CN117274821A - Multi-polarization SAR farmland flood detection method and system considering rainfall influence - Google Patents
Multi-polarization SAR farmland flood detection method and system considering rainfall influence Download PDFInfo
- Publication number
- CN117274821A CN117274821A CN202311541115.7A CN202311541115A CN117274821A CN 117274821 A CN117274821 A CN 117274821A CN 202311541115 A CN202311541115 A CN 202311541115A CN 117274821 A CN117274821 A CN 117274821A
- Authority
- CN
- China
- Prior art keywords
- polarization
- farmland
- sar
- image
- formula
- 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.)
- Granted
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 78
- 230000010287 polarization Effects 0.000 claims abstract description 74
- 230000008859 change Effects 0.000 claims abstract description 41
- 238000000034 method Methods 0.000 claims abstract description 36
- 239000011159 matrix material Substances 0.000 claims abstract description 17
- 238000007781 pre-processing Methods 0.000 claims abstract description 13
- 238000012545 processing Methods 0.000 claims abstract description 4
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 22
- 125000003275 alpha amino acid group Chemical group 0.000 claims description 18
- 230000006870 function Effects 0.000 claims description 13
- 238000001914 filtration Methods 0.000 claims description 9
- 238000010586 diagram Methods 0.000 claims description 8
- 230000003313 weakening effect Effects 0.000 claims description 5
- 230000009977 dual effect Effects 0.000 claims description 4
- 230000011218 segmentation Effects 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 230000017105 transposition Effects 0.000 claims description 3
- 238000006243 chemical reaction Methods 0.000 claims description 2
- 238000003709 image segmentation Methods 0.000 claims description 2
- 239000003550 marker Substances 0.000 claims description 2
- 238000011084 recovery Methods 0.000 claims description 2
- 238000010276 construction Methods 0.000 claims 1
- 239000000126 substance Substances 0.000 abstract 1
- 230000009466 transformation Effects 0.000 abstract 1
- 230000008901 benefit Effects 0.000 description 5
- 238000013135 deep learning Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000011524 similarity measure Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000005388 cross polarization Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002787 reinforcement Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/89—Radar or analogous systems specially adapted for specific applications for mapping or imaging
- G01S13/90—Radar or analogous systems specially adapted for specific applications for mapping or imaging using synthetic aperture techniques, e.g. synthetic aperture radar [SAR] techniques
- G01S13/9021—SAR image post-processing techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/72—Data preparation, e.g. statistical preprocessing of image or video features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
- G06V10/765—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects using rules for classification or partitioning the feature space
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/84—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using probabilistic graphical models from image or video features, e.g. Markov models or Bayesian networks
- G06V10/85—Markov-related models; Markov random fields
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/188—Vegetation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Remote Sensing (AREA)
- Software Systems (AREA)
- Multimedia (AREA)
- Radar, Positioning & Navigation (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Electromagnetism (AREA)
- Probability & Statistics with Applications (AREA)
- Radar Systems Or Details Thereof (AREA)
- Image Processing (AREA)
Abstract
The invention relates to the technical field of radar image processing, in particular to a multi-polarization SAR farmland flood detection method and system taking rainfall influence into consideration, comprising the steps of preprocessing dual-time-phase multi-polarization SAR image data to generate a polarization covariance matrix; respectively calculating the intensity ratio of the co-polarized and cross-polarized channelsHLTThe method comprises the steps of counting measures, constructing SAR image polarization difference measures, and calculating pixel by pixel to generate a polarization difference map; multi-polarization SAR image transformation by using Markov random field modelPerforming chemical detection; according to the change detection result, reducing the interference of mountain shadow on the detection result by using a digital elevation model, and extracting a farmland disaster result; and performing geocoding on the farmland disaster result under the SAR coordinate system, and outputting a final farmland flooding detection result under the geographic coordinate system. According to the method, false detection caused by rainfall can be effectively avoided, the applicability of farmland flood detection of the multi-polarization SAR image in rainfall weather is improved, and the detection precision of the polarization SAR flood is improved.
Description
Technical Field
The invention relates to the technical field of radar image processing, in particular to a multi-polarization SAR farmland flood detection method and system considering rainfall influence.
Background
Synthetic Aperture Radar (SAR) remote sensing is an active microwave imaging remote sensing means, can penetrate cloud and fog, and has the advantages of all-day and all-weather operation. Because rainfall is often accompanied in the occurrence process of farmland flood disasters, the traditional optical remote sensing technology faces difficulty in acquiring ground surface images. Therefore, monitoring of flood disasters using SAR remote sensing technology has shown great potential. In recent years, along with the mass application of observation platforms such as multi-polarization SAR satellites, the acquisition cost of the polarization SAR data is continuously reduced, and the application of the multi-polarization SAR data is also becoming wider and wider. Compared to monopolar SAR images, multi-polarized SAR images have multiple polarization channels including co-polarization and cross-polarization, which enables it to extract more information.
In the flood disaster detection process, change detection is a common means. Change detection is a process in the remote sensing field that analyzes changes in a property of a particular region over two or more time periods. Currently, the multi-polarization SAR-based change detection method mainly comprises the following steps:
(1) Polarized SAR image change detection method based on similarity measure
The detection method based on the similarity measurement is an unsupervised change detection strategy, and the basic flow is to construct a difference map by utilizing multi-temporal images, and then identify a change region by analyzing the difference map. Since the covariance matrix of the multi-temporal polarized SAR image follows the complex Wishare distribution, metrics such as SRW distance, geodesic distance, kullback-Leibler divergence, etc. can be deduced. The similarity measure can effectively evaluate the similarity degree of the dual/multi-temporal polarized SAR images, construct a difference map between different time intervals and further realize the change detection of a target area.
(2) PCC-based change detection method
The post-classification comparison method (Post Classification Comparison, PCC) is a method of classifying images to be detected first and performing change detection by comparing the results of the double/multiple temporal classification. Compared with an unsupervised method based on similarity measure, the method adopts a supervised mode, can directly obtain the change category, and the single-phase accuracy depends on the single-phase classification result, so that the condition that the change detection accuracy is reduced due to single-phase classification errors is easy to occur; meanwhile, due to the supervision characteristic, the influence of the subjective factors is larger.
(3) Deep learning-based change detection method
In recent years, with the development of neural network technologies such as convolutional neural networks and cyclic neural networks, a change detection method based on deep learning has also been greatly advanced. The deep learning model can automatically extract image features and has stronger space-time information expression capability. Current telemetry data sets are difficult to satisfy because deep learning requires a large number of well-marked samples. Therefore, the weakly supervised deep learning method and the reinforcement learning method constitute the main research direction of the change detection method based on deep learning.
The above-mentioned change detection methods all have better performance in specific fields, but because these methods do not consider the rainfall influence possibly encountered in the flood disaster change detection, the change detection accuracy still needs to be improved.
Disclosure of Invention
The invention aims to provide a multi-polarization SAR farmland flood detection method considering rainfall influence, so as to solve the problems in the background technology.
In order to achieve the above purpose, the invention provides a multi-polarization SAR farmland flood detection method considering rainfall influence, comprising the following steps:
(1) Preprocessing dual-phase multi-polarization SAR image data: registering, multi-vision and filtering multi-polarization SAR data of two time phases to generate a polarization covariance matrixC；
(2) Generating a polarization difference diagram: based on the polarization covariance matrix generated in step (1)CRespectively calculating the intensity ratio of the co-polarized and cross-polarized channels and the Hotelling-Laururi trace statistical measure, and constructing the multi-polarized SAR image polarization difference measure considering rainfall influenceDFurther calculating pixel by pixel to generate a polarization difference map;
(3) Multi-polarization SAR image change detection based on Markov random field: pre-segmenting the polarization difference map obtained in the step (2) by using an Ojin method, inputting a binary matrix obtained by segmentation into a Markov random field model as a tag field, and detecting multi-polarization SAR image change by using the Markov random field model;
(4) Extracting a farmland flood disaster area: according to the change detection result of the step (3), weakening the interference of mountain shadows on the detection result by using a digital elevation model, and extracting a farmland disaster result;
(5) Outputting a farmland flood disaster detection result: and after extracting the farmland disaster result, performing geocoding on the farmland disaster result under the SAR coordinate system, and outputting a final farmland flooding detection result under the geographic coordinate system.
Further, in the step (1), the data preprocessing of the dual-phase multi-polarization SAR image specifically includes: registering the image of the second time phase to the first time phase, and performing multi-view and refinement Lee filtering processing to generate a polarization covariance matrixC：
1）；
In the formula (1), the amino acid sequence of the formula (1),Hrepresents the operation of conjugate transposition,krepresenting a multi-polarized scattering vector.
Further, in the step (2), a polarization intensity ratio is constructed based on the intensities of the co-polarized and cross-polarized channels：
2）；
In the formula 2), the amino acid sequence of the formula (II),intensity value representing the homopolar channel, +.>Representing the intensity values of the cross-polarized channels; based on the difference of the water body pixels under the influence of rainfall on the two different polarization channels, the water body pixels are added with the water body pixels>The water body area disturbed by rain fall can be identified.
Further, polarization covariance matrix based on two time phasesAnd->The statistical measure of Hotelling-Laurushi can be calculated>：
3）；
In the formula 3), the amino acid sequence of the formula (III),representing trace operations +.>Representing maximum value operation>Representing the polarization dimension +.>In the case of dual polarization +.>；
Taking statistical measure of Hotelling-LaurushiAnd polarization ratio->Is used for constructing polarization difference measure considering rainfall effectD：
4）；
Calculating a difference measure from 4) pixel-by-pixelPolarization difference map can be obtained>。
Further, in the step (3), the polarization difference map is based on the oxford methodThe method comprises the following specific steps: polarization difference map->Pre-segmentation is performed to obtain an unchanged area and a changed area, and a polarization difference diagram is calculated by using the Ojin method ++>Inter-class variance corresponding to gray value +.>Polarization difference map is +.>The segmentation is performed as a result of a change or no change, and the image segmentation result can be used as an initial marker field of the Markov random field model.
Further, calculating a polarization difference map by using the Ojin methodWhen the gray value is the gray value, the inter-class variance is +.>The calculation formula of (2) is as follows:
5）；
in the formula 5), the amino acid sequence of the formula,representing the proportion of the pixels of the two types of samples, namely the changed type and the unchanged type, to the total pixel when the threshold value T is taken>Representing the average gray value of the pixels of both types of samples, respectively>Representing the total average gray value of the image.
Further, the polarization difference diagram is formedInputting a Markov random field model as observation field data, fusing eight neighborhood information, and constructing an energy function +.>And generating a priori probabilities of Markov random field +.>：
6）；
In the formula 6), the amino acid sequence of the formula,is a normalization factor, ++>Is the temperature parameter->Is an exponential function;
then, assume a polarization difference measureObeying a gaussian distribution, based on varying classes and unchanged
Average value of similar pixelsSum of variances->Calculating likelihood functions of the two classes respectively +.>：
7）；
In the formula (7), the amino acid sequence of the formula (I),is an exponential function; furthermore, the posterior probability can be calculated>：
8）；
Based onJudging the change type pixel by pixel according to the principle of maximum probability, updating the tag field, iterating the processes from the formula 6) to the formula 8), until the maximum iteration number is reached, and outputting a change detection result.
Further, in the step (4), the extracting of the disaster-affected area of the farmland specifically includes:
firstly, extracting the terrain gradient based on a digital elevation model, and setting a gradient threshold valueAnd->The method comprises the steps of carrying out a first treatment on the surface of the Wherein, the gradient is defined to be smaller than the gradient threshold +.>The pixel of (2) is water, and the gradient is defined to be between gradient threshold +.>And->The pixels in between are weak water body candidate points;
then searching whether a strong candidate point exists in a neighborhood window of each weak candidate point, and if so, updating the pixel as the strong candidate point;
then traversing all weak candidate point pixels, and screening out pixels belonging to the water body based on the neighborhood information;
and finally, combining the strong candidate point results extracted twice to generate a farmland disaster result.
Further, in the step (5), the conversion of the obtained farmland disaster recovery result from the SAR coordinate system to the geographic coordinate system is specifically as follows:
firstly, calculating the initial position of an image point based on the length of radar echo time and echo Doppler characteristics, and generating an initial lookup table of a slant range pixel and a geographic pixel;
and then, calculating an analog SAR intensity image based on the digital elevation model, matching the analog SAR intensity image with the real SAR intensity image to obtain a matching polynomial, generating a fine lookup table, completing image positioning based on the lookup table, giving geographic coordinates to each SAR disaster detection result image pixel, and finally realizing drawing and release of farmland flood disaster detection results.
The invention also provides a multi-polarization SAR farmland flood detection system considering rainfall influence, which comprises the following steps:
the polarized SAR data preprocessing module is used for carrying out registration, multi-view and filtering preprocessing operation on the dual-phase multi-polarized SAR data;
the polarization difference map generation module is used for calculating the polarization difference measure of the preprocessed image so as to generate a polarization difference map;
the polarization SAR change detection module is used for dividing the image into a changed type and an unchanged type;
the farmland flood disaster area extraction module is used for weakening the influence of mountain shadows by utilizing digital elevation model data based on a Markov random field model and extracting a disaster area;
and the geocoding module is used for giving geographic coordinates to each SAR disaster detection result image element and outputting a final farmland flooding detection result under a geographic coordinate system.
Compared with the prior art, the invention has the following beneficial effects:
according to the multi-polarization SAR farmland flood detection method considering rainfall influence, the rainfall influence is described by using the intensity ratio of homopolarization and cross polarization channels, the variation detection distance is constructed by combining Hotelling-Law trace statistical measure, and the Markov random field model is introduced to realize farmland flood disaster detection. In addition, the method has the advantages of low calculation cost, high result reliability and the like, has wide application prospect, can be applied to the field of farmland flood disaster detection, and has potential to influence other fields depending on remote sensing image change detection. The method of the invention has practical value, important commercial value and social benefit, and makes a certain contribution to the progress of flood disaster detection technology.
In addition to the objects, features and advantages described above, the present invention has other objects, features and advantages. The invention will be described in further detail with reference to the accompanying drawings.
Drawings
The accompanying drawings are included to provide a further understanding of embodiments of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain, without limitation, the embodiments of the invention. In the drawings:
FIG. 1 is a flow chart of one embodiment of the present invention;
FIG. 2 is a schematic diagram of a result of detecting a change in a flood disaster in an embodiment of the present invention; wherein, (a) is a pre-disaster multi-polarization SAR pseudo-color synthetic image, (b) is a post-disaster multi-polarization SAR pseudo-color synthetic image, (c) is a polarization difference image, and (d) is a flood disaster monitoring result;
FIG. 3 is a schematic diagram of a detection system according to an embodiment of the present invention.
Detailed Description
The present invention will be described in detail below with reference to the embodiments shown in the drawings, but it should be understood that the embodiments are not limited to the present invention, and functional, method, or structural equivalents and alternatives according to the embodiments are within the scope of protection of the present invention by those skilled in the art.
Referring to fig. 1 and fig. 2, an embodiment of the present invention provides a method for detecting multi-polarization SAR farmland flooding in consideration of rainfall effect, in which a Sentinel-1 dual-polarization spaceborne SAR system is used as an example for detecting farmland flooding disaster variation in an yang lake region, and the region has more ground features, including farmland, city, lake and the like, and the situation is complex; the post-disaster images are greatly affected by rain fall. The multi-polarization SAR farmland flood detection method comprises the following steps:
step 1, double time phases are multiplePreprocessing polarized SAR image data: preprocessing operations such as registration, multi-view and filtering are carried out on multi-polarization SAR data of the first time phase and the second time phase to obtain a polarization covariance matrixC. The registration operation uses the first time phase as a reference, registers the image of the second time phase to the first time phase, and adopts an exquisite Lee filter for filtering. The resulting polarization covariance matrixCAs shown in formula 1):
1）；
in the formula (1), the amino acid sequence of the formula (1),Hrepresents the operation of conjugate transposition,krepresenting a multi-polarized scattering vector.
Step 2, generating a polarization difference diagram: based on the polarization covariance matrix generated in step 1CRespectively calculating the intensities of the co-polarized and cross-polarized channels to construct a polarization intensity ratio：
2）；
In the formula 2), the amino acid sequence of the formula (II),intensity value representing the homopolar channel, +.>Representing the intensity values of the cross-polarized channels; based on the difference of the water body pixels under the influence of rainfall on the two different polarization channels, the water body pixels are added with the water body pixels>The water body area disturbed by rain fall can be identified. Further, polarization covariance matrix based on two phases +.>And->A statistical measure of Hotelling-Lawley trace (HLT) can be calculated>：
3）；
In the formula 3), the amino acid sequence of the formula (III),representing trace operations +.>Representing maximum value operation>Representing the polarization dimension +.>In the case of dual polarization +.>. Afterwards, take->Statistical measure->And polarization ratio->Is used for constructing polarization difference measure considering rainfall effectD：
4）；
Calculating a difference measure from 4) pixel-by-pixelDA polarization difference map can be obtained。
Step 3, multi-polarization SAR image change detection based on Markov random field (Markov Random Field, MRF): pre-segmenting the polarization difference map constructed in the step (2) by using an Ojin method, and firstly calculating the polarization difference mapInter-class variance corresponding to gray value +.>The specific formula is as follows:
5）；
in the formula 5), the amino acid sequence of the formula,representing the proportion of the pixels of the two types of samples, namely the changed type and the unchanged type, to the total pixel when the threshold value T is taken>Average gray value of pixels of two types of samples representing changed type and unchanged type, < + >>Representing the total average gray value of the image. By inter-class variance->Taking the corresponding threshold T at the maximum value as a threshold, taking a binary matrix obtained by dividing the threshold T as a label field, inputting the label field into a Markov random field model, fusing eight neighborhood information, and constructing an energy function +.>And generating a priori probabilities of Markov random field +.>：
6）；
In the formula 6), the amino acid sequence of the formula,is a normalization factor, ++>Is the temperature parameter->Is an exponential function. Then, assume the polarization difference measure +.>Obeying Gaussian distribution, based on mean value of pixels of changed type and unchanged type->Sum of variances->Calculating likelihood functions of the two classes respectively +.>：
7）；
In the formula (7), the amino acid sequence of the formula (I),is an exponential function; furthermore, the posterior probability can be calculated>：
8）；
Based onJudging the change type pixel by pixel according to the maximum probability principle, and updating the tag fieldAnd iterating the above process until the maximum iteration number is reached, and outputting a change detection result.
Step 4, extracting a farmland flood disaster area: according to the change detection result of the step 3, extracting the terrain gradient by using a digital elevation model (Digital Elevation Model, DEM) and setting a gradient threshold valueAnd->. Wherein, the gradient is defined to be smaller than the gradient threshold +.>The pixel of (2) belongs to the water body, and the gradient is between gradient threshold +.>And->The pixels in the middle can belong to water bodies and also can belong to mountain shadows, and the gradient of the pixels in the part is in the overlapping area of the two types of sample histogram distribution, which is called as a weak water body candidate point; then searching whether a strong candidate point exists in a neighborhood window of each weak candidate point, and if so, updating the pixel as the strong candidate point; then traversing all weak candidate point pixels, and screening out pixels belonging to the water body based on the neighborhood information; and finally, combining the strong candidate point results extracted twice to generate a farmland disaster result.
Step 5, outputting a farmland flood disaster detection result: after a disaster area is extracted, performing geocoding on a disaster result under an SAR coordinate system, calculating an initial position of an image point based on the length of radar echo time and echo Doppler characteristics, and generating an initial lookup table of an oblique image element and a geographic image element; then, calculating a simulated SAR intensity image based on the DEM, matching the simulated SAR intensity image with a real SAR intensity image to obtain a matching polynomial, generating a fine lookup table, finishing image positioning based on the lookup table, endowing geographic coordinates to each SAR disaster detection result image pixel, and outputting a final farmland flooding detection result under a geographic coordinate system.
As shown in fig. 3, the present invention further provides a multi-polarization SAR farmland flooding detection system considering rainfall effect, comprising:
the polarized SAR data preprocessing module 1 is used for carrying out registration, multi-view and filtering preprocessing operation on the dual-phase multi-polarized SAR data;
the polarization difference map generating module 2 is used for calculating the polarization difference measure of the preprocessed image so as to generate a polarization difference map;
a polarization SAR change detection module 3 for classifying the image into a changed class and an unchanged class;
the farmland flood disaster area extraction module 4 is used for weakening the influence of mountain shadows by utilizing digital elevation model data based on a Markov random field model and extracting a disaster area;
and the geocoding module 5 gives geographic coordinates to each SAR disaster detection result image element and outputs a final farmland flooding detection result under a geographic coordinate system.
The modules included in the multi-polarization SAR farmland flood detection system provided in the present embodiment belong to the function implementation modules of the multi-polarization SAR farmland flood detection method in the above embodiment, and the specific working principle and implementation manner are the same as those described in the above method embodiment, and are not repeated here.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. The multi-polarization SAR farmland flood detection method considering rainfall influence is characterized by comprising the following steps:
(1) Preprocessing dual-phase multi-polarization SAR image data: registering, multi-vision and filtering preprocessing are carried out on multi-polarization SAR data of two time phases to generate a polarization covariance matrixC；
(2) Difference in polarizationAnd (3) generating a diagram: based on the polarization covariance matrix generated in step (1)CRespectively calculating the intensity ratio of the co-polarized and cross-polarized channels and the Hotelling-Laururi trace statistical measure, and constructing the multi-polarized SAR image polarization difference measure considering rainfall influenceDFurther calculating pixel by pixel to generate a polarization difference map;
(3) Multi-polarization SAR image change detection based on Markov random field: pre-segmenting the polarization difference map obtained in the step (2) by using an Ojin method, inputting a binary matrix obtained by segmentation into a Markov random field model as a tag field, and detecting multi-polarization SAR image change by using the Markov random field model;
(4) Extracting a farmland flood disaster area: according to the change detection result of the step (3), weakening the interference of mountain shadows on the detection result by using a digital elevation model, and extracting a farmland disaster result;
(5) Outputting a farmland flood disaster detection result: and after extracting the farmland disaster result, performing geocoding on the farmland disaster result under the SAR coordinate system, and outputting a final farmland flooding detection result under the geographic coordinate system.
2. The method for detecting the flood of the multi-polarization SAR farmland according to claim 1, wherein in the step (1), the data preprocessing of the dual-temporal multi-polarization SAR image is specifically: registering the image of the second time phase to the first time phase, and performing multi-view and refinement Lee filtering processing to generate a polarization covariance matrixC：
1）；
In the formula (1), the amino acid sequence of the formula (1),Hrepresents the operation of conjugate transposition,krepresenting a multi-polarized scattering vector.
3. The method for detecting flood in multi-polarization SAR farmland according to claim 2, wherein in said step (2), the construction is based on the intensities of co-polarized and cross-polarized channelsRatio of polarization intensity：
2）；
In the formula 2), the amino acid sequence of the formula (II),intensity value representing the homopolar channel, +.>Representing the intensity values of the cross-polarized channels; based on the difference of the water body pixels under the influence of rainfall on the two different polarization channels, the water body pixels are added with the water body pixels>The water body area disturbed by rain fall can be identified.
4. The multi-polarization SAR farmland flooding detection method according to claim 3, wherein the polarization covariance matrix based on two phasesAnd->The statistical measure of Hotelling-Laurushi can be calculated>：
3）；
In the formula 3), the amino acid sequence of the formula (III),the trace-out operation is represented by a trace-out operation,/>representing maximum value operation>Representing the dimension of polarization, in the case of full polarizationIn the case of dual polarization +.>；
Taking statistical measure of Hotelling-LaurushiAnd polarization ratio->Is used for constructing polarization difference measure considering rainfall effectD：
4）；
Calculating a difference measure from 4) pixel-by-pixelPolarization difference map can be obtained>。
5. The method for detecting flood in multi-polarization SAR farmland according to claim 4, wherein in said step (3), the polarization difference map is based on the oxford methodThe method comprises the following specific steps: polarization difference map->Pre-segmentation is performed to obtain an unchanged area and a changed area, and a polarization difference diagram is calculated by using the Ojin method ++>Inter-class variance corresponding to gray value +.>Polarization difference map is +.>The segmentation is performed as a result of a change or no change, and the image segmentation result can be used as an initial marker field of the Markov random field model.
6. The multi-polarization SAR farmland flood detection method according to claim 5, wherein said inter-class varianceWhen any threshold value T is taken, the calculation formula is as follows:
5）；
in the formula 5), the amino acid sequence of the formula,representing the proportion of the sample pixels of the changed type and the unchanged type to the total pixels when the threshold T is taken>Average gray value of pixels of two types of samples representing changed type and unchanged type, < + >>Representing the total average gray value of the image.
7. The multi-polarization SAR farmland flood detection method according to claim 5, wherein the polarization difference map is obtained byInputting a Markov random field model as observation field data, fusing eight neighborhood information, and constructing an energy function +.>And generating a priori probabilities of Markov random field +.>：
6）；
In the formula 6), the amino acid sequence of the formula,is a normalization factor, ++>Is the temperature parameter->Is an exponential function;
then, assume a polarization difference measureObeying Gaussian distribution, based on mean value of pixels of changed type and unchanged type->Sum of variances->Calculating likelihood functions of the two classes respectively +.>：
7）；
In the formula (7), the amino acid sequence of the formula (I),is an exponential function; furthermore, the posterior probability can be calculated>：
8）；
Based onJudging the change type pixel by pixel according to the principle of maximum probability, updating the tag field, iterating the formula 6) to the formula 8), until the maximum iteration number is reached, and outputting a change detection result.
8. The method for detecting multi-polarization SAR farmland flooding according to claim 1, wherein in the step (4), the extracting of the disaster-stricken area of the farmland is specifically:
firstly, extracting the terrain gradient based on a digital elevation model, and setting a gradient threshold valueAnd->The method comprises the steps of carrying out a first treatment on the surface of the Wherein, the gradient is defined to be smaller than the gradient threshold +.>The pixel of (2) is water, and the gradient is defined to be between gradient threshold +.>And->The pixels in between are weak water body candidate points;
then searching whether a strong candidate point exists in a neighborhood window of each weak candidate point, and if so, updating the pixel as the strong candidate point;
then traversing all weak candidate point pixels, and screening out pixels belonging to the water body based on the neighborhood information;
and finally, combining the strong candidate point results extracted twice to generate a farmland disaster result.
9. The method for detecting the flood of the multi-polarization SAR farmland according to claim 8, wherein in the step (5), the conversion of the obtained disaster recovery result of the farmland from the SAR coordinate system to the geographic coordinate system is specifically as follows:
firstly, calculating the initial position of an image point based on the length of radar echo time and echo Doppler characteristics, and generating an initial lookup table of a slant range pixel and a geographic pixel;
and then, calculating an analog SAR intensity image based on the digital elevation model, matching the analog SAR intensity image with the real SAR intensity image to obtain a matching polynomial, generating a fine lookup table, completing image positioning based on the lookup table, giving geographic coordinates to each SAR disaster detection result image pixel, and finally realizing drawing and release of farmland flood disaster detection results.
10. A system based on the multi-polarization SAR farmland flooding detection method of any one of claims 1-9, comprising:
the polarized SAR data preprocessing module (1) is used for carrying out registration, multi-view and filtering preprocessing operation on the dual-phase multi-polarized SAR data;
the polarization difference map generation module (2) is used for calculating the polarization difference measure of the preprocessed image so as to generate a polarization difference map;
a polarization SAR change detection module (3) for dividing the image into a changed class and an unchanged class;
the farmland flood disaster area extraction module (4) is used for weakening the influence of mountain shadows by utilizing digital elevation model data based on a Markov random field model and extracting a disaster area;
and the geocoding module (5) is used for endowing geographic coordinates to each SAR disaster detection result image pixel and outputting a final farmland flooding detection result under a geographic coordinate system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311541115.7A CN117274821B (en) | 2023-11-20 | 2023-11-20 | Multi-polarization SAR farmland flood detection method and system considering rainfall influence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311541115.7A CN117274821B (en) | 2023-11-20 | 2023-11-20 | Multi-polarization SAR farmland flood detection method and system considering rainfall influence |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117274821A true CN117274821A (en) | 2023-12-22 |
| CN117274821B CN117274821B (en) | 2024-02-02 |
Family
ID=89216362
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311541115.7A Active CN117274821B (en) | 2023-11-20 | 2023-11-20 | Multi-polarization SAR farmland flood detection method and system considering rainfall influence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117274821B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118537788A (en) * | 2024-05-10 | 2024-08-23 | 南京北斗创新应用科技研究院有限公司 | Unsupervised flood detection method and system considering space-time polarization information |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105321163A (en) * | 2014-07-31 | 2016-02-10 | 中国科学院遥感与数字地球研究所 | Method and apparatus for detecting variation region of fully polarimetric SAR (Synthetic Aperture Radar) image |
| CN112558066A (en) * | 2020-10-30 | 2021-03-26 | 西南电子技术研究所（中国电子科技集团公司第十研究所） | Dual-polarization SAR image system |
| CN113567981A (en) * | 2021-06-28 | 2021-10-29 | 中国电建集团华东勘测设计研究院有限公司 | SAR image-based flood risk area automatic extraction method |
| CN114067152A (en) * | 2022-01-14 | 2022-02-18 | 南湖实验室 | Refined flood inundated area extraction method based on satellite-borne SAR image |
| CN114265065A (en) * | 2021-12-23 | 2022-04-01 | 中国电子科技集团公司第十四研究所 | Multi-channel SAR moving target detection method fusing multi-polarization images |
| CN115019192A (en) * | 2022-05-30 | 2022-09-06 | 杭州电子科技大学 | Flood change detection method and system based on dual-channel backbone network and joint loss function |
| JP2022185296A (en) * | 2021-06-02 | 2022-12-14 | 清水建設株式会社 | Positioning method of satellite image |
| CN115984778A (en) * | 2023-01-09 | 2023-04-18 | 北京深蓝空间遥感技术有限公司 | Multi-feature optimization based method for rapidly and dynamically monitoring Sentinel-1 data in flood |
| US11747498B1 (en) * | 2022-09-01 | 2023-09-05 | Chengdu University Of Technology | Method, system, device and medium for landslide identification based on full polarimetric SAR |
| CN116977311A (en) * | 2023-08-02 | 2023-10-31 | 中国人民解放军61540部队 | Flood disaster area detection method, system, electronic equipment and storage medium |
-
2023
- 2023-11-20 CN CN202311541115.7A patent/CN117274821B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105321163A (en) * | 2014-07-31 | 2016-02-10 | 中国科学院遥感与数字地球研究所 | Method and apparatus for detecting variation region of fully polarimetric SAR (Synthetic Aperture Radar) image |
| CN112558066A (en) * | 2020-10-30 | 2021-03-26 | 西南电子技术研究所（中国电子科技集团公司第十研究所） | Dual-polarization SAR image system |
| JP2022185296A (en) * | 2021-06-02 | 2022-12-14 | 清水建設株式会社 | Positioning method of satellite image |
| CN113567981A (en) * | 2021-06-28 | 2021-10-29 | 中国电建集团华东勘测设计研究院有限公司 | SAR image-based flood risk area automatic extraction method |
| CN114265065A (en) * | 2021-12-23 | 2022-04-01 | 中国电子科技集团公司第十四研究所 | Multi-channel SAR moving target detection method fusing multi-polarization images |
| CN114067152A (en) * | 2022-01-14 | 2022-02-18 | 南湖实验室 | Refined flood inundated area extraction method based on satellite-borne SAR image |
| CN115019192A (en) * | 2022-05-30 | 2022-09-06 | 杭州电子科技大学 | Flood change detection method and system based on dual-channel backbone network and joint loss function |
| US11747498B1 (en) * | 2022-09-01 | 2023-09-05 | Chengdu University Of Technology | Method, system, device and medium for landslide identification based on full polarimetric SAR |
| CN115984778A (en) * | 2023-01-09 | 2023-04-18 | 北京深蓝空间遥感技术有限公司 | Multi-feature optimization based method for rapidly and dynamically monitoring Sentinel-1 data in flood |
| CN116977311A (en) * | 2023-08-02 | 2023-10-31 | 中国人民解放军61540部队 | Flood disaster area detection method, system, electronic equipment and storage medium |
Non-Patent Citations (4)
| Title |
|---|
| MOSER G等: "Unsupervised change detection from multichannel SAR images", IEEE GEOSCIENCE AND REMOTE SENSING LETTERS * |
| 杨萌;张弓;: "遥感图像变化区域的无监督压缩感知", 中国图象图形学报, no. 11 * |
| 王娜;张景发;: "SAR图像变化检测技术方法综述", 地壳构造与地壳应力文集, no. 01 * |
| 赵金奇;: "多时相极化SAR影像变化检测方法研究", 测绘学报, no. 04 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118537788A (en) * | 2024-05-10 | 2024-08-23 | 南京北斗创新应用科技研究院有限公司 | Unsupervised flood detection method and system considering space-time polarization information |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117274821B (en) | 2024-02-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Jiang et al. | The potential of more accurate InSAR covariance matrix estimation for land cover mapping | |
| CN107329139B (en) | Double-temporal dual-polarization interference SAR image water area extraction method | |
| CN107507200B (en) | SAR image high-precision large-range water area extraction method based on connectivity detection and noise suppression | |
| Liao et al. | Urban change detection based on coherence and intensity characteristics of SAR imagery | |
| Lê et al. | Multiscale framework for rapid change analysis from SAR image time series: Case study of flood monitoring in the central coast regions of Vietnam | |
| CN117274821B (en) | Multi-polarization SAR farmland flood detection method and system considering rainfall influence | |
| Xiang et al. | Adaptive statistical superpixel merging with edge penalty for PolSAR image segmentation | |
| CN103294792A (en) | Polarimetric SAR (synthetic aperture radar) terrain classification method based on semantic information and polarimetric decomposition | |
| Guo et al. | KD-tree-based euclidean clustering for tomographic SAR point cloud extraction and segmentation | |
| CN115019192B (en) | Flood variation detection method and system based on dual-channel backbone network and joint loss function | |
| CN114325706B (en) | Distributed scatterer filtering method | |
| CN110570462A (en) | flood inundation range automatic extraction method based on polarized radar remote sensing image | |
| Mararakanye et al. | Gully features extraction using remote sensing techniques | |
| CN116704331A (en) | Beach change information extraction method and device and electronic equipment | |
| CN117310705B (en) | Flood disaster rapid detection method based on dual-polarized SAR image | |
| Wang et al. | Modified statistically homogeneous pixels’ selection with multitemporal SAR images | |
| Dong et al. | A novel VHR image change detection algorithm based on image fusion and fuzzy C-means clustering | |
| CN113095375B (en) | North sea ice classification method based on medium-method ocean satellite scatterometer | |
| Chen et al. | Statistic Ratio Attention Guided Siamese U-Net for SAR Image Semantic Change Detection | |
| Han et al. | Runways detection based on scattering similarity and structural characteristics | |
| Liu et al. | Change detection of polarimetric SAR images applied to specific land cover type | |
| Liao et al. | Urban change detection using coherence and intensity characteristics of multi-temporal ERS-1/2 Imagery | |
| Iglesias et al. | Towards Unsupervised Flood Mapping Generation Using Automatic Thresholding and Classification Aproaches | |
| Shi et al. | Multi-scale Features Fusion Network for Unsupervised Change Detection in Heterogeneous Optical and SAR Images | |
| Wu et al. | An improved D-LinkNet method for road extraction from high resolution remote sensing images |
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 |