWI for the data set, extract the water body threshold by the maximum inter-class variance method, and extract the river water body range by masking. The formula is: NDWI= (Green-NIR) / (Green+NIR).

[0033] Execute step 103, remove small holes and smooth water body contours in the water body by morphology, and delete the area with a large slope in the water body range; the method of removing small holes and smooth water body contours in the water body by morphology includes: performing image opening and image closing operations on the image, The image opening operation smoothes the contours of the object, disconnects narrow necks and eliminates thin protrusions; The image closing operation bridges narrow discontinuities and elongated gullies, eliminates small holes, and fills in breaks in contour lines; The image opening operation first erodes and then dilates; the image closing operation first dilates and then erodes.

[0034] The method for deleting the area with a larger slope in the water body range includes: performing slope calculation on the DEM obtained in the GEE platform, and performing masking on the water body to extract the area with a smaller slope.

[0035] Execute step 104, apply a skeletonization algorithm to extract the center line of the river, calculate the angle of the river cross section to obtain the median orthogonal angle, and extract the river width; Among them, Green is the green band and NIR is the near infrared band.

[0036] Execute step 110 to extract the river width by radar image preprocessing, including screening Sentinel-1 GRD image data with an interference wide stripe mode of IW, and performing Refined Lee filtering denoising, geographic correction, and radiometric normalization data processing to obtain a river radar image dataset.

[0037] Execute step 120 to extract water level by preprocessing the altimetry satellite, including downloading ICEsat-2 ATL03 data, denoising the image using the adaptive DBSCAN algorithm, performing Gaussian fitting on the water level to obtain the expectation, calculating the elevation conversion coefficient for elevation conversion, and obtaining the river water level in a unified coordinate system.

[0038] The method for calculating the elevation conversion coefficient for elevation conversion includes converting the coordinates between different geodetic coordinate systems into corresponding coordinates through seven parameters. The formula is: X' = ​​X + dx-Y*rz + Z*ry + S Y' = Y + X*rZ +Z*rx + S Z'=ZX*ry + Y*rx + S Among them, (X, Y, Z) are the coordinates of the original coordinate system, (X', Y', Z') are the coordinates of the target coordinate system, dx, dy and dz are translation parameters, rx, ry and rz are rotation parameters, and s is the scale parameter.

[0039] Execute step 130 to invert the river topography using the elevation boundary, including obtaining the multi-period river boundary elevation using the water level extracted from the multi-period altimetry satellite data in combination with the river boundary information, interpolating the multi-period boundary elevation using the Kriging interpolation method to obtain the river boundary elevation points above the lowest water level, and then applying the Kriging interpolation method to extrapolate the entire river elevation boundary to invert the underwater elevation, and combining the inverted underwater topography to obtain the complete river topography.

[0040] Execute step 140 to invert the river topography from remote sensing images, including obtaining the river topography by interpolating the elevation boundary and calculating the water depth value of the corresponding point using the multi-period water level data obtained by the altimetry satellite, using the blue and green band radiation brightness and water depth of the optical remote sensing image Sentinel-2 as the prior data of the logarithmic ratio model, calculating the relevant parameters, establishing the logarithmic ratio water depth inversion model, and inverting the multi-period underwater topography.

[0041] Execute step 150 to construct river terrain, including retaining the overlapping areas of the river terrain inverted from the elevation boundary and remote sensing images, and using the river terrain results inverted from the elevation boundary as the main non-overlapping areas. The multi-period river terrain inverted from the remote sensing images is used as a reference to fuse the river terrain to obtain the river terrain. The blue and green band radiation brightness and water depth of the optical remote sensing image Sentinel-2 are used as the prior data of the logarithmic ratio model, and the relevant parameters are calculated to establish the construction of the logarithmic ratio water depth inversion model:

[0042] In the formula, n , m 1. m 0 is the empirical parameter, and is the radiance value of the two bands, and is the water depth value.

[0043] Execute step 160, river flow inversion, including calculating the cross-sectional area of ​​the multi-period river channel based on the inverted river channel topography and the multi-period water level obtained by the altimetry satellite, establishing the cross-sectional area-flow empirical formula based on the cross-sectional area and the measured flow, and performing accuracy evaluation. The cross-sectional area-flow empirical formula is: Q ≈ k A+ b in, Q is the river flow, A is the cross-sectional area of ​​the water flow, k and b is the empirical relationship coefficient.

[0044] The accuracy evaluation method includes the root mean square error RMSE and the correlation coefficient R For accuracy assessment, the calculation formula of RMSE is:

[0045] Correlation coefficient R The calculation formula is

[0046] in, Q p To measure river flow, q i To predict river flow, Nis the total number of samples.

[0047] Example 2 like Figure 2 As shown, a river flow inversion method based on the coordinated application of multi-mode remote sensing satellites of the present invention comprises: Step S1, optical image river width extraction and preprocessing: On the Google Earth Engine platform, screen the Sentinel-2 Level-2A surface reflectance products with a value less than 10%, and preprocess the data to obtain a river optical image dataset. Calculate the NDWI (Normalized Difference Water Index) for the dataset, extract the water body threshold using the OTSU threshold method (maximum inter-class variance method), and extract the river water body range using a mask. Use morphological methods to remove small holes in the water body and smooth water body contours, delete areas with larger slopes in the water body range, and other data processing to improve the accuracy of water body extraction. Apply the skeleton algorithm to extract the centerline of the river, calculate the river cross-section angle to obtain the median orthogonal angle, and extract the river width. Wherein, NDWI (normalized water index): NDWI= (Green-NIR) / (Green+NIR); Step S2, radar image preprocessing and river width extraction: Sentinel-1 GRD images with interference wide strip mode of IW are selected on the Google Earth Engine platform, and the data are subjected to Refined Lee filtering denoising, geo-correction, and radiometric normalization data processing to obtain a river radar image dataset. Calculate the SDWI (dual polarization water index) for the dataset, extract the water body threshold by the OTSU threshold method (maximum inter-class variance method), and extract the river water body range by masking. Through morphological removal of small holes and smooth water body contours in the water body, and deletion of areas with large slopes in the water body range, the accuracy of water body extraction is improved; the skeleton algorithm is used to extract the center line of the river, and the cross-sectional angle of the river is calculated to obtain the median orthogonal angle to extract the river width. Wherein SDWI (dual polarization water index): SDWI=EXP(VV*VH / 1000) Step S3, altimetry satellite preprocessing and water level extraction: ICEsat-2 ATL03 data is downloaded from the National Aeronautics and Space Administration (NASA), the image is denoised using the adaptive DBSCAN algorithm, the water level is Gaussian fitted to obtain the expected value, the elevation conversion coefficient is calculated for elevation conversion, and the river water level in a unified coordinate system is obtained.

[0048] Step S4, inverting river channel topography from elevation boundary: The water level extracted from multi-period altimetry satellite data is combined with river boundary information to obtain multi-period river channel boundary elevations, and the multi-period boundary elevations are used to perform Kriging interpolation to obtain river channel boundary elevation points above the lowest water level. The Kriging interpolation method is then applied to extrapolate the entire river elevation boundary to invert the underwater elevation, and the complete river channel topography is obtained by combining the inverted underwater topography.

[0049] Step S5: Invert river topography from remote sensing images: The river topography is obtained by interpolating the elevation boundary and the multi-period water level data obtained by the altimetry satellite to calculate the water depth value of the corresponding point. The blue and green band radiation brightness and water depth of the optical remote sensing image Sentinel-2 are used as the prior data of the logarithmic ratio model, and the relevant parameters are calculated to establish a logarithmic ratio water depth inversion model to invert the multi-period underwater topography.

[0050] Step S6, constructing river topography: retaining the overlapping areas of the elevation boundary and the river topography inverted by remote sensing images, and taking the river topography results inverted by the elevation boundary as the main non-overlapping areas, and taking the multi-period river topography inverted by remote sensing images as a reference, and performing river topography fusion to obtain the river topography.

[0051] Step S7, river flow inversion: Based on the inverted river channel topography and the multi-period water levels obtained by the altimetry satellite, the multi-period river channel cross-sectional area can be calculated, and the cross-sectional area and measured flow are used to establish the cross-sectional area-flow empirical formula, and the accuracy is evaluated.

[0052] In step S5, the blue and green band radiance and water depth of the optical remote sensing image Sentinel-2 are used as the prior data of the logarithmic ratio model, and the relevant parameters are calculated to establish the construction of the logarithmic ratio water depth inversion model:

[0053] In the formula n ,, are empirical parameters, , are the radiance values ​​of the two bands, and is the water depth value.

[0054] The river flow inversion accuracy evaluation in step S7 includes the following: (1) Root mean square error (RMSE), calculated as:

[0055] (2) Correlation coefficient (R), calculated as:

[0056] Q p : Measured river flow; q i : Predict river flow; N : Total number of samples.

[0057] Through the above technical scheme disclosed in the present invention, the following beneficial results are obtained: the river width and water level of multiple periods are extracted by remote sensing satellite data to accurately invert the river topography, and the flow is inverted by constructing a truncation-flow empirical formula, thereby achieving accurate prediction of river flow and providing effective technical support for river flow inversion.

[0058] Example 3 A river flow inversion system based on the coordinated application of multi-mode remote sensing satellites, comprising: The remote sensing data processing module is configured to obtain Sentinel-1, 2 satellite and ICEsat-2 satellite data products, and perform data preprocessing, using Sentinel-1, 2 satellite images to extract river width and ICEsat-2 satellite images to extract water level.

[0059] The river topography inversion module is configured to use river width and water level to extract multi-temporal river boundaries and interpolate to obtain river topography; use optical images and water depth to establish a water depth inversion model to obtain multi-temporal river topography, and fuse the two data to obtain the river topography.

[0060] The flow prediction module is configured to calculate the cross-sectional area of ​​the river channel for multiple periods through the inverted river channel topography and the multiple water levels obtained by the altimetry satellite, and to establish the cross-sectional area-flow empirical formula based on the cross-sectional area and the measured flow.

[0061] The accuracy detection module is configured to detect the accuracy of river flow prediction by applying a root mean square error formula.

[0062] In order to better understand the present invention, the above is described in detail in conjunction with the specific embodiments of the present invention, but it is not intended to limit the present invention. Any simple modifications made to the above embodiments based on the technical essence of the present invention still fall within the scope of the technical solution of the present invention. Each embodiment in this specification focuses on the differences from other embodiments, and the same or similar parts between the embodiments can be referenced to each other. For the system embodiment, since it basically corresponds to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

Claims

1. A method for estimating river flow by using altimetry satellite to deduce river flow area, including optical image river width extraction and preprocessing, characterized in that: The following steps are also included: Step 1: Radar image preprocessing and river width extraction; Step 2: Altimetry satellite preprocessing water level extraction; Step 3: Invert river terrain using elevation boundaries; Step 4: Invert river topography from remote sensing images; Step 5: Construct river terrain; Step 6: Inversion of river flow: Based on the inverted river topography and the multi-period water levels obtained by the altimetry satellite, the multi-period river cross-sectional area is calculated, and the cross-sectional area and the measured flow are used to establish the cross-sectional-flow empirical formula, and the accuracy is evaluated. The accuracy evaluation method includes performing accuracy evaluation through the root mean square error RMSE, and the calculation formula is: ， Correlation coefficient R The calculation formula is ， in, Q p To measure river flow, q i To predict river flow, N is the total number of samples.

2. The method for estimating river flow by using altimetry satellite to deduce river flow area as claimed in claim 1, characterized in that: The optical image river width extraction and preprocessing includes the following sub-steps: step 01: Filter and preprocess the surface reflectance product data to obtain the river optical image dataset; Step 02: Calculate the normalized water index NDWI for the data set, extract the water body threshold by the maximum inter-class variance method, and extract the river water body range by masking. The formula is: NDWI= (Green-NIR) / (Green+NIR); Step 03: Use morphology to remove small holes and smooth water contours in the water body, and delete areas with larger slopes in the water body; Step 04: Apply the skeletonization algorithm to extract the center line of the river, calculate the cross-sectional angle of the river to obtain the median orthogonal angle, and extract the river width; Among them, Green is the green band and NIR is the near infrared band.

3. The method for estimating river flow by using altimetry satellite to deduce river flow area as claimed in claim 2, characterized in that: The method for removing small holes and smooth water body contours in water bodies by morphology comprises: performing image opening and image closing operations on the image, The image opening operation smoothes the contours of the object, disconnects narrow necks and eliminates thin protrusions; The image closing operation bridges narrow discontinuities and elongated gullies, eliminates small holes, and fills in breaks in contour lines; The image opening operation first erodes and then dilates; the image closing operation first dilates and then erodes.

4. The method for estimating river flow by using altimetry satellite to deduce river flow area as claimed in claim 3, characterized in that: The method for deleting the area with a larger slope in the water body range includes: performing slope calculation on the DEM obtained in the GEE platform, and performing masking on the water body to extract the area with a smaller slope.

5. The method for estimating river flow by using altimetry satellite to deduce river flow area as claimed in claim 4, characterized in that: The step 1 includes screening Sentinel-1 GRD image data with an interference wide-band mode of IW, and performing Refined Lee filtering denoising, geographic correction, and radiometric normalization data processing to obtain a river radar image data set.

6. The method for estimating river flow by using altimetry satellite to deduce river flow area as claimed in claim 5, characterized in that: The step 2 includes downloading ICEsat-2 ATL03 data, denoising the image using an adaptive DBSCAN algorithm, performing Gaussian fitting on the water level to obtain the expectation, calculating the elevation conversion coefficient for elevation conversion, and obtaining the river water level in a unified coordinate system.

7. The method for estimating river flow by using altimetry satellite to deduce river flow area as claimed in claim 6, characterized in that: The method for calculating the elevation conversion coefficient for elevation conversion includes converting the coordinates between different geodetic coordinate systems into corresponding coordinates through seven parameters. The formula is: X' = ​​X + dx-Y*rz + Z*ry + S Y' = Y + X*rZ +Z*rx + S Z'=ZX*ry + Y*rx + S Among them, (X, Y, Z) are the coordinates of the original coordinate system, (X', Y', Z') are the coordinates of the target coordinate system, dx, dy and dz are translation parameters, rx, ry and rz are rotation parameters, and s is the scale parameter.

8. The method for estimating river flow by using altimetry satellite to deduce river flow area as claimed in claim 7, characterized in that: The step 3 includes obtaining multi-period river boundary elevations by combining the water level extracted from the multi-period altimetry satellite data with the river boundary information, interpolating the multi-period boundary elevations using the Kriging interpolation method to obtain the river boundary elevation points above the lowest water level, and then applying the Kriging interpolation method to extrapolate the entire river elevation boundary to invert the underwater elevation, and combining the inverted underwater topography to obtain the complete river channel topography.

9. The method for estimating river flow by using altimetry satellite to deduce river flow area as claimed in claim 8, characterized in that: The step 4 includes obtaining the river topography through interpolation of the elevation boundary and calculating the water depth value of the corresponding point through the multi-period water level data obtained by the altimetry satellite, using the blue and green band radiation brightness and water depth of the optical remote sensing image Sentinel-2 as the prior data of the logarithmic ratio model, calculating the relevant parameters, establishing the logarithmic ratio water depth inversion model, and inverting the multi-period underwater topography.

10. The method for estimating river flow by using altimetry satellite to deduce river flow area as claimed in claim 9, characterized in that: The step 5 includes retaining the overlapping areas of the elevation boundary and the river topography inverted by the remote sensing image, and the non-overlapping areas are mainly based on the river topography results inverted by the elevation boundary, and the multi-period river topography inverted by the remote sensing image is used as a reference to fuse the river topography to obtain the river topography.

Images