River morphological feature extraction method and system based on urban planning

Abstract

The invention discloses a river morphological feature extraction method and system based on urban planning, and relates to the technical field of digital image processing and analysis. The method comprises the following steps: obtaining river channel data, and preprocessing the river channel data; extracting a river channel binary image based on the preprocessed river channel data; and according to the extracted river channel binary image, river width information and river center information of the river are extracted. According to the method and the system provided by the invention, the performance requirement on a computer in a data calculation process can be reduced, meanwhile, the precision of extracting the quantitative information of the river form is effectively improved so as to extract the river width information and the river center information in the river form characteristics, and the method and the system can be applied to urban planning in a city planning process. And carrying out identification and accurate positioning on regional key problems.

Description

Technical Field

[0001] The present invention relates to the technical field of digital image processing and analysis, and in particular to a method and system for extracting river morphological features based on urban planning. Background Art

[0002] River morphology affects the speed and flow of water, and plays an important role in flood prevention and disaster reduction, water resources management, river management, ecological protection, etc. The morphological characteristics of rivers mainly include curvature, width, and river islands. Currently, the above information is mainly obtained by interpreting satellite remote sensing images.

[0003] At present, river width has always been a hot topic in remote sensing hydrological research. The commonly used method is to use the center line of the river as a vertical line, and the distance between the intersection of the vertical line and the two banks is the river width. This method has a complex algorithm for determining the vertical line of the river center line or the direction of the river, and the measurement results are inaccurate. River islands are of great significance for studying the hydrodynamic processes and ecological service functions of river systems. The commonly used extraction method is to use machine learning or deep learning algorithms with the help of satellite remote sensing images to achieve river island classification. This method achieves high-precision results but often consumes a lot of computing power and has high requirements on computer performance. In addition, in the process of urban planning and design, the lack of the above quantitative information and algorithms on river morphology often affects the identification and precise positioning of key regional issues, resulting in low accuracy in river width calculation and river island extraction. Summary of the invention

[0004] In order to overcome the problems in the prior art of complex algorithms and inaccurate measurement results for the vertical line of the river centerline or the river direction, high requirements on computer performance, which affect the identification and precise positioning of regional key issues, the present invention proposes a river morphology feature extraction method and system based on urban planning, which can effectively improve the accuracy of river width calculation and river island extraction, so that in the process of urban planning, more accurate river morphology quantitative information can be extracted to identify and accurately locate regional key issues.

[0005] To achieve the purpose of the present invention, the present invention adopts the following technical solutions:

[0006] A method for extracting river morphological features based on urban planning, the method comprising the following steps:

[0007] Acquiring river data and preprocessing the river data;

[0008] Extracting river binary map based on preprocessed river data;

[0009] According to the extracted river binary map, the river width information and river island information of the river are extracted.

[0010] In the above technical solution, preprocessing the river data can improve the reliability and utilization efficiency of the river data and reduce the performance requirements for the computer; based on the preprocessed river data, the extracted river binary map can more intuitively observe the river data from the data, thereby making it more convenient to process the data to extract the river width information and river island information in the river morphological characteristics, and then provide more accurate river morphological quantitative information in the process of urban planning, so as to identify and accurately locate key regional issues.

[0011] Furthermore, the process of preprocessing the river data includes:

[0012] According to the image time span and cloud coverage, the acquired river data images are screened to obtain multispectral image data;

[0013] According to the multispectral image data, the improved normalized difference water index is calculated, and the expression is:

[0014]

[0015] Based on the calculated normalized difference water index, a normalized difference water index image is obtained;

[0016] Among them, ρ Green represents the green band, ρ SWIR represents the short-wave near-infrared band, Band3 represents the green band of the multispectral image, Band 11 Band 11 representing the shortwave near-infrared band of the multispectral image.

[0017] Furthermore, the OTSU algorithm is used to perform binary threshold segmentation on the normalized difference water index image to obtain a binary map of the river.

[0018] The OTSU algorithm represents the Otsu algorithm; in the river binary map, the river is 1 and the others are 0.

[0019] In the above technical scheme, the images of the acquired river data are screened according to the image time span and cloud coverage, which can effectively reduce the time required for data processing and eliminate some unnecessary image data in the process of river morphological feature extraction, thereby reducing the requirements for computer performance; the OTSU algorithm is used to perform binary threshold segmentation on the normalized difference water index image, which can further simplify the data processing process and increase the data calculation rate, so as to quickly obtain the river binary map required for river morphological feature extraction and improve the efficiency of river morphological feature extraction.

[0020] Furthermore, the process of extracting the river width information of the river includes:

[0021] Smoothing and denoising the river binary image;

[0022] Based on the smoothed and denoised river binary map, the river bankline and river centerline are calculated;

[0023] Calculate the river width based on the river bank line and river center line.

[0024] Furthermore, the process of calculating the river bankline and the river centerline includes:

[0025] Extract the binary image edge from the river binary image, generate a binary image matrix, and calculate the river bank line. The expression is:

[0026]

[0027] Iteratively erode the generated binary image matrix, calculate the number of transitions from non-zero values ​​to zero values ​​of pixels, and iteratively and gradually extract other pixels outside the skeleton to obtain the center line of the river. The expression is:

[0028] if A(x i ,y j-1 )+A(x i ,y j+1 )+A(x i-1 ,y i )+A(x i+1 ,y j )<4then A(x i ,y j )=0;

[0029] Among them, (x, y) represents the coordinate axis, represents the gradient value of element f, i and j both represent elements in the matrix, and A represents an element in the matrix.

[0030] Furthermore, based on the river bankline and the river centerline, the process of calculating the river width includes:

[0031] A neural network model is set to learn the data changes of the river bankline and the river centerline, and a river width model that grasps the data changes of the river bankline and the river centerline is obtained;

[0032] The river width model calculates the river centerline element (x i ,y j )The Euclidean distance of the nearest non-zero element is obtained to obtain the distance image, which is expressed as:

[0033]

[0034] Based on the calculated distance image, the Euclidean distance from the centerline of the river to the nearest river shoreline is multiplied by two to obtain the river width;

[0035] Among them, (x′ i ,y′ j ) represents the element (x i ,y j ) nearest non-zero element.

[0036] In the above technical scheme, the river binary map is smoothed and denoised, which can reduce the impact of the uneven river edges in the river binary map on the poor calculation results. The set river width model can learn to more accurately extract the river width features based on the river bank and river centerline, and improve the calculation efficiency, so as to quickly extract the river width information in the river morphological features, and then provide more accurate river morphological quantitative information in the process of urban planning, so as to identify and accurately locate key regional issues.

[0037] Furthermore, the process of extracting river island information includes:

[0038] Set the river network model to learn the river binary map and obtain the river center island model that grasps the river binary map information;

[0039] The Jiangxinzhou model uses a flood filling algorithm to fill holes in the river according to the binary map of the river channel, and traverses all elements in the binary map matrix to find the element A(x i ,y j ) value is zero, check the value of the 4-connected position element of the element, when the 4-connected element has a non-zero value, the position element is assigned 1, and the filling matrix is ​​obtained. The expression is:

[0040] if A(x i ,y j )=0and(A(x i ,y j-1 )orA(x i-1 ,y j )orA(x i+1 ,y j )orA(x i ,y j+1 ))

[0041] then A(x i ,y j )=1;

[0042] The filling matrix is ​​subtracted from the binary image matrix to obtain the Jiangxinzhou matrix. Based on the Jiangxinzhou matrix, Jiangxinzhou is transformed.

[0043] Furthermore, the river bankline and river centerline are calculated using a morphological algorithm.

[0044] In the above technical scheme, the river island model set up can more accurately extract the river island information in the river according to the river binary map information, and improve the calculation efficiency, so as to quickly extract the river width information in the river morphological characteristics, and then provide more accurate river morphological quantitative information in the process of urban planning, so as to identify and accurately locate key regional issues; the river shoreline and river centerline are calculated using morphological algorithms, which can improve the calculation efficiency and accuracy of the river shoreline and river centerline.

[0045] A river morphological feature extraction system based on urban planning, the system comprising:

[0046] Acquisition module, used to obtain river data;

[0047] A processing module, used for preprocessing the river data;

[0048] The river morphology feature extraction module extracts the river binary map based on the preprocessed river data;

[0049] The river morphology feature extraction module is also used to extract river width information and river island information of the river according to the extracted river binary map.

[0050] An electronic device comprises a memory, a processor and a computer program stored in the memory and running on the processor, wherein when the processor executes the computer program, the steps of a method for extracting river morphological features based on urban planning are implemented.

[0051] Compared with the prior art, the present invention has the following beneficial effects:

[0052] The present invention proposes a method and system for extracting river morphological features based on urban planning. First, the river data is preprocessed to improve the reliability and utilization efficiency of the river data and reduce the performance requirements for the computer. Then, based on the preprocessed river data, the river binary map extracted can more intuitively observe the river data from the data, so that it is more convenient to process the data to extract the river width information and river island information in the river morphological features, and then in the process of urban planning, provide more accurate river morphological quantitative information to identify and accurately locate key regional issues. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 A flow chart of the steps of a method for extracting river morphological features based on urban planning provided in an embodiment of the present application;

[0054] Figure 2 A schematic diagram of a structure based on a river width model and a river island model provided in an embodiment of the present application;

[0055] Figure 3 This is a diagram showing the effect of river data smoothing and noise reduction based on closed operation provided in an embodiment of the present application;

[0056] Figure 4 A result diagram of river shoreline, centerline, and intersection points calculated based on morphology provided in an embodiment of the present application;

[0057] Figure 5 A result diagram of river width calculated based on distance conversion provided in an embodiment of the present application;

[0058] Figure 6 The effect diagram before and after the river data processing based on flood filling provided in the embodiment of the present application;

[0059] Figure 7 A result map of river island extraction based on flood filling provided in an embodiment of the present application;

[0060] Figure 8 A schematic diagram of the structure of a river morphological feature extraction system based on urban planning provided in an embodiment of the present application. DETAILED DESCRIPTION

[0061] In order to facilitate the understanding of the present invention, the present invention will be described more fully below with reference to the relevant drawings. Preferred embodiments of the present invention are provided in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present invention more thorough and comprehensive.

[0062] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0063] Embodiment 1:

[0064] This embodiment provides a method for extracting river morphological features based on urban planning. Figure 1 , the method comprises the following steps:

[0065] S1: Acquire river data and pre-process the river data;

[0066] S2: Extracting river binary map based on preprocessed river data;

[0067] S3: Extract river width information and river island information of the river according to the extracted river binary map.

[0068] As a preferred embodiment, in step S1, see Figure 2 , the process of preprocessing the river data includes:

[0069] According to the image time span and cloud coverage, the acquired river data images are screened to obtain multispectral image data;

[0070] According to the multispectral image data, the improved normalized difference water index is calculated, and the expression is:

[0071]

[0072] Based on the calculated normalized difference water index, a normalized difference water index image is obtained;

[0073] Among them, ρ Green represents the green band, ρ SWIR represents the short-wave near-infrared band, Band3 represents the green band of the multispectral image, Band 11 Band 11 representing the shortwave near-infrared band of the multispectral image.

[0074] Furthermore, the OTSU algorithm is used to perform binary threshold segmentation on the normalized difference water index image to obtain a binary map of the river.

[0075] The OTSU algorithm represents the Otsu algorithm; in the river binary map, the river is 1 and the others are 0.

[0076] For example, see Figure 2 , extract river information through Google Earth Engine open source remote sensing images;

[0077] Google Earth Engine can provide open source images such as Landsat series long-term images and Sentinel-1 / 2. The example of the present invention uses Sentinel-2 multispectral images with a spatial resolution of 10 meters.

[0078] Sentinel-2 multispectral images were selected based on the image time span (2023.1.1-2023.12.31), cloud cover (≤20%), and research scope (Guangzhou administrative boundary and Beijiang River Basin). OTSU threshold segmentation was used to extract water bodies from MNDWI images.

[0079] The OTSU algorithm calculates the grayscale image based on the maximum inter-class variance and automatically finds the optimal threshold to achieve binary segmentation of the image. The binary image obtained by segmentation is the river (water body 1, non-water body 0). The process of finding the optimal threshold is to traverse all possible thresholds, calculate the variance between the two groups under each threshold, and select the threshold with the largest variance as the optimal threshold.

[0080] It can be understood that screening the images of the acquired river data according to the image time span and cloud coverage can effectively reduce the time required for data processing and eliminate some unnecessary image data in the process of river morphological feature extraction, thereby reducing the requirements for computer performance; using the OTSU algorithm to perform binary threshold segmentation on the normalized difference water index image can further simplify the data processing process and increase the rate of data calculation, thereby quickly obtaining the river binary map required for river morphological feature extraction and improving the efficiency of river morphological feature extraction.

[0081] As a preferred embodiment, in step S2, the process of extracting the river width information of the river includes:

[0082] Smoothing and denoising the river binary image;

[0083] Based on the smoothed and denoised river binary map, the river bankline and river centerline are calculated;

[0084] Calculate the river width based on the river bank line and river center line.

[0085] For example, see Figure 3 , Smoothing and denoising of river binary images. The uneven river edges of the extracted river binary images will affect the subsequent river calculation results, so the binary image is processed by closed operation morphology, first dilation and then corrosion to achieve the effect of smoothing and denoising the river edge.

[0086] Further, see Figure 2 and Figure 4 , the process of calculating river bankline and river centerline includes:

[0087] Extract the binary image edge from the river binary image, generate a binary image matrix, and calculate the river bank line. The expression is:

[0088]

[0089] Iteratively erode the generated binary image matrix, calculate the number of transitions from non-zero values ​​to zero values ​​of pixels, and iteratively and gradually extract other pixels outside the skeleton to obtain the center line of the river. The expression is:

[0090] if A(x i ,y j-1 )+A(xi ,y j+1 )+A(x i-1 ,y i )+A(x i+1 ,y j )<4then A(x i ,y j )=0;

[0091] Among them, (x, y) represents the coordinate axis, represents the gradient value of element f, i and j both represent elements in the matrix, and A represents an element in the matrix.

[0092] Shoreline calculation logic: The focus of shoreline calculation is to calculate the gradient to extract the edge of the binary image. It traverses all elements of the binary image matrix (m×n). The gradient value of element f is the joint function of the x and y axis derivatives of the element, that is, The large value of the image gradient indicates that the pixel points of the element on the x-axis and y-axis change quickly, and the element is located at the edge, that is, all points with large gradient values ​​are river banks.

[0093] Centerline calculation logic: The focus of centerline calculation is to calculate the number of transitions from non-zero values ​​to zero values ​​of pixels, and to iterate and gradually extract other pixels outside the skeleton, that is, the last iteration before the elements are eroded to an empty set. The binary image matrix (m×n) elements are iterated multiple times for erosion calculation. The logic of erosion calculation is to traverse all elements of the binary image matrix (m×n), and for each element A(x, y), the value of the element in the four fields of the element is searched, and the following formula is used to calculate if A(x i ,y j-1 )+A(x i ,y j+1 )+A(x i-1 ,y i )+A(x i+1 ,y j )<4then A(x i ,y j )=0.

[0094] Further, see Figure 2 and Figure 5 , based on the river bankline and river centerline, the process of calculating the river width includes:

[0095] A neural network model is set to learn the data changes of the river bankline and the river centerline, and a river width model that grasps the data changes of the river bankline and the river centerline is obtained;

[0096] The river width model calculates the river centerline element (x i ,y j)The Euclidean distance of the nearest non-zero element is obtained to obtain the distance image, which is expressed as:

[0097]

[0098] Based on the calculated distance image, the Euclidean distance from the centerline of the river to the nearest river shoreline is multiplied by two to obtain the river width;

[0099] Among them, (x′ i ,y′ j ) represents the element (x i ,y j ) nearest non-zero element.

[0100] For example, the embodiment of the present invention uses the distance conversion algorithm to calculate the distance image of the river shoreline binary image (shoreline is 1, other elements are 0) obtained by morphological calculation. The distance at the shoreline is 0, and other elements are reassigned to their Euclidean distance to the nearest shoreline. Distance conversion calculation logic: traverse all elements of the matrix (m×n) and find the element (x i ,y j ) the nearest non-zero element, and calculate the element and its nearest non-zero element (x′ i ,y′ j ) i,j , assign the distance to the position.

[0101] The calculation results are expressed as

[0102] like Figure 5 As shown, the embodiment of the present invention uses MATLAB to find the location of the center line of the river, and uses the location to index the distance image to obtain the distance result (the center line is the Euclidean distance to the nearest shoreline, and the others are zero), and the center line distance matrix is ​​multiplied by 2 and the data resolution (10 meters in the present invention) to obtain the river width.

[0103] It can be understood that smoothing and denoising the binary river map can reduce the impact of uneven river edges in the binary river map on poor calculation results. The set river width model can learn to more accurately extract river width features based on river shorelines and river centerlines, and improve calculation efficiency, so as to quickly extract river width information from river morphological features and provide more accurate quantitative information on river morphology in the process of urban planning, so as to identify and accurately locate key regional issues.

[0104] As a preferred embodiment, see Figure 2 and Figure 6 ,The process of extracting river island information includes:

[0105] Set the river network model to learn the river binary map and obtain the river center island model that grasps the river binary map information;

[0106] The Jiangxinzhou model uses a flood filling algorithm to fill holes in the river according to the binary map of the river channel, and traverses all elements in the binary map matrix to find the element A(x i ,y j ) value is zero, check the value of the 4-connected position element of the element, when the 4-connected element has a non-zero value, the position element is assigned 1, and the filling matrix is ​​obtained. The expression is:

[0107] if A(x i ,y j )=0and(A(x i ,y j-1 )orA(x i-1 ,y j )orA(x i+1 ,y j )orA(x i ,y j+1 ))

[0108] then A(x i ,y j )=1;

[0109] The filling matrix is ​​subtracted from the binary image matrix to obtain the Jiangxinzhou matrix. Based on the Jiangxinzhou matrix, Jiangxinzhou is transformed.

[0110] like Figure 6 As shown, the embodiment of the present invention uses the river binary map (river is 1, others are 0) to fill the holes in the river using the 4-connectivity flood filling algorithm in MATLAB. When the pixel values ​​of the 4-connected positions of the 0-valued element are all 1, the position will be reassigned to 1. Flood filling calculation logic: traverse all elements of the matrix (m×n), find the position where the element A(xi, yj) value is zero, check the element value of the 4-connected position of the element, and when the 4-connected element has a non-zero value, the element at the position is assigned 1. The calculation process is expressed as:

[0111] Jiangxinzhou Extract: Figure 2 and Figure 7 As shown, in the embodiment of the present invention, the filled matrix is ​​subtracted from the original river matrix to obtain the Jiangxin Island image, and the Jiangxin Island matrix result is converted into a vector to obtain the Jiangxin Island.

[0112] As a preferred embodiment, a morphological algorithm is used to calculate the river bankline and the river centerline.

[0113] It can be understood that the set river island model can more accurately extract the river island information in the river according to the river binary map information, and improve the calculation efficiency, so as to quickly extract the river width information in the river morphological characteristics and then provide more accurate river morphological quantitative information in the process of urban planning, so as to identify and accurately locate key regional issues; the river shoreline and river centerline are calculated using morphological algorithms, which can improve the calculation efficiency and accuracy of the river shoreline and river centerline.

[0114] In this embodiment, preprocessing the river data can improve the reliability and utilization efficiency of the river data and reduce the performance requirements for the computer; based on the preprocessed river data, the extracted river binary map can more intuitively observe the river data from the data, thereby making it more convenient to process the data to extract the river width information and river island information in the river morphological characteristics, and then provide more accurate river morphological quantitative information in the process of urban planning, so as to identify and accurately locate key regional issues.

[0115] Embodiment 2:

[0116] This embodiment provides a river morphological feature extraction system based on urban planning, see Figure 8 , the system comprising:

[0117] Acquisition module, used to obtain river data;

[0118] A processing module, used for preprocessing the river data;

[0119] The river morphology feature extraction module extracts the river binary map based on the preprocessed river data;

[0120] The river morphology feature extraction module is also used to extract river width information and river island information of the river according to the extracted river binary map.

[0121] Embodiment three:

[0122] This embodiment provides an electronic device, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the processor executes the computer program, the steps of a method for extracting river morphological features based on urban planning are implemented.

[0123] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for extracting river morphological features based on urban planning, characterized in that: The method comprises the following steps: Acquiring river data and preprocessing the river data; Extracting river binary map based on preprocessed river data; According to the extracted river binary map, the river width information and river island information of the river are extracted.

2. The method for extracting river morphological features based on urban planning according to claim 1, characterized in that: The process of preprocessing the river data includes: According to the image time span and cloud coverage, the acquired river data images are screened to obtain multispectral image data; According to the multispectral image data, the improved normalized difference water index is calculated, and the expression is: Based on the calculated normalized difference water index, a normalized difference water index image is obtained; Among them, ρ Green represents the green band, ρ SWIR represents the short-wave near-infrared band, Band3 represents the green band of the multispectral image, Band 11 Band 11 representing the shortwave near-infrared band of the multispectral image.

3. The method for extracting river morphological features based on urban planning according to claim 2 is characterized in that: The OTSU algorithm is used to perform binary threshold segmentation on the normalized difference water index image to obtain a river binary map. The OTSU algorithm represents the Otsu algorithm; in the river binary map, the river is 1 and the others are 0.

4. The method for extracting river morphological features based on urban planning according to claim 3 is characterized in that: The process of extracting river width information of a river includes: Smoothing and denoising the river binary image; Based on the smoothed and denoised river binary map, the river bankline and river centerline are calculated; Calculate the river width based on the river bank line and river center line.

5. The method for extracting river morphological features based on urban planning according to claim 4 is characterized in that: The process of calculating river banks and river centerlines includes: Extract the binary image edge from the river binary image, generate a binary image matrix, and calculate the river bank line. The expression is: Iteratively erode the generated binary image matrix, calculate the number of transitions from non-zero values ​​to zero values ​​of pixels, and iteratively and gradually extract other pixels outside the skeleton to obtain the center line of the river. The expression is: if A(x i ,y j-1 )+A(x i ,y j+1 )+A(x i-1 ,y i )+A(x i+1 ,y j )<4then A(x i ,y j )=0; Among them, (x, y) represents the coordinate axis, represents the gradient value of element f, i and j both represent elements in the matrix, and A represents an element in the matrix.

6. The method for extracting river morphological features based on urban planning according to claim 5, characterized in that: Based on the river bank line and the river center line, the process of calculating the river width includes: A neural network model is set to learn the data changes of the river bankline and the river centerline, and a river width model that grasps the data changes of the river bankline and the river centerline is obtained; The river width model calculates the river centerline element (x i ,y j )The Euclidean distance of the nearest non-zero element is obtained to obtain the distance image, which is expressed as: Based on the calculated distance image, the Euclidean distance from the centerline of the river to the nearest river shoreline is multiplied by two to obtain the river width; Among them, (x′ i ,y′ j ) represents the element (x i ,y j ) nearest non-zero element.

7. The method for extracting river morphological features based on urban planning according to claim 6, characterized in that: The process of extracting river island information includes: Set the river network model to learn the river binary map and obtain the river center island model that grasps the river binary map information; The Jiangxinzhou model uses a flood filling algorithm to fill holes in the river according to the binary map of the river channel, and traverses all elements in the binary map matrix to find the element A(x i ,y j ) value is zero, check the value of the 4-connected position element of the element, when the 4-connected element has a non-zero value, the position element is assigned 1, and the filling matrix is ​​obtained. The expression is: if A(x i ,and j )=0and(A(x i ,and j-1 )orA(x i-1 ,and j )orA(x i+1 ,and j )orA(x i ,and j+1 )) then A(x i ,y j )=1; The filling matrix is ​​subtracted from the binary image matrix to obtain the Jiangxinzhou matrix. Based on the Jiangxinzhou matrix, Jiangxinzhou is transformed.

8. The method for extracting river morphological features based on urban planning according to claim 5, characterized in that: The river bankline and river centerline are calculated using morphological algorithms.

9. A river morphology feature extraction system based on urban planning, characterized in that: The system comprises: Acquisition module, used to obtain river data; A processing module, used for preprocessing the river data; The river morphology feature extraction module extracts the river binary map based on the preprocessed river data; The river morphology feature extraction module is also used to extract river width information and river island information of the river according to the extracted river binary map.

10. An electronic device, characterized in that: The method comprises a memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the processor executes the computer program, the steps of the method according to any one of claims 1 to 8 are implemented.

Images