CN112016536B - Method and system for rapidly extracting levee flood break information in plain river network area based on remote sensing - Google Patents
Method and system for rapidly extracting levee flood break information in plain river network area based on remote sensing Download PDFInfo
- Publication number
- CN112016536B CN112016536B CN202011161071.1A CN202011161071A CN112016536B CN 112016536 B CN112016536 B CN 112016536B CN 202011161071 A CN202011161071 A CN 202011161071A CN 112016536 B CN112016536 B CN 112016536B
- Authority
- CN
- China
- Prior art keywords
- water body
- disaster
- levee
- edge
- burst
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/13—Satellite images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Astronomy & Astrophysics (AREA)
- Remote Sensing (AREA)
- Image Processing (AREA)
Abstract
The invention provides a remote sensing-based rapid extraction method and system for levee flood and flood break information in a plain river network area, wherein the method comprises the steps of obtaining multi-temporal remote sensing image data and generating a thematic map according to an extraction result, and further comprises the following steps: extracting and analyzing the burst water body and generating a burst water body file; performing edge detection according to the original image, extracting the levee and generating a change edge file; and performing superposition verification on the burst water body file and the change edge file, and calculating parameters.
Description
Technical Field
The invention relates to the technical field of hydraulic simulation, in particular to a method and a system for rapidly extracting levee flood break information in a plain river network area based on remote sensing.
Background
Flood disasters occur more and more frequently in the world at present, and the flood disasters which occur every year can bring serious threats to national economy and the economic and property safety of people of all countries. Due to the burst property, the occurrence time and the uncertainty of the range of the flood disaster, the prevention and the treatment of the flood disaster are very difficult.
One of the important prevention and control measures for flood is to construct levees in river bank areas with frequent flood disasters so as to reduce the threat of flood to river bank lands. When flood occurs, the levee is often burst due to overlarge intensity of the flood, so that the flood submerges the land in the levee and threatens the economic and property safety of people. At present, a great deal of research is carried out on flood disaster monitoring by a remote sensing method, but because the randomness and the discreteness of the spatial distribution of the levees are strong, the distribution range is wide, and the large-range remote sensing monitoring and extraction specially aiming at the levees are difficult, a method for comparing systems for monitoring and extracting the information of the levees after flood breakdown does not exist.
The Welch Feihui, the Liuwen, the Welch Wentai, the Wentai, the residual waves and the Sun Source 'shape parameter-based remote sensing image' same-spectrum foreign object 'target distinction' are published in the fourteenth national image graphics academic conference discourse in 2008, 5 months, and the article aims at expressing different attributes of ground object targets and distribution conditions thereof through the height difference (reflecting the spectral characteristics of the ground objects) and the spatial variation (reflecting the spatial distribution of the ground objects) of pixel gray values of the remote sensing image. In remote sensing image classification, the phenomenon of same-spectrum foreign matter and same-object different-spectrum is one of the main reasons influencing the classification precision. In order to inhibit images of the images and improve the classification precision, the remote sensing image classification idea based on object knowledge is set forth, namely, the high-resolution remote sensing image is utilized to have distribution characteristics, shape characteristics, topological characteristics, texture characteristics and the like besides spectral characteristics. When in classification, the shape characteristics of the ground objects are increased, so that the classification precision of the images is improved, and the problem of 'same-spectrum foreign objects' is solved to a certain extent. This article uses an object-oriented approach to process high-resolution remote sensing images, and extracts a specific study object by segmenting it using different features of the object. Among the shape features proposed in this article, there are many parameters for expressing the shape features, which are cumbersome to use, and the most important shape index is defined as an index of 1 when the figure is a square, which is not flexible enough in applying whether the figure definition is regular or not.
Wanfang data of 29 days 12 and 12 months 2015 discloses a Wanyan paper flood area rapid extraction method research based on multi-source data, and the paper researches cloud and shadow removal methods suitable for the two sensors according to data characteristics of HJ1 and landsat 8, and further researches a flood area rapid extraction method on a multi-source remote sensing image. The remote sensing data selected by the method is an optical image, and the removal of cloud layers and shadows specific to the optical image is discussed at a large extent. Since the flood disaster area is usually accompanied by thick cloud cover and rainfall, the method of using optical image can not completely eliminate cloud cover and shadow.
Disclosure of Invention
In order to solve the technical problems, the invention provides a method and a system for rapidly extracting levee flood break information in a plain river network area based on remote sensing. By analyzing the shape parameters of the flood submerging water body, regular dam breaking submerging water bodies, namely the water bodies are broken and determined, are extracted from the large-range flood submerging water bodies. The edge of the levee with phase change at any time, namely the changing edge, is extracted by respectively carrying out edge detection and superposition analysis on the original images in a certain range around the burst water body. And performing superposition verification on the obtained breach water body file and the change edge file, and calculating parameters such as the length of the breach, the position of the breach and the like.
The invention aims to provide a remote sensing-based rapid extraction method for levee flood break information in a plain river network area, which comprises the steps of obtaining multi-temporal remote sensing image data and generating a thematic map according to an extraction result, and is characterized by further comprising the following steps of:
step 1: extracting and analyzing the burst water body and generating a burst water body file;
step 2: performing edge detection according to the original image, extracting the levee and generating a change edge file;
and step 3: and performing superposition verification on the burst water body file and the change edge file, and calculating parameters.
Preferably, the acquiring the multi-temporal remote sensing image data includes selecting a microwave remote sensing image with a minimum time difference before and after occurrence of a disaster as a pre-disaster image and a post-disaster image respectively.
In any of the above schemes, preferably, the acquiring the multi-temporal remote sensing image data includes preprocessing the multi-temporal remote sensing image data by at least one of application of a track file, radiometric calibration, filtering, multi-view, and ortho-correction, so as to obtain a preprocessing influence.
In any of the above schemes, preferably, the step 1 comprises the following sub-steps:
step 11: extracting water bodies in the pre-disaster image and the post-disaster image by using a threshold value method;
step 12: carrying out comparative analysis, and extracting the original water body and the newly added water body by superposition analysis;
step 13: and analyzing the shape parameters of the newly added water body elements to distinguish the irregular water body from the regular water body.
In any of the above schemes, preferably, the threshold method is to take a peak and a valley in the gray histogram as a threshold, and the image spot with the gray value lower than the threshold of the peak and the valley is the water body.
In any of the above schemes, preferably, the step 12 includes performing superposition analysis on the extracted pre-disaster water body and the extracted post-disaster water body, defining a superposed part of the two as an original water body, and defining a part of the post-disaster water body that is not superposed with the pre-disaster water body as a newly added water body, that is, a submerged water body in the current flood disaster, and storing an original water body file and a newly added water body file respectively.
In any of the above aspects, preferably, the calculation formula of the shape parameter is
Wherein P is the perimeter of the ground feature unit, and A is the area of the ground feature unit.
In any of the above schemes, preferably, the step 13 includes selecting appropriate shape parameter values through a test, extracting regular elements by using a threshold method, importing perimeter and area data into an attribute table, and storing the perimeter and area data as a crash water body file.
In any of the above schemes, preferably, the step 2 includes the following sub-steps:
step 21: selecting a monitoring range;
step 22: performing transition type edge analysis on the image before the disaster; carrying out ridge type edge analysis on the post-disaster image;
step 23: and extracting the levee.
In any of the above schemes, preferably, the step 21 includes performing buffer processing on the obtained water body to generate a mask file including the water body within a certain range.
In any of the above schemes, it is preferable that the pre-disaster original image and the post-disaster original image are clipped into a partial image capable of performing edge analysis using the mask file.
In any of the above schemes, it is preferable that the method for detecting the transitional edge is that in the pre-disaster image, the two sides of the levee are respectively the water body and the land, because the backscattering coefficient of the water body is lower than that of the levee and the land, the gray level difference between the water body and the levee and the land on the remote sensing image is large, and the levee and the land are regarded as a whole as compared with the water body because the gray level difference is small, so that the gradient change of the gray level of the path of the water body-the levee-the land is from low to high, and the levee which is determined is extracted by performing transitional edge analysis on the local image before the disaster.
In any of the above schemes, preferably, the ridge-type edge detection method is that in the post-disaster image, because the levee breaks, flood water submerges the ground inside the levee, water bodies are arranged on both sides of the levee, both sides of the levee appear dark on a remote sensing image, the levee is not covered by the water body and still appears bright, the gray level gradient change corresponding to the path of the water body-the levee-the water body is low-high-low, and an obvious ridge-type edge is formed, and the levee with the break is extracted by performing ridge-type edge analysis on the post-disaster local image.
In any of the above schemes, preferably, the method for extracting the levee is to perform superposition analysis on the transition type edge in the extracted pre-disaster image and the ridge type edge in the post-disaster image, define the edge which is overlapped, i.e., the edge which is changed from the transition type to the ridge type, as the levee after the break, and save the overlapped part as the change edge file.
In any of the above schemes, preferably, the method for stack verification is to stack the regular water body file, i.e., the breach water body file and the change edge file, extract the edge of the overlapped part and store the edge as the levee file.
In any of the above schemes, preferably, the parameter calculating method calculates the perimeter of the element of the breach water body and the length of the edge of the levee, and calculates the difference between the perimeter and the length of the edge of the levee, namely the length of the breach; and defining the center position of the burst as a burst coordinate, and introducing the burst coordinate into a burst water body attribute table.
The invention also provides a system for rapidly extracting flood break information of the levee in the plain river network area based on remote sensing, which comprises a data acquisition module for acquiring multi-temporal remote sensing image data, an image generation module for generating a thematic map according to an extraction result, and the system further comprises the following modules:
the water body extraction module: the method is used for extracting and analyzing the burst water body and generating a burst water body file;
the levee extraction module: the levee is extracted according to the original image to generate a change edge file;
a parameter calculation module: and the method is used for performing superposition verification on the burst water body file and the change edge file and calculating parameters.
Preferably, the acquiring the multi-temporal remote sensing image data includes selecting a microwave remote sensing image with a minimum time difference before and after occurrence of a disaster as a pre-disaster image and a post-disaster image respectively.
In any of the above schemes, preferably, the acquiring the multi-temporal remote sensing image data includes preprocessing the multi-temporal remote sensing image data by at least one of application of a track file, radiometric calibration, filtering, multi-view, and ortho-correction, so as to obtain a preprocessing influence.
In any of the above aspects, preferably, the method for extracting the burst water body comprises the following sub-steps:
step 11: extracting water bodies in the pre-disaster image and the post-disaster image by using a threshold value method;
step 12: carrying out comparative analysis, and extracting the original water body and the newly added water body by superposition analysis;
step 13: and analyzing the shape parameters of the newly added water body elements to distinguish the irregular water body from the regular water body.
In any of the above schemes, preferably, the threshold method is to take a peak and a valley in the gray histogram as a threshold, and the image spot with the gray value lower than the threshold of the peak and the valley is the water body.
In any of the above schemes, preferably, the step 12 includes performing superposition analysis on the extracted pre-disaster water body and the extracted post-disaster water body, defining a superposed part of the two as an original water body, and defining a part of the post-disaster water body that is not superposed with the pre-disaster water body as a newly added water body, that is, a submerged water body in the current flood disaster, and storing an original water body file and a newly added water body file respectively.
In any of the above aspects, preferably, the calculation formula of the shape parameter is
Wherein P is the perimeter of the ground feature unit, and A is the area of the ground feature unit.
In any of the above schemes, preferably, the step 13 includes selecting appropriate shape parameter values through a test, extracting regular elements by using a threshold method, importing perimeter and area data into an attribute table, and storing the perimeter and area data as a crash water body file.
In any of the above schemes, preferably, the method for extracting the levee comprises the following substeps:
step 21: selecting a monitoring range;
step 22: performing transition type edge analysis on the image before the disaster; carrying out ridge type edge analysis on the post-disaster image;
step 23: and extracting the levee.
In any of the above schemes, preferably, the step 21 includes performing buffer processing on the obtained water body to generate a mask file including the water body within a certain range.
In any of the above schemes, it is preferable that the pre-disaster original image and the post-disaster original image are clipped into a partial image capable of performing edge analysis using the mask file.
In any of the above schemes, it is preferable that the method for detecting the transitional edge is that in the pre-disaster image, the two sides of the levee are respectively the water body and the land, because the backscattering coefficient of the water body is lower than that of the levee and the land, the gray level difference between the water body and the levee and the land on the remote sensing image is large, and the levee and the land are regarded as a whole as compared with the water body because the gray level difference is small, so that the gradient change of the gray level of the path of the water body-the levee-the land is from low to high, and the levee which is determined is extracted by performing transitional edge analysis on the local image before the disaster.
In any of the above schemes, preferably, the ridge-type edge detection method is that in the post-disaster image, because the levee breaks, flood water submerges the ground inside the levee, water bodies are arranged on both sides of the levee, both sides of the levee appear dark on a remote sensing image, the levee is not covered by the water body and still appears bright, the gray level gradient change corresponding to the path of the water body-the levee-the water body is low-high-low, and an obvious ridge-type edge is formed, and the levee with the break is extracted by performing ridge-type edge analysis on the post-disaster local image.
In any of the above schemes, preferably, the method for extracting the levee is to perform superposition analysis on the transition type edge in the extracted pre-disaster image and the ridge type edge in the post-disaster image, define the edge which is overlapped, i.e., the edge which is changed from the transition type to the ridge type, as the levee after the break, and save the overlapped part as the change edge file.
In any of the above schemes, preferably, the method for stack verification is to stack the regular water body file, i.e., the breach water body file and the change edge file, extract the edge of the overlapped part and store the edge as the levee file.
In any of the above schemes, preferably, the parameter calculating method calculates the perimeter of the element of the breach water body and the length of the edge of the levee, and calculates the difference between the perimeter and the length of the edge of the levee, namely the length of the breach; and defining the center position of the burst as a burst coordinate, and introducing the burst coordinate into a burst water body attribute table.
The invention provides a method and a system for quickly extracting levee flood break information in a plain river network area based on remote sensing, which can quickly and accurately extract the levee break information caused by flood disasters and provide powerful support for disaster prevention and relief. In the method, the special spatial characteristics of levee bursting and the special change of the levee between different time phases are utilized to extract the levee bursting information from different angles, so that the accuracy of extracting the bursting information is improved; the specific shape of the burst water body is identified by using the shape parameters, so that the influence of a large amount of information in an original image is avoided, the calculated amount is reduced, and more background interference is avoided; by using different states of different time phase edges, errors generated by edge detection on a single image are effectively avoided.
Drawings
Fig. 1 is a flowchart of a preferred embodiment of the method for rapidly extracting levee flood break information in the flat river network area based on remote sensing according to the present invention.
Fig. 2 is a block diagram of a preferred embodiment of the remote sensing-based rapid extraction system for levee flood break information in the flat river network area according to the present invention.
Fig. 3 is a flowchart of an embodiment of a water body extraction method of the method for rapidly extracting levee flood break information in the flat river network area based on remote sensing according to the invention.
Fig. 4 is a flowchart of an embodiment of the levee extraction method of the method for rapidly extracting levee flood break information in the flat river network area based on remote sensing according to the present invention.
Fig. 5 is a system work flow diagram of another preferred embodiment of the method for rapidly extracting levee flood break information in the flat river network area based on remote sensing according to the invention.
Detailed Description
The invention is further illustrated with reference to the figures and the specific examples.
Example one
As shown in fig. 1 and 2, step 100 is executed, and the data obtaining module 200 obtains multi-temporal remote sensing image data, and performs preprocessing on the multi-temporal remote sensing image data by using at least one operation of applying a track file, radiometric calibration, filtering, multi-view and orthorectification, so as to obtain a preprocessing influence. Acquiring multi-temporal remote sensing image data comprises selecting a microwave remote sensing image with the minimum time difference before and after a disaster as a pre-disaster image and a post-disaster image respectively.
Step 110 is executed, the water body extraction module 210 extracts and analyzes the burst water body, and generates a burst water body file. As shown in fig. 3, step 111 is executed to extract the water body in the pre-disaster image and the post-disaster image by using a threshold value method. The threshold method is that the peak and the valley are taken in the gray level histogram as threshold values, and the pattern spots with gray values lower than the threshold values of the peak and the valley are the water body.
And step 112, performing comparative analysis, and performing superposition analysis to extract the original water body and the newly added water body. Performing superposition analysis on the extracted pre-disaster water body and the extracted post-disaster water body, defining the superposed part of the two water bodies as an original water body, defining the part of the post-disaster water body which is not superposed with the pre-disaster water body as a newly added water body, namely the submerged water body in the flood disaster, and respectively storing an original water body file and a newly added water body file
And 113, analyzing the shape parameters of the newly added water body elements to distinguish irregular water bodies from regular water bodies, selecting appropriate shape parameter values through tests, extracting the regular elements by using a threshold method, importing perimeter and area data into an attribute table, and storing the perimeter and area data as a burst water body file. The calculation formula of the shape parameter is
Wherein P is the perimeter of the ground feature unit, and A is the area of the ground feature unit.
And executing the step 120, the levee extraction module 220 performs edge detection according to the original image, extracts the levee, and generates a change edge file. As shown in fig. 4, step 121 is executed to select a monitoring range, perform buffer processing on the obtained burst water body, and generate a mask file including the burst water body within a certain range; cutting the original image before disaster and the original image after disaster into local images capable of performing edge analysis by using the mask file
Step 122 is executed to perform transition type edge analysis on the pre-disaster image. The method for detecting the jump-type edge comprises the steps that in an image before a disaster, water and soil are respectively arranged on two sides of a levee, the backscattering coefficient of the water is lower than that of the levee and the soil, the gray level difference between the water and the levee and the soil on a remote sensing image is large, the levee and the soil can be regarded as a whole due to the fact that the gray level difference is small compared with the water, therefore, the gradient change of the gray level of the path of the water, the levee and the soil is from low to high, and the levee which is determined to be broken is extracted by performing jump-type edge analysis on a local image before the disaster.
Step 123 is executed to perform ridge-type edge analysis on the post-disaster image. The ridge type edge detection method is characterized in that in an image after disaster, because the polder is broken, flood water bodies submerge the soil in the polder, the water bodies are arranged on the two sides of the polder, the two sides of the polder appear dark colors on a remote sensing image, the polder is not covered by the water bodies and still appears bright colors, the gray level gradient change corresponding to the path of the water bodies, namely the polder and the water bodies, namely the gray level gradient change is low-high-low, the obvious ridge type edge is formed, and the polder with the break is extracted by performing ridge type edge analysis on a local image after disaster.
Step 124 is executed to extract the levee. The method for extracting the levee comprises the steps of conducting superposition analysis on a transition type edge in the extracted pre-disaster image and a roof ridge type edge in the post-disaster image, defining the edge which is overlapped, namely the edge changed from the transition type to the roof ridge type, as the levee after burst, and storing the overlapped part as a change edge file.
Executing step 130, the parameter calculating module 230 performs overlay verification on the breach water body file and the change edge file, and calculates parameters. The superposition verification method is to superpose the regular water body file, namely the burst water body file and the change edge file, extract the edge of the superposed part and store the edge as the levee file. Calculating the perimeter of the elements of the burst water body and the edge length of the levee by a parameter calculating method, and calculating the difference between the perimeter and the edge length of the levee to obtain the burst length; and defining the center position of the burst as a burst coordinate, and introducing the burst coordinate into a burst water body attribute table.
In step 140, the image generation module 240 generates a thematic map according to the extraction result.
Example two
The system selects a proper multi-temporal remote sensing image manually, takes the preprocessed image as input data, and outputs the length of the levee breach, the area of the breach water body and the center point coordinate of the breach water body after a series of analysis and processing. As shown in fig. 5 below.
1. Multi-temporal remote sensing image
1.1 remote sensing image acquisition
Monitoring flood disasters, firstly, acquiring accurate time of flood disasters, and selecting microwave remote sensing images with minimum time difference before and after the disasters as pre-disaster images and post-disaster images respectively based on the officially reported disaster occurrence time.
1.2 remote sensing image preprocessing
The obtained radar remote sensing image cannot be directly subjected to image processing, and a certain necessary preprocessing process is required to be carried out on the remote sensing image, wherein the preprocessing process comprises operations such as application of track files, radiometric calibration, filtering processing, multi-view and orthometric correction, and therefore an original image capable of being further processed is obtained.
2. Burst water body
2.1 Water extraction
And extracting water bodies in the pre-disaster image and the post-disaster image by using a threshold value method.
In the radar remote sensing image, an original image obtained through preprocessing is a gray level image, and a calm water surface can be regarded as specular reflection, and the backscattering coefficient is smaller than backscattering coefficients of other ground objects, so that a water body is dark or black in the radar remote sensing image. And the flood disaster places are mostly big rivers, rivers and lakes, the proportion of the water body in the image is large, and the gray levels of the target water body and the background are obviously different, so that the gray level histogram shows obvious double peaks. And taking the peak valley as a threshold value, and obtaining the water body by using the pattern spots with the gray values lower than the threshold value of the peak valley. The method is used for extracting the water body before the disaster and the water body after the disaster respectively.
2.2 comparative analysis
And (5) performing superposition analysis to extract the original water body and the newly added water body.
And performing superposition analysis on the extracted pre-disaster water body and the extracted post-disaster water body, defining the superposed part of the two water bodies as an original water body, defining the part, which is not superposed with the pre-disaster water body, in the post-disaster water body as a newly-added water body, and respectively extracting the two water bodies as separate files. The newly added water body is the submerged water body in the flood disaster.
2.3 shape parameter analysis
And analyzing the shape parameters of the newly added water body elements to distinguish an irregular water body (natural water body) and a regular water body (burst water body).
And performing combined treatment on the water body before the disaster and the water body after the disaster to obtain an original water body file and a newly added water body file, and defining the newly added water body as a set of the natural water body and the burst water body inside the burst levee. After a flood disaster occurs, the natural water body follows the natural law, flows to the low-lying part of the terrain and is accumulated, and irregular sawtooth-shaped spreading is presented on the remote sensing image; the levee of the burst water body can separate the burst water body from other water bodies in shape, and the burst water body presents a regular and smooth figure which is the same as the levee and is obviously different from the natural water body in shape. Assuming a plurality of patterns having the same area, the longer the circumference, the more irregular the pattern, i.e., the larger the ratio of the circumference to the area, the more irregular the pattern. And analyzing the relationship between the perimeter and the area of the element by analyzing the shape coefficient of the newly added water body, and finding a proper threshold value to distinguish the irregular water body from the regular water body.
The shape factor expression method used here is to define the shape factor of a circle as 1 with the circle as the most regular shape, and the formula:
wherein P is the perimeter of the ground feature unit, and A is the area of the ground feature unit. When the element is a circle, the shape factor F is 1; when the elements are squares, the shape factor F is 4/pi; when the elements are regular triangles, the shape factor F is
The larger the form factor, the more irregular the element. After the test, a proper shape parameter value can be selected, regular elements are extracted by using a threshold value method, perimeter and area data are imported into an attribute table, and the data are stored as a burst water body file.
3. Levee
And performing buffer area treatment on the obtained bursting water body to generate a mask file in a certain range including the bursting water body.
Performing transition type edge analysis on the image before the disaster; and (5) carrying out ridge type edge analysis on the post-disaster image.
After the flood disaster occurs, because the levee breaks and flood water enters the interior of the levee, ground features around the levee present inconsistent remote sensing images before and after the flood occurrence time on the remote sensing image.
3.1 select monitoring Range
3.1.1 buffer processing
A flooded file is obtained.
And performing buffer area treatment on the obtained bursting water body to generate a mask file in a certain range including the bursting water body.
3.1.2 cropping the original image
And obtaining a local image before disaster and a local image after disaster.
And cutting the original image before the disaster and the original image after the disaster into local images capable of carrying out edge analysis by using a mask file.
3.2 edge analysis
3.2.1 transition type edge detection
In the image before the disaster, the two sides of the levee are respectively a water body and a land. The backscattering coefficient of the water body is lower than that of the levee and the land, so that the gray level difference between the water body and the levee and the land on the remote sensing image is large. The levee and the land have smaller gray level difference and can be regarded as a whole compared with the water body, so that the gradient change of the gray level of the path of the water body, the levee and the land is from low to high. The position of the levee is the edge of the gray level change, and the type of the levee is a transition type edge. The levee for the break can be extracted by performing transition type edge analysis on the local image before the disaster.
3.2.2 Ridge-type edge detection
In the post-disaster image, because the levee breaks, flood water submerges the land inside the levee, the water bodies are arranged on both sides of the levee, the two sides of the levee appear dark on the remote sensing image, and the levee is not covered by the water body and still appears bright. The gray level gradient change corresponding to the path of the water body, the levee and the water body is low-high-low, and the levee-water polder is obvious in ridge-shaped edge. The levee with the break can be extracted by carrying out ridge type edge analysis on the local image after the disaster.
3.3 levee extraction
In summary, the change of the edge type of the levee in the remote sensing image at any time is as follows: the transition edge with larger gray scale difference on two sides is changed into a ridge edge with consistent gray scale on two sides. And other ground object types submerged by the flood water body present coherent natural water bodies, and the edge type of the water body before and after the disaster is a transition type edge. Therefore, by detecting the change, i.e., by performing superposition analysis on the transition type edge in the image before disaster and the ridge type edge in the image after disaster, the edge which is overlapped, i.e., changed from the transition type to the ridge type, can be defined as the levee after the break. Saving the overlay as a variant edge file
4. Overlay verification, calculating parameters
4.1 overlay verification, reduce errors
And superposing the regular water body file, namely the burst water body file and the change edge file, extracting the edge of the superposed part and storing the edge as a levee file (line file), and reducing errors caused by edge analysis.
4.2 calculating parameters
Calculating the perimeter of the elements of the burst water body and the edge length of the levee, and calculating the difference between the perimeter of the elements and the edge length of the levee to obtain the burst length; and defining the center position of the burst as a burst coordinate, and introducing the burst coordinate into a burst water body attribute table.
5. Outputting the result
And generating a thematic map according to the extraction result, wherein the basic format of the thematic map is fixed.
5.1 thematic map names
The output result of the system is a remote sensing monitoring thematic map of the dam break radar in the region, and the region name is manually and externally input.
5.2 image content
And scaling and adapting the local image to a fixed picture frame, marking the area of the burst water body by using the levee file, and marking the center point of the burst according to the position information of the burst in the attribute table of the burst water body. Importing an auxiliary layer, comprising: administrative divisions at all levels of county, city and province, and the marked characters of main rivers.
5.3 auxiliary elements
According to the drawing frame, the necessary auxiliary elements such as a drawing, a scale and a compass are inserted.
5.3.1 illustration
And generating a proper scale according to each element in the frame. The content comprises the following steps: marking the center position of the breach, marking all levels of administrative districts, marking the levee, marking the submersion range of the breach water body and the like.
5.3.2 Scale
Insert the appropriate scale in meters or kilometers.
5.3.3 North arrow
Inserting the north arrow, and adjusting the direction to point to the north direction.
5.4 text
And generating necessary texts, such as image sources, imaging time, drawing units, the area of the breach water body, the length of the breach, the coordinates of the breach and the like.
5.5 output image
And outputting the image as an editable interface, manually adjusting the thematic map, or adding other elements.
For a better understanding of the present invention, the foregoing detailed description has been given in conjunction with specific embodiments thereof, but not with the intention of limiting the invention thereto. Any simple modifications of the above embodiments according to the technical essence of the present invention still fall within the scope of the technical solution of the present invention. In the present specification, each embodiment is described with emphasis on differences from other embodiments, and the same or similar parts between the respective embodiments may be referred to each other. For the system embodiment, since it basically corresponds to the method embodiment, the description is relatively simple, and for the relevant points, reference may be made to the partial description of the method embodiment.
Claims (3)
1. A method for rapidly extracting levee flood break information in a plain river network area based on remote sensing comprises the steps of obtaining multi-temporal remote sensing image data and generating a thematic map according to an extraction result, and is characterized by further comprising the following steps:
the acquiring of the multi-temporal remote sensing image data comprises selecting a microwave remote sensing image with the minimum time difference before and after occurrence of a disaster as a pre-disaster image and a post-disaster image respectively;
step 1: extracting and analyzing the burst water body and generating a burst water body file, comprising the following substeps:
step 11: extracting water bodies in the pre-disaster image and the post-disaster image by using a threshold value method;
step 12: performing comparative analysis, performing superposition analysis to extract an original water body and a newly added water body, performing superposition analysis on the extracted pre-disaster water body and the extracted post-disaster water body, defining the superposed part of the two as the original water body, and defining the part of the post-disaster water body which is not superposed with the pre-disaster water body as the newly added water body, namely the submerged water body in the flood disaster at this time, and respectively storing an original water body file and a newly added water body file;
step 13: analyzing the shape parameters of the newly added water body elements to distinguish irregular water bodies from regular water bodies, selecting proper shape parameter values through tests, extracting the regular elements by using a threshold value method, importing perimeter and area data into an attribute table, and storing the perimeter and area data as a burst water body file;
step 2: carrying out edge detection according to an original image, extracting an levee and generating a change edge file, and comprising the following substeps:
step 21: selecting a monitoring range;
step 22: performing transition type edge analysis on the image before the disaster; carrying out ridge type edge analysis on the post-disaster image;
step 23: extracting the levee, wherein the extraction method of the levee comprises the steps of performing superposition analysis on a transition edge in the extracted pre-disaster image and a ridge edge in the post-disaster image, defining the edge which is overlapped, namely changed from the transition type to the ridge type, as the levee after burst, and storing the overlapped part as a change edge file;
and step 3: performing superposition verification on the breach water body file and the change edge file, and calculating parameters, wherein the superposition verification method comprises the steps of superposing a regular water body file, namely the breach water body file and the change edge file, extracting the edge of the superposed part and storing the edge as an levee file; the parameter calculating method calculates the perimeter of the elements of the burst water body and the edge length of the levee, and calculates the difference between the perimeter and the edge length of the levee to obtain the burst length; and defining the center position of the burst as a burst coordinate, and introducing the burst coordinate into a burst water body attribute table.
2. The method for rapidly extracting levee flood break information in flat river network areas based on remote sensing as claimed in claim 1, wherein the calculation formula of the shape parameter is
Wherein P is the perimeter of the ground feature unit, and A is the area of the ground feature unit.
3. The utility model provides a quick extraction system of levee flood burst information in plain river network area based on remote sensing, includes the data acquisition module that is used for acquireing multi-temporal remote sensing image data and the image generation module that generates the thematic map according to the extraction result, acquire multi-temporal remote sensing image data including selecting the microwave remote sensing image that the time difference is minimum before the calamity takes place as image before the calamity and image after the calamity respectively, its characterized in that still includes following module:
the water body extraction module: the method is used for extracting and analyzing the burst water body and generating a burst water body file, and the extraction method of the burst water body comprises the following sub-steps:
step 11: extracting water bodies in the pre-disaster image and the post-disaster image by using a threshold value method;
step 12: performing comparative analysis, performing superposition analysis to extract an original water body and a newly added water body, performing superposition analysis on the extracted pre-disaster water body and the extracted post-disaster water body, defining the superposed part of the two as the original water body, and defining the part of the post-disaster water body which is not superposed with the pre-disaster water body as the newly added water body, namely the submerged water body in the flood disaster at this time, and respectively storing an original water body file and a newly added water body file;
step 13: analyzing the shape parameters of the newly added water body elements to distinguish irregular water bodies from regular water bodies, selecting proper shape parameter values through tests, extracting the regular elements by using a threshold value method, importing perimeter and area data into an attribute table, and storing the perimeter and area data as a burst water body file;
the levee extraction module: the method for extracting the levee comprises the following substeps:
step 21: selecting a monitoring range;
step 22: performing transition type edge analysis on the image before the disaster; carrying out ridge type edge analysis on the post-disaster image;
step 23: extracting the levee, wherein the extraction method of the levee comprises the steps of performing superposition analysis on a transition edge in the extracted pre-disaster image and a ridge edge in the post-disaster image, defining the edge which is overlapped, namely changed from the transition type to the ridge type, as the levee after burst, and storing the overlapped part as a change edge file;
a parameter calculation module: the method for the superposition verification comprises the steps of superposing regular water body files, namely the burst water body file and the change edge file, extracting the edge of a superposed part and storing the edge as an levee file; the parameter calculating method calculates the perimeter of the elements of the burst water body and the edge length of the levee, and calculates the difference between the perimeter and the edge length of the levee to obtain the burst length; and defining the center position of the burst as a burst coordinate, and introducing the burst coordinate into a burst water body attribute table.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011161071.1A CN112016536B (en) | 2020-10-27 | 2020-10-27 | Method and system for rapidly extracting levee flood break information in plain river network area based on remote sensing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011161071.1A CN112016536B (en) | 2020-10-27 | 2020-10-27 | Method and system for rapidly extracting levee flood break information in plain river network area based on remote sensing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112016536A CN112016536A (en) | 2020-12-01 |
| CN112016536B true CN112016536B (en) | 2021-04-13 |
Family
ID=73528394
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011161071.1A Active CN112016536B (en) | 2020-10-27 | 2020-10-27 | Method and system for rapidly extracting levee flood break information in plain river network area based on remote sensing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112016536B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108681651A (en) * | 2018-05-29 | 2018-10-19 | 成都理工大学 | A kind of EARLY RECOGNITION method of the moraine dike ice-lake breach of no ice core |
| CN109272545A (en) * | 2018-08-27 | 2019-01-25 | 郭云扬 | A kind of method of the geometric parameter relationship and its scale of quantitatively characterizing river flood fan |
-
2020
- 2020-10-27 CN CN202011161071.1A patent/CN112016536B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108681651A (en) * | 2018-05-29 | 2018-10-19 | 成都理工大学 | A kind of EARLY RECOGNITION method of the moraine dike ice-lake breach of no ice core |
| CN109272545A (en) * | 2018-08-27 | 2019-01-25 | 郭云扬 | A kind of method of the geometric parameter relationship and its scale of quantitatively characterizing river flood fan |
Non-Patent Citations (2)
| Title |
|---|
| Menghua Li et al..Retrieval of historical surface displacements of the Baige landslide from time-series SAR observations for retrospective analysis of the collapse event.《Remot Sensing of Environment》.2020,第240卷1-11. * |
| 何兴祥等.冻错曲冰湖溃决风险及坝体稳定性研究.《四川水利发电》.2020,第39卷(第4期),8-12. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112016536A (en) | 2020-12-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Hagenaars et al. | On the accuracy of automated shoreline detection derived from satellite imagery: A case study of the sand motor mega-scale nourishment | |
| Yamazaki et al. | Development of a global~ 90 m water body map using multi-temporal Landsat images | |
| CN107145874B (en) | Ship target detection and identification method in complex background SAR image | |
| Qiao et al. | 55-year (1960–2015) spatiotemporal shoreline change analysis using historical DISP and Landsat time series data in Shanghai | |
| Smeeckaert et al. | Large-scale classification of water areas using airborne topographic lidar data | |
| Pardo-Pascual et al. | Automatic extraction of shorelines from Landsat TM and ETM+ multi-temporal images with subpixel precision | |
| Billa et al. | Semi-automated procedures for shoreline extraction using single RADARSAT-1 SAR image | |
| Souza-Filho et al. | Use of synthetic aperture radar for recognition of Coastal Geomorphological Features, land-use assessment and shoreline changes in Bragança coast, Pará, Northern Brazil | |
| CN113177183B (en) | Seawater pollution monitoring and early warning method and system based on ocean remote sensing image | |
| Faour | Evaluating urban expansion using remotely-sensed data in Lebanon | |
| Hu et al. | Monitoring coastline variations in the Pearl River Estuary from 1978 to 2018 by integrating Canny edge detection and Otsu methods using long time series Landsat dataset | |
| CN114821355B (en) | Coastline automatic identification method and device | |
| Liu et al. | Automated extraction of tidal creeks from airborne laser altimetry data | |
| McCormack et al. | A methodology for mapping annual flood extent using multi-temporal Sentinel-1 imagery | |
| Baek et al. | Oil spill detection of Kerch strait in November 2007 from dual-polarized TerraSAR-X image using artificial and convolutional neural network regression models | |
| Melkamu et al. | Detecting and mapping flood inundation areas in Fogera-Dera Floodplain, Ethiopia during an extreme wet season using Sentinel-1 data | |
| CN112016536B (en) | Method and system for rapidly extracting levee flood break information in plain river network area based on remote sensing | |
| Wu et al. | Coastal reclamation shaped narrower and steeper tidal flats in Fujian, China: Evidence from time-series satellite data | |
| Natih et al. | Study on shoreline changes using Landsat imagery in Sangsit Region, Bali Province | |
| CN108875636A (en) | A kind of diameter radar image sea land dividing method based on subsidiary discriminant strategy | |
| Zeng et al. | Urban water body detection from the combination of high-resolution optical and SAR images | |
| CN114219800A (en) | Flooding detection method based on SAR image | |
| Yu et al. | Semi-Automatic Extraction of The Threshold Segmentation of Coastline Based on Coastline Type | |
| Hua et al. | The research of artificial shoreline extraction based on airborne LIDAR data | |
| Patel et al. | Identifying probable submergence area of Surat City using digital elevation model and Geographical Information System |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |