Method and system for characterizing complex form of braided river

Abstract

The invention discloses a braided river complex form characterization method and system, and the method comprises the steps: 1, carrying out the preliminary screening and cutting of an original remote sensing image in a set time period and a set research region, setting a cloud cover ratio threshold value of the research region, carrying out the screening, obtaining an initial remote sensing image, and carrying out the preprocessing; step 2, calculating an initial remote sensing image MNDWI index to obtain an MNDWI index image set; 3, calculating a segmentation threshold value of each pixel in the images in the MNDWI index image set, and carrying out preliminary extraction on the water body of the braided river image to obtain a preliminary extraction water body set, and 4, obtaining a final water body set based on the preliminary extraction water body set; and step 5, based on the extracted final water body set and the delimited river boundary, performing morphological sampling, calculating morphological parameters, and returning a morphological parameter set of the river reach in the form of a data table. According to the method, the time required for extracting morphological features of the braided river is remarkably shortened, and braided river research based on remote sensing images is facilitated.

Description

Technical Field

[0001] The invention belongs to the technical field of remote sensing image processing, and in particular relates to a method and system for characterizing the complex morphology of a braided river. Background Art

[0002] Rivers in nature have various shapes and large variations in different scales. Most rivers originate from plateaus and mountains, which are long and dangerous and difficult to access. At present, there is a lack of high-precision measured topography and water and sediment data for plateau rivers (source of the Yangtze River, Yellow River, Lancang River, etc.). It is difficult to obtain measured data using traditional manual measurement methods. How to efficiently obtain large-scale river prototype data and study the evolution of riverbeds has always been a problem that scholars have paid close attention to.

[0003] The field river prototype observation method mainly uses surveying instruments such as total stations, theodolites, RTK, and drones to survey the study area. Plateau rivers have changeable morphology, complex terrain, and harsh natural conditions. There are many difficulties in field investigations (altitude and hypoxia, inconvenient transportation, bad weather, transportation costs, etc.). The measurement accuracy and measurement area are greatly constrained by human factors and natural conditions, which consumes a lot of manpower and material resources and cannot carry out long-term field work. Remote sensing images have high temporal resolution and comprehensive spatial coverage, and have become an important method for studying river evolution, especially in remote areas (such as plateau rivers). At present, the use of remote sensing imagery to study the evolution of braided rivers has the following technical problems: 1. The utilization rate of data is not high, and remote sensing image information is not fully mined: The traditional way to obtain remote sensing images is to download them from websites, such as USGS and geospatial data clouds. The remote sensing images obtained by this method are complete images, but researchers of river evolution often only need a small part of the image, and the screening and downloading process is very time-consuming. 2. The traditional method for extracting water bodies in braided rivers is the global Otsu algorithm, which ignores small branches and leads to an underestimated complexity of the water body. 3. For the extracted water bodies, the morphological parameters of braided rivers are mainly calculated manually, which is time-consuming and labor-intensive, and there are human errors, which reduces the reliability and scientificity of the data processing results. Therefore, the development of fast and batch remote sensing image screening, downloading, high-precision water body extraction, and morphological parameter calculation methods has important scientific significance and practical value for making full use of remote sensing images to explore braided river information and study the riverbed evolution of plateau rivers. Summary of the invention

[0004] The present invention is carried out to solve the above problems, and its purpose is to provide a method and analysis system for characterizing the complex morphology of braided rivers. The method is simple and easy to learn, with low time and labor costs, greatly improving the efficiency and accuracy of braided river morphological feature extraction, and providing a scientific and reliable technical process for further research on the evolution law of braided rivers.

[0005] In order to achieve the above purpose, the present invention adopts the following scheme:

[0006] A method for characterizing the complex morphology of a braided river comprises the following steps:

[0007] Step 1: Preliminarily screen and crop the original remote sensing images within the set time period and the set study area, set the cloud cover ratio threshold of the study area, further screen and obtain the initial remote sensing images and preprocess them;

[0008] Step 2, calculate the initial remote sensing image MNDWI index to obtain the MNDWI index image set;

[0009] Step 3: Calculate the segmentation threshold of each pixel in the image in the MNDWI index image set, and perform preliminary extraction of water bodies in the braided river image based on the segmentation threshold to obtain a preliminary extracted water body set.

[0010] Step 4: Obtain the final water body set based on the preliminary extracted water body set;

[0011] Step 5: Based on the extracted final water body set and the delineated river boundary, perform morphological sampling, calculate morphological parameters, and return the morphological parameter set of the river section in the form of a data table.

[0012] Furthermore, in the step 1, the ROI is used as the cropping area, and the preprocessing includes: removing the image of the river covered by clouds, ice and snow.

[0013] Furthermore, the step 3 comprises:

[0014] Step 3.1, convert the MNDWI image into image array npy, geographic transformation parameter p, and projection information prj, and store them separately;

[0015] Step 3.2, crop the image based on the delineated river boundary;

[0016] Step 3.3, using the maximum inter-class variance as the classification goodness indicator, calculate the segmentation threshold for each pixel of the image Xiaohong; and determine whether the pixel is a water body based on the segmentation threshold, and perform preliminary extraction of the water body of the braided river image to obtain a preliminary water body extraction image;

[0017] Furthermore, the step 3.3 also includes removing non-water body noise on land and filling water body cavity noise in wider branches, and updating the water body of the extracted braided river image to obtain a preliminary extracted water body set.

[0018] Furthermore, the step 4 comprises:

[0019] Step 4.1, calculate the local minimum point-saddle point pair in the water body image in the final water body set;

[0020] Step 4.2, for each pair of local minimum point-saddle point, the lowest path between local minimum point-saddle point:

[0021] Step 4.3, repeat step 4.2 until all local minimum points are connected to the saddle point, and all minimum paths form a branch path network;

[0022] Step 4.4, traverse the lowest path in the branch path network, delete the objects in the lowest path whose MNDWI average value is less than the MNDWI set threshold and whose branch path network length is less than the length set threshold, and update the branch path network to complete the water body extraction of the braided river and obtain the final water body set.

[0023] Furthermore, the step 4.1 comprises:

[0024] Step 4.11, calculate the local minimum point in the water body image in the final water body set;

[0025] Step 4.12, calculate all saddle points of the water body image in the final water body set;

[0026] Step 4.13: For each local minimum point, traverse all saddle points and determine the saddle point closest to the local minimum point, that is, find a pair of local minimum point-saddle point.

[0027] Furthermore, step 4.2 includes: taking the saddle point as the starting point, considering the lowest point among the 8 surrounding pixel points, calculating the distance between the point and the local lowest point, and then taking the point as the starting point, calculating the lowest point among the 8 surrounding pixel points, and repeating this step until the calculated path reaches the local lowest point.

[0028] Furthermore, the step 5 comprises:

[0029] Step 5.1, based on the delineated river boundary, equidistantly encrypt the coordinates of the left and right river banks to generate an encrypted boundary line, and generate the river centerline, while calculating the river width at each point on the centerline, and counting the length of the river section;

[0030] Step 5.2, generate sections according to a certain step length, and count the number of branches, water surface width, and water surface width on each section by traversing the sections;

[0031] Step 5.3, calculate the braid strength, water surface width and water surface width of each section of the river;

[0032] Step 5.4: Based on the designated river channel boundary, count the water area and river channel area of ​​the river section, and calculate the water area ratio; count the average sandbar area ratio of the river section; calculate the average river channel width based on the river length obtained in step 5.1; and calculate the branch channel density based on the braided strength of the river section scale.

[0033] Furthermore, the step 5 also includes: collecting the morphological parameters calculated in the process of water body sampling and morphological parameter calculation into the same data table.

[0034] On the other hand, the present invention also provides a braided river complex morphology characterization system, comprising:

[0035] The initial remote sensing image acquisition module is used to preliminarily screen and crop the original remote sensing images within the set time period and the set study area, set the cloud cover ratio threshold of the study area, and further screen and obtain the initial remote sensing images and preprocess them;

[0036] The MNDWI index image set acquisition module is used to calculate the MNDWI index of the initial remote sensing image to obtain the MNDWI index image set;

[0037] A preliminary water body set acquisition module is used to calculate the segmentation threshold of each pixel in the image in the MNDWI index image set, and to obtain a preliminary water body set by preliminary extraction of the water body of the braided river image based on the segmentation threshold.

[0038] A final water body set acquisition module, which is used to acquire a final water body set based on the preliminary extracted water body set;

[0039] The morphological parameter calculation module is used to perform morphological sampling and calculate morphological parameters based on the extracted final water body set and the delineated river channel boundary, and return the morphological parameter set of the river section in the form of a data table.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] The present invention can accurately extract braided river water bodies, and combined with the GEE platform, can quickly generate a set of braided river morphological parameters. For a river section, the only things that need to be set in the entire process are the study area, the river boundary, and three thresholds. The present invention can significantly shorten the time required for extracting braided river morphological features, which is beneficial to the study of braided rivers based on remote sensing images. By making full use of remote sensing image data, more morphological evolution laws can be discovered, and the results are reliable and accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is the technical roadmap of the method of the present invention.

[0043] Figure 2 This is a diagram of calculation results of braided river water extraction according to the present invention, wherein (a)-(d) are water extraction processes respectively.

[0044] Figure 3Schematic diagrams comparing the braided river water extraction method of the present invention with the traditional method (Global Otsu method), wherein (a) is a schematic diagram of water extraction by the present invention; (b) is a schematic diagram of water extraction by the traditional method; (c) is a schematic diagram of water extraction by the present invention. T3 Compared with traditional method B T3 Comparative schematic diagram; (d) is the R W Compared with the traditional method R W Comparison diagram. DETAILED DESCRIPTION

[0045] In order to facilitate those skilled in the art to understand and implement the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and the present embodiment. It should be understood that the implementation examples described herein are only used to illustrate and explain the present invention and are not used to limit the present invention.

[0046] Example 1

[0047] like Figure 1 As shown, the present invention provides a method for characterizing the complex morphology of a braided river, comprising the following steps:

[0048] Step 1: Preliminarily screen and crop the original remote sensing images within the set time period and the set study area, set the cloud cover ratio threshold of the study area, further screen and obtain the initial remote sensing images and preprocess them;

[0049] Step 2, calculate the initial remote sensing image MNDWI index to obtain the MNDWI index image set;

[0050] Step 3: Calculate the segmentation threshold of each pixel in the image in the MNDWI index image set, and perform preliminary extraction of water bodies in the braided river image based on the segmentation threshold to obtain a preliminary extracted water body set.

[0051] Step 4: Obtain the final water body set based on the preliminary extracted water body set;

[0052] Step 5: Based on the extracted final water body set and the delineated river boundary, perform morphological sampling, calculate morphological parameters, and return the morphological parameter set of the river section in the form of a data table.

[0053] In this embodiment, in step 1, in the GEE-Python platform, the ROI of the study area of ​​the Tuotuo River section in the source area of ​​the Yangtze River is delineated, and Landsat 8 is selected for calculation. The images are initially screened according to the ROI, year 2013-2020, month June-September and other conditions. The images screened in step 1 are cropped, with ROI as the cropping area. Use the ee.Algorithms.Landsat.simpleCloudScore(image) function built into GEE to calculate the cloud cover for the cropped image in step 2. According to the proportion of cloud cover in the study area, the images are further screened when the cloud cover is <20%. For the image set filtered according to cloud cover, images of rivers covered by clouds, ice and snow are removed through manual inspection. The image screening is now complete.

[0054] In step 2, the normalizedDifference(['B3','B6']) method of the built-in ee.Image type data in GEE is used to calculate the MNDWI index image set. The MNDWI index image is calculated because it can clearly distinguish between land and water.

[0055] Call the geemap.ee_export_image_collection() function in the Python third-party library geemap to download the calculated MNDWI indicator image set to the local computer for subsequent processing.

[0056] Step 3 includes:

[0057] Step 3.1, convert the MNDWI image into image array npy, geographic transformation parameter p, and projection information prj, and store them separately;

[0058] Step 3.2, crop the image based on the delineated river boundary;

[0059] Step 3.3, using the maximum inter-class variance as the classification goodness indicator, calculate the segmentation threshold for each pixel of the image Xiaohong; and based on the segmentation threshold, determine whether the pixel is a water body, and perform preliminary extraction of the water body of the braided river image to obtain a preliminary water body extraction image.

[0060] Specifically: the downloaded MNDWI image is converted into npy (image array), p (geographic conversion parameters), prj (projection information), based on the custom read_geo() and save() functions. The converted files are stored in the GEE_IMG\LC08\b1 folder under the project folder, in the three subfolders of geo, img, and prj.

[0061] Prepare the delineated riverbank boundary file (shapefile) and crop the image according to the river boundary. The purpose of cropping the image is to reduce the amount of calculation and facilitate the use of subsequent algorithms.

[0062] Use the maximum inter-class variance as the classification goodness indicator to calculate the segmentation threshold for each pixel in the image. Now there is a central pixel, and only consider all pixels within a radius of 5 pixels from the central pixel. These points will participate in the calculation of the central pixel segmentation threshold. Sort the pixels involved in the calculation from small to large, and calculate the inter-class variance for each pixel value: in p i is the pixel value of the i-th pixel. For all pixel values, each inter-class variance is calculated accordingly, and the pixel value corresponding to the maximum inter-class variance is the segmentation threshold of the central pixel.

[0063] Traverse each pixel of the image, calculate the segmentation threshold for each pixel, and judge whether the pixel is a water body according to the segmentation threshold: if the pixel value is greater than the segmentation threshold, it is a water body, otherwise it is a non-water body. The preliminary extraction of the water body of the braided river image is completed, that is, the mask.

[0064] For the first initially extracted water body mask, set denoise_thresh = 0. Traverse each independent water body object, delete the object whose MNDWI average value is less than denoise_thresh in the mask, and update the mask. This step is to remove land non-water body noise.

[0065] According to the mask obtained in the previous step, find the hole (i.e. sandbar) object in the mask, i.e. bar.

[0066] Based on the obtained bar, set bar_thresh = 0.2, delete the objects whose MNDWI average value in bar is less than bar_thresh, and update bar. This step is to fill the water hole noise in the wider branch channel (caused by the local MNDWI value in the branch channel being lower than the surrounding water body).

[0067] Add the updated bar and the mask updated in step 10 to update the mask again, and get Water mask = mask + bar.

[0068] Step 4 includes:

[0069] Step 4.1, calculate the local minimum point-saddle point pair in the water body image in the final water body set;

[0070] Step 4.2, for each pair of local minimum point-saddle point, the lowest path between local minimum point-saddle point:

[0071] Step 4.3, repeat step 4.2 until all local minimum points are connected to the saddle point, and all minimum paths form a branch path network;

[0072] Step 4.4, traverse the lowest path in the branch path network, delete the objects whose MNDWI average value in the lowest path is less than the MNDWI set threshold and the branch path network length is less than the length set threshold, and update the branch path network to complete the water body extraction of the braided river to obtain the final water body set, such as Figure 2 shown.

[0073] Specifically: Calculate the local minimum point: Use the python module skimage.morphology.local_minima to calculate all local minimum points.

[0074] Calculate all saddle points of the image: max_h is the maximum value of each row, min_l is the minimum value of each column; store the maximum value of each row in list h, and store the minimum value of each column in list l; determine whether there are the same numbers in lists h and l, if so, it is a saddle point.

[0075] For each local minimum point, traverse all saddle points and determine the saddle point closest to the local minimum point, that is, find a pair of local minimum point-saddle point.

[0076] For each pair of local minimum point-saddle point, calculate the lowest path link between the two points: starting from the saddle point, consider the lowest point among the 8 pixel points around it, and calculate the distance between this point and the local minimum point. Starting from this point again, calculate the lowest point among the 8 pixel points around it, and repeat this step until the calculated path reaches the local minimum point. In this process, if the distance between the point and the local minimum point is greater than twice the distance between the local minimum point and the saddle point, it means that the path is wrong. Delete the path around the saddle point and repeat the step again until the lowest path between the local minimum point and the saddle point is calculated.

[0077] Repeat the previous step until all local minimum points are connected to the saddle point, and all minimum paths form a branch path network net.

[0078] Traverse each link in the net, delete the links whose MNDWI average value is less than link_thresh=-0.2, and update the net. This step is to delete the noise links that are misjudged as branches in the calculation results.

[0079] Based on the net obtained in the previous step, delete the objects whose length is less than 10 pixels and update the net.

[0080] Add the net obtained in the previous step to the water mask obtained, and use the mask supplemented by the net, that is, mask = mask + net. Get the Water Segment and complete the water extraction of the braided river. Figure 3 Shown is a schematic diagram comparing the braided river water extraction method of this embodiment with the traditional method (Global Otsu method).

[0081] Figure 3 B in c T3 (Real) is the braid strength value obtained by visual interpretation.

[0082] The step 5 comprises:

[0083] Step 5.1, based on the delineated river boundary, equidistantly encrypt the coordinates of the left and right river banks to generate an encrypted boundary line, and generate the river centerline, while calculating the river width at each point on the centerline, and counting the length of the river section;

[0084] Step 5.2, generate sections according to a certain step length, and count the number of branches, water surface width, and water surface width on each section by traversing the sections;

[0085] Step 5.3, calculate the braid strength, water surface width and water surface width of each section of the river;

[0086] Step 5.4: Based on the designated river channel boundary, count the water area and river channel area of ​​the river section, and calculate the water area ratio; count the average sandbar area ratio of the river section; calculate the average river channel width based on the river length obtained in step 5.1; and calculate the branch channel density based on the braided strength of the river section scale.

[0087] Furthermore, the step 5 also includes: collecting the morphological parameters calculated in the process of water body sampling and morphological parameter calculation into the same data table.

[0088] Specifically: Based on the generated boundary shp file, the coordinates of the left and right river banks are encrypted at equal intervals so that the number of coordinate points on the left and right banks is equal, which is 3 times the maximum number of coordinate points in the original left and right bank lines.

[0089] Based on the encrypted boundary line, the center line axis of the river (the center point sequence connecting the left and right bank sequential points) is generated, and the river width of each point on the center line (the straight-line distance between the left and right bank sequential points) is calculated, and the river length l (center line length) of the river section is counted.

[0090] According to the generated axis, generate sections at a certain step length, and set the section step length Crs_step to 200 (m).

[0091] Cross-section sampling statistics of water body information. For a water body shapefile, by traversing the cross sections, count the number of channels, water surface width, and water surface width on each cross section. Water surface width is the width of the water body that the cross section passes through (including the width of the sandbar in the water body), and water surface width is the width of the water body that the cross section passes through (excluding the width of the sandbar). This step is completed based on the intersection method of the geometry object in the shapely library.

[0092] Calculate the morphological parameter B T3 ,L,L W , respectively characterize the braiding strength, water surface widening and water surface width at the river section scale. The calculation method of each parameter is the average value of all sections.

[0093] Generate the shapefile of the river section and river channel according to the boundary file, and count the water area A of the river section W , river area A, and calculate the water area ratio R W =A W / A, average sandbar area ratio of statistical river sections According to the obtained river length l, calculate the average river width W = A / l. According to the obtained B T3 , calculate the branch channel density Rr = B T3 / W.

[0094] The above steps are used for sampling and morphological parameter calculation process of water bodies. The morphological parameters calculated from each image are collected into the same data table, and the table header is each morphological parameter and the image shooting date. If the user has the hydrological data of the river section, the daily hydrological information such as flow Q will be stored synchronously in the data table.

[0095] Example 2

[0096] This embodiment provides a braided river complex morphology characterization system, including:

[0097] The initial remote sensing image acquisition module is used to preliminarily screen and crop the original remote sensing images within the set time period and the set study area, set the cloud cover ratio threshold of the study area, and further screen and obtain the initial remote sensing images and preprocess them;

[0098] The MNDWI index image set acquisition module is used to calculate the MNDWI index of the initial remote sensing image to obtain the MNDWI index image set;

[0099] A preliminary water body set acquisition module is used to calculate the segmentation threshold of each pixel in the image in the MNDWI index image set, and to obtain a preliminary water body set by preliminary extraction of the water body of the braided river image based on the segmentation threshold.

[0100] A final water body set acquisition module, which is used to acquire a final water body set based on the preliminary extracted water body set;

[0101] The morphological parameter calculation module is used to perform morphological sampling and calculate morphological parameters based on the extracted final water body set and the delineated river channel boundary, and return the morphological parameter set of the river section in the form of a data table.

[0102] The above is only a preferred specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed in the present application should be covered within the protection scope of the present application.

[0103] It should be understood that parts not elaborated in detail in this specification belong to the prior art.

[0104] It should be understood that the above description of the preferred embodiment is relatively detailed and cannot be regarded as limiting the scope of patent protection of the present invention. Under the enlightenment of the present invention, ordinary technicians in this field can also make substitutions or modifications without departing from the scope of protection of the claims of the present invention, which all fall within the scope of protection of the present invention. The scope of protection requested for the present invention shall be based on the attached claims.

Claims

1. A method for characterizing the complex morphology of a braided river, characterized in that: The steps include: Step 1: Preliminarily screen and crop the original remote sensing images within the set time period and the set study area, set the cloud cover ratio threshold of the study area, further screen and obtain the initial remote sensing images and preprocess them; Step 2, calculate the initial remote sensing image MNDWI index to obtain the MNDWI index image set; Step 3: Calculate the segmentation threshold of each pixel in the image in the MNDWI index image set, and perform preliminary extraction of water bodies in the braided river image based on the segmentation threshold to obtain a preliminary extracted water body set. Step 4: Obtain the final water body set based on the preliminary extracted water body set; Step 5: Based on the extracted final water body set and the delineated river boundary, perform morphological sampling, calculate morphological parameters, and return the morphological parameter set of the river section in the form of a data table.

2. A method for characterizing complex morphology of braided rivers according to claim 1, characterized in that: In the step 1, the ROI is used as the cropping area, and the preprocessing includes: removing the image of the river covered by clouds, ice and snow.

3. A method for characterizing complex morphology of braided rivers according to claim 1, characterized in that: The step 3 comprises: Step 3.1, convert the MNDWI image into image array npy, geographic transformation parameter p, and projection information prj, and store them separately; Step 3.2, crop the image based on the delineated river boundary; Step 3.3, using the maximum inter-class variance as the classification goodness indicator, calculate the segmentation threshold for each pixel of the image Xiaohong; and based on the segmentation threshold, determine whether the pixel is a water body, and perform preliminary extraction of the water body of the braided river image to obtain a preliminary water body extraction image.

4. A method for characterizing complex morphology of braided rivers according to claim 3, characterized in that: The step 3.3 also includes removing non-water body noise on land and filling water body cavity noise in wider branches, and updating the water body of the extracted braided river image to obtain a preliminary extracted water body set.

5. A method for characterizing complex morphology of braided rivers according to claim 4, characterized in that: The step 4 comprises: Step 4.1, calculate the local minimum point-saddle point pair in the water body image in the final water body set; Step 4.2, for each pair of local minimum point-saddle point, the lowest path between local minimum point-saddle point: Step 4.3, repeat step 4.2 until all local minimum points are connected to the saddle point, and all minimum paths form a branch path network; Step 4.4, traverse the lowest path in the branch path network, delete the objects in the lowest path whose MNDWI average value is less than the MNDWI set threshold and whose branch path network length is less than the length set threshold, and update the branch path network to complete the water body extraction of the braided river and obtain the final water body set.

6. A method for characterizing complex morphology of braided rivers according to claim 5, characterized in that: The step 4.1 comprises: Step 4.11, calculate the local minimum point in the water body image in the final water body set; Step 4.12, calculate all saddle points of the water body image in the final water body set; Step 4.13: For each local minimum point, traverse all saddle points and determine the saddle point closest to the local minimum point, that is, find a pair of local minimum point-saddle point.

7. A method for characterizing complex morphology of braided rivers according to claim 5, characterized in that: The step 4.2 includes: taking the saddle point as the starting point, considering the lowest point among the 8 surrounding pixel points, calculating the distance between the point and the local lowest point, and then taking the point as the starting point, calculating the lowest point among the 8 surrounding pixel points, and repeating this step until the calculated path reaches the local lowest point.

8. A method for characterizing complex morphology of braided rivers according to claim 5, characterized in that: The step 5 comprises: Step 5.1, based on the delineated river boundary, equidistantly encrypt the coordinates of the left and right river banks to generate an encrypted boundary line, and generate the river centerline, while calculating the river width at each point on the centerline, and counting the length of the river section; Step 5.2, generate sections according to a certain step length, and count the number of branches, water surface width, and water surface width on each section by traversing the sections; Step 5.3, calculate the braid strength, water surface widening and water surface width of each section of the river; Step 5.4: Based on the designated river channel boundary, count the water area and river channel area of ​​the river section, and calculate the water area ratio; count the average sandbar area ratio of the river section; calculate the average river channel width based on the river length obtained in step 5.1; and calculate the branch channel density based on the braided strength of the river section scale.

9. A method for characterizing complex morphology of braided rivers according to claim 8, characterized in that: The step 5 also includes: collecting the morphological parameters calculated in the process of water body sampling and morphological parameter calculation into the same data table.

10. A complex morphology characterization system for braided rivers, characterized in that: include: The initial remote sensing image acquisition module is used to preliminarily screen and crop the original remote sensing images within the set time period and the set study area, set the cloud cover ratio threshold of the study area, and further screen and obtain the initial remote sensing images and preprocess them; The MNDWI index image set acquisition module is used to calculate the MNDWI index of the initial remote sensing image to obtain the MNDWI index image set; A preliminary water body set acquisition module is used to calculate the segmentation threshold of each pixel in the image in the MNDWI index image set, and to obtain a preliminary water body set by preliminary extraction of the water body of the braided river image based on the segmentation threshold. A final water body set acquisition module, which is used to acquire a final water body set based on the preliminary extracted water body set; A morphological parameter calculation module is used to perform morphological sampling and calculate morphological parameters based on the extracted final water body set and the delineated river channel boundary, and return the morphological parameter set of the river section in the form of a data table; The complex morphology characterization system for a braided river is used to execute the steps in the complex morphology characterization method for a braided river as described in any one of claims 1-9.

Images