CN114545410A - Crop lodging monitoring method based on synthetic aperture radar dual-polarization data coherence - Google Patents
Crop lodging monitoring method based on synthetic aperture radar dual-polarization data coherence Download PDFInfo
- Publication number
- CN114545410A CN114545410A CN202210157010.0A CN202210157010A CN114545410A CN 114545410 A CN114545410 A CN 114545410A CN 202210157010 A CN202210157010 A CN 202210157010A CN 114545410 A CN114545410 A CN 114545410A
- Authority
- CN
- China
- Prior art keywords
- lodging
- crop
- polarization
- coherence
- coefficient
- 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
- 238000012544 monitoring process Methods 0.000 title claims abstract description 91
- 238000000034 method Methods 0.000 title claims abstract description 86
- 230000008859 change Effects 0.000 claims abstract description 32
- 238000009826 distribution Methods 0.000 claims abstract description 17
- 238000012216 screening Methods 0.000 claims abstract description 13
- 230000004660 morphological change Effects 0.000 claims abstract description 6
- 230000010287 polarization Effects 0.000 claims description 96
- 239000011159 matrix material Substances 0.000 claims description 79
- 238000012545 processing Methods 0.000 claims description 28
- 230000008030 elimination Effects 0.000 claims description 16
- 238000003379 elimination reaction Methods 0.000 claims description 16
- 238000004590 computer program Methods 0.000 claims description 12
- 238000004422 calculation algorithm Methods 0.000 claims description 10
- 238000012360 testing method Methods 0.000 claims description 9
- 238000012549 training Methods 0.000 claims description 9
- 238000013145 classification model Methods 0.000 claims description 8
- 230000001427 coherent effect Effects 0.000 claims description 8
- 238000003860 storage Methods 0.000 claims description 8
- 238000013507 mapping Methods 0.000 claims description 7
- 238000000354 decomposition reaction Methods 0.000 claims description 6
- 230000009977 dual effect Effects 0.000 claims description 6
- 238000012417 linear regression Methods 0.000 claims description 5
- 230000000007 visual effect Effects 0.000 claims description 5
- 238000006073 displacement reaction Methods 0.000 claims description 4
- 238000007781 pre-processing Methods 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 238000005457 optimization Methods 0.000 claims description 2
- 230000003287 optical effect Effects 0.000 abstract description 7
- 238000010586 diagram Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 238000004891 communication Methods 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 239000002689 soil Substances 0.000 description 2
- 238000001228 spectrum Methods 0.000 description 2
- 240000008042 Zea mays Species 0.000 description 1
- 235000005824 Zea mays ssp. parviglumis Nutrition 0.000 description 1
- 235000002017 Zea mays subsp mays Nutrition 0.000 description 1
- 235000005822 corn Nutrition 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000002401 inhibitory effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010238 partial least squares regression Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000003746 surface roughness Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
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
- G01S13/9023—SAR image post-processing techniques combined with interferometric techniques
-
- 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
- G01S13/904—SAR modes
- G01S13/9058—Bistatic or multistatic SAR
-
- 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
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
- G01S7/024—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using polarisation effects
-
- 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
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
- G01S7/41—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
-
- 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
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
- G01S7/41—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
- G01S7/411—Identification of targets based on measurements of radar reflectivity
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Abstract
The invention provides a crop lodging monitoring method based on synthetic aperture radar dual-polarization data coherence, which comprises the following steps: extracting backscattering coefficients before and after the operation area is laid down; determining a backscattering variation characteristic, a coherence coefficient and an average value texture characteristic; screening to obtain lodging crop monitoring sensitive characteristics in the backscattering change characteristics, the coherence coefficient and the average texture characteristics; and identifying the lodging crops by adopting a partial least square classification method to obtain the spatial distribution information of the lodging crops in the operation area. According to the method, primary difference and phase unwrapping are carried out on primary and secondary phase images provided by double-time-phase bipolar synthetic aperture radar data to generate the coherence coefficients before and after crop lodging, morphological changes before and after crop lodging are represented through phase difference, meanwhile, the second-order probability texture features based on spatial neighborhood statistics are fused to monitor crop lodging, backscattering change features and noise interference of the coherence coefficients are inhibited, and the accuracy of crop lodging monitoring results under the condition of optical data loss is improved.
Description
Technical Field
The invention relates to the technical field of agricultural informatization, in particular to a crop lodging monitoring method based on the dual-polarization data coherence of a synthetic aperture radar.
Background
Crop lodging is a common agricultural disaster phenomenon, which seriously affects the normal growth and development of crops and threatens grain safety. With the rapid development of remote sensing technology, it is more and more common to apply the remote sensing technology to obtain the disaster information of crops. According to different remote sensing data sources, the remote sensing monitoring method for lodging crops can be divided into an optical monitoring method and a radar monitoring method.
The optical monitoring method is mainly based on the space-time difference rule of lodging crops and non-lodging crops in visible light images, a series of sensitive indexes capable of distinguishing the lodging crops and the non-lodging crops are constructed, and the indexes mainly comprise: spectrum, texture, vegetation index and canopy structure characteristics; and screening the characteristics by using a characteristic screening method to obtain the optimal characteristics. And then identifying the lodging crops and the non-lodging crops in the operation area based on the most characteristic combined with a machine learning or linear regression classification method. Although researchers have developed various features and methods for identifying lodging crops based on optical data and achieved good results, the optical data are susceptible to atmospheric environment interference, and the application of the methods is limited; secondly, the existence of the phenomenon of 'foreign matter in the same spectrum' enables the identification degree of lodging crops and non-lodging crops to be reduced, and therefore the identification precision is reduced.
The Radar monitoring method is generally a technology of Synthetic Aperture Radar (SAR), which is a technology of actively transmitting microwave signals and receiving echo signals to produce a scattering image of an observation area. The SAR remote sensing technology does not depend on solar illumination, has small atmospheric attenuation, can work under any weather conditions, and has the capability of all-weather earth observation. In addition, the SAR remote sensing technology is sensitive to surface roughness and ground object geometric shapes, can provide multi-band, multi-polarization and multi-angle scattering information, and provides powerful technical support for monitoring crop lodging. At present, the traditional radar monitoring method mainly detects the change of different polarization backscattering coefficients (VV, VH, HV, HH) in time series, but the following problems exist in the researches: (1) the disaster weather is often accompanied by rainstorm, and in addition, the backscattering coefficient is sensitive to moisture, so that the scattering value is easy to distort, and the separability of lodging crops and non-lodging crops is poor; (2) the canopy structure of vegetation is comparatively complicated, and the randomness of scattering characteristic is stronger, and multiple key elements such as soil background, vegetation, water are overlapped the coupling problem in the backscattering process, are difficult to categorised lodging to the contribution of backscattering.
Disclosure of Invention
The invention provides a crop lodging monitoring method and a crop lodging monitoring system based on synthetic aperture radar dual-polarization data coherence, which are used for overcoming the defects in the prior art.
In a first aspect, the invention provides a crop lodging monitoring method based on synthetic aperture radar dual-polarization data coherence, which comprises the following steps:
collecting crop samples with different lodging degrees in an operation area, acquiring dual-polarized Synthetic Aperture Radar (SAR) remote sensing data before and after lodging, and extracting backscattering coefficients before and after the lodging in the operation area based on the dual-polarized SAR remote sensing data;
determining backscattering change characteristics, coherence coefficients and average texture characteristics before and after lodging based on the backscattering coefficients;
screening and obtaining lodging crop monitoring sensitive characteristics in the backscattering change characteristics, the coherence coefficient and the average value texture characteristics based on a random forest-recursive characteristic elimination algorithm;
and identifying the lodging crops by adopting a partial least square classification method to obtain the spatial distribution information of the lodging crops in the operation area.
According to the crop lodging monitoring method based on the coherence of the synthetic aperture radar dual-polarization data, provided by the invention, the crop samples with different lodging degrees in the operation area are collected, the dual-polarization synthetic aperture radar SAR remote sensing data before and after lodging are obtained, and the backscattering coefficients before and after the lodging of the operation area are extracted based on the dual-polarization SAR remote sensing data, and the method comprises the following steps:
determining a pre-lodging crop sample and a post-lodging crop sample, acquiring a single-view complex Sentinel-1SLC image in the dual-polarization SAR remote sensing data, and preprocessing the Sentinel-1SLC image to obtain a ground-distance multi-view Sentinel-1GRD image;
and respectively acquiring a VV polarization backscattering coefficient before the operation area is lodged, a VH polarization backscattering coefficient before the operation area is lodged, a VV polarization backscattering coefficient after the operation area is lodged and a VH polarization backscattering coefficient after the operation area is lodged from the Sentinel-1GRD image.
According to the crop lodging monitoring method based on the dual-polarized data coherence of the synthetic aperture radar, the backward scattering change characteristic, the coherence coefficient and the average texture characteristic before and after lodging are determined based on the backward scattering coefficient, and the crop lodging monitoring method comprises the following steps:
dividing the VH polarization backscattering coefficient before the operation area is laid by the VV polarization backscattering coefficient before the operation area is laid to obtain a coefficient ratio before the operation area is laid;
dividing the VH polarization backscattering coefficient after the operation area is lodged by the VV polarization backscattering coefficient after the operation area is lodged to obtain a coefficient ratio after lodging;
respectively making a difference between the VH polarization backscattering coefficient before the operation area is lodged and the VH polarization backscattering coefficient after the operation area is lodged, making a difference between the VV polarization backscattering coefficient before the operation area is lodged and the VV polarization backscattering coefficient after the operation area is lodged, and making a difference between the ratio of the coefficient before the lodging and the coefficient after the lodging, so as to obtain the backscattering variation characteristic;
constructing an average texture feature based on second-order probability statistics based on the backscattering variation feature;
and carrying out interference processing on the Sentinel-1SLC image to obtain a VV polarization coherent coefficient and a VH polarization coherent coefficient, and representing morphological changes before and after crop lodging based on phase difference information in the VV polarization coherent coefficient and the VH polarization coherent coefficient.
According to the crop lodging monitoring method based on the dual-polarized data coherence of the synthetic aperture radar, the average value texture feature based on second-order probability statistics is constructed based on the backscattering variation feature, and the method comprises the following steps:
determining a two-dimensional matrix of the Sentinel-1SLC image, and obtaining a normalized SAR image value based on the maximum value and the minimum value in the two-dimensional matrix;
based on the backscattering variation characteristics, counting a gray level co-occurrence matrix of the normalized SAR image value according to a preset direction and a preset displacement;
and calculating the average value texture characteristics of the gray level co-occurrence matrix based on the optimized preset neighborhood window.
According to the crop lodging monitoring method based on the dual-polarized data coherence of the synthetic aperture radar, the optimized preset neighborhood window comprises the following steps:
acquiring the average texture feature of the crop sample before lodging, the average texture feature of the crop sample after lodging, the covariance matrix of the crop sample before lodging and the covariance matrix of the crop sample after lodging;
determining the optimized preset neighborhood window by adopting J-M distance based on the average texture feature of the crop sample before lodging, the average texture feature of the crop sample after lodging, the covariance matrix of the crop sample before lodging and the covariance matrix of the crop sample after lodging;
the J-M distance comprises a preset feasible region, and the preset feasible region is sequentially divided into a first feasible region interval, a second feasible region interval and a third feasible region interval based on the separability degree of the textural features.
According to the crop lodging monitoring method based on the synthetic aperture radar dual-polarization data coherence, the Sentinel-1SLC image is subjected to interference processing to obtain a VV polarization coherence coefficient and a VH polarization coherence coefficient, and the method comprises the following steps:
respectively taking phase information provided by the Sentinel-1SLC images before and after lodging as a main phase and an auxiliary phase for calculating a coherence coefficient;
calculating a coordinate mapping relation between the main phase and the auxiliary phase homonymy points, and sampling an auxiliary phase image into a pixel grid of the main phase image according to the coordinate mapping relation so that the homonymy points of the main phase image and the auxiliary phase image correspond to the same ground resolution unit;
marking main and auxiliary phase values of the same spatial position in VV polarization as a pixel pair, carrying out complex conjugate multiplication on the pixel pair by pixel pair to obtain an interference complex value, and resolving a primary differential interference phase pixel by pixel based on a real part and an imaginary part of the interference complex value;
phase unwrapping is carried out on the primary differential interference phase of the VV polarization, the fuzzy whole-cycle phase 2k pi in the phase main value is recovered, and the VV polarization coherence coefficient is obtained;
marking main and auxiliary phase values at the same spatial position in VH polarization as a pixel pair, carrying out complex conjugate multiplication on the pixel pair by pixel pair to obtain an interference complex value, and resolving a primary differential interference phase pixel by pixel based on a real part and an imaginary part of the interference complex value;
and performing phase unwrapping on the primary differential interference phase of the VH polarization, and recovering the blurred whole-cycle phase 2k pi in the phase main value to obtain the coherence coefficient of the VH polarization.
According to the crop lodging monitoring method based on the dual-polarized data coherence of the synthetic aperture radar, provided by the invention, lodging crop monitoring sensitive characteristics in the backscatter change characteristics, the coherence coefficient and the average value texture characteristics are obtained by screening based on a random forest-recursive characteristic elimination algorithm, and the method comprises the following steps:
superposing the pre-lodging crop sample and the post-lodging crop sample to a Google Earth remote sensing image, and obtaining multi-sample data through visual interpretation;
extracting characteristic values of the multi-sample data based on the backscattering variation characteristic, the coherence coefficient and the average value texture characteristic to obtain a sample point characteristic data set;
and inputting the sample point characteristic data set and the corresponding sample label into a random forest-recursive characteristic elimination filter to obtain the lodging crop monitoring sensitive characteristic.
According to the crop lodging monitoring method based on the synthetic aperture radar dual-polarization data coherence, lodging crops are identified by adopting a partial least square classification method, and lodging crop spatial distribution information of the working area is obtained, and the method comprises the following steps:
performing centralized processing on the matrix of the lodging crop monitoring sensitive characteristics and the corresponding sample label set matrix to respectively obtain a sensitive characteristic principal component score matrix, a sensitive characteristic principal component decomposition load matrix and a sensitive characteristic principal component error matrix, and a sample label set principal component score matrix, a sample label set principal component decomposition load matrix and a sample label set principal component error matrix;
performing multiple linear regression on the sensitive feature principal component score matrix and the sample label set principal component score matrix to obtain an unknown pixel sensitive feature score matrix, and obtaining an unknown pixel label based on the unknown pixel sensitive feature score matrix;
and dividing the matrixes of the lodging crop monitoring sensitive characteristics and the corresponding sample label set matrixes into a training set and a testing set according to a preset proportion, inputting the training set and the testing set to a partial least squares regression classification model, and performing model tuning to obtain lodging crop distribution information of the operation area.
In a second aspect, the present invention further provides a crop lodging monitoring system based on the dual-polarized data coherence of the synthetic aperture radar, which includes:
the system comprises a first processing module, a second processing module and a third processing module, wherein the first processing module is used for collecting crop samples with different lodging degrees in an operation area, acquiring dual-polarized Synthetic Aperture Radar (SAR) remote sensing data before and after lodging, and extracting backscattering coefficients before and after the operation area lodging based on the dual-polarized SAR remote sensing data;
the second processing module is used for determining backscattering change characteristics, coherence coefficients and mean texture characteristics before and after lodging based on the backscattering coefficients;
the third processing module is used for screening and obtaining lodging crop monitoring sensitive characteristics in the backscattering change characteristics, the coherence coefficient and the average value texture characteristics based on a random forest-recursive characteristic elimination algorithm;
and the fourth processing module is used for identifying the lodging crops by adopting a partial least square classification method to obtain the spatial distribution information of the lodging crops in the operation area.
In a third aspect, the present invention further provides an electronic device, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the method for monitoring crop lodging based on the coherence of the dual-polarized data of the synthetic aperture radar as described in any one of the above.
In a fourth aspect, the present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when being executed by a processor, carries out the steps of the method for crop lodging monitoring based on the coherence of the dual polarized data of synthetic aperture radar as described in any of the above.
In a fifth aspect, the present invention also provides a computer program product comprising a computer program which, when being executed by a processor, realizes the steps of the method for monitoring crop lodging based on the coherence of the dual polarized data of the synthetic aperture radar as described in any one of the above.
According to the crop lodging monitoring method and system based on the multi-base synthetic aperture radar dual-polarization data coherence, the coherence coefficient provided by the double-temporal multi-polarization SAR single vision complex data is applied to crop lodging monitoring, morphological change characteristics before and after crop lodging are fully considered, second-order probability texture characteristics based on spatial proximity statistics are fused, backscattering change characteristics and noise interference of the coherence coefficient are suppressed to a certain extent, and the precision of a crop lodging monitoring result is further improved.
Drawings
In order to more clearly illustrate the technical solutions of the present invention or the prior art, the drawings needed for the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.
FIG. 1 is a schematic flow chart of a crop lodging monitoring method based on the dual-polarization data coherence of a synthetic aperture radar, provided by the invention;
FIG. 2 is a schematic structural diagram of a crop lodging monitoring system based on the dual-polarization data coherence of the synthetic aperture radar provided by the invention;
fig. 3 is a schematic structural diagram of an electronic device provided in the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is obvious that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Aiming at various limitations of remote sensing monitoring of lodging crops in the prior art, the invention adopts dual-polarization coherence analysis to solve the bottleneck problem of low classification result precision caused by interference of backscattering coefficients on moisture sensitivity, strong scattering characteristic randomness and the like in remote sensing monitoring of lodging crops based on the SAR interference principle.
Fig. 1 is a schematic flow chart of a crop lodging monitoring method based on synthetic aperture radar dual-polarization data coherence, as shown in fig. 1, including:
step S1, collecting crop samples with different lodging degrees in an operation area, acquiring dual-polarized Synthetic Aperture Radar (SAR) remote sensing data before and after lodging, and extracting backscattering coefficients before and after the operation area lodging based on the dual-polarized SAR remote sensing data;
firstly, collecting crop samples with different lodging degrees in an operation area, acquiring Sentinel-1 dual-polarized SAR remote sensing data before and after lodging in the operation area, and extracting a backward scattering coefficient;
the Sentinel-1 (Sentinel 1) satellite is an earth observation satellite in the European space agency Colbriy program (GMES), and consists of two polar orbit satellites A and B, wherein the sensors carried by the two satellites are SAR and belong to active microwave remote sensing satellites. The sensor carries the C band.
Step S2, determining backscattering change characteristics, coherence coefficients and mean texture characteristics before and after lodging based on the backscattering coefficients;
and then constructing a strong robustness lodging monitoring index based on the backscatter change characteristics before and after lodging and the coherence coefficient, wherein the strong robustness lodging monitoring index comprises the backscatter change characteristics, the coherence coefficient and the average texture characteristics.
Step S3, based on a random forest-recursive feature elimination algorithm, screening and obtaining lodging crop monitoring sensitive features in the backscattering variation feature, the coherence coefficient and the average value texture feature;
and (4) inputting the backscatter change characteristics, the coherence coefficient and the texture characteristics obtained in the step (S2) into a random forest-recursive characteristic elimination filter, and screening out lodging crop monitoring feeling characteristics.
And step S4, identifying the lodging crops by adopting a partial least square classification method to obtain the lodging crop spatial distribution information of the operation area.
And finally, realizing lodging area identification by using a partial least square classification method, thereby obtaining lodging crop distribution information of the whole operation area.
According to the invention, the coherence coefficient provided by the double-time-phase dual-polarization SAR single-vision complex data is applied to crop lodging monitoring, morphological change characteristics before and after crop lodging are fully considered, second-order probability texture characteristics based on spatial proximity statistics are fused, backscattering change characteristics and noise interference of the coherence coefficient are inhibited to a certain extent, and the precision of a crop lodging monitoring result is further improved.
Based on the above embodiment, the method step S1 includes:
determining a pre-lodging crop sample and a post-lodging crop sample, acquiring a single-view complex Sentinel-1SLC image in the dual-polarization SAR remote sensing data, and preprocessing the Sentinel-1SLC image to obtain a ground-distance multi-view Sentinel-1GRD image;
and respectively acquiring a VV polarization backscattering coefficient before the operation area is lodged, a VH polarization backscattering coefficient before the operation area is lodged, a VV polarization backscattering coefficient after the operation area is lodged and a VH polarization backscattering coefficient after the operation area is lodged from the Sentinel-1GRD image.
Specifically, a Sentinel-1SLC (Sentinel-1 Single Look Complex, Sentinel No. 1 Single Look Complex) image is preprocessed to obtain a Sentinel-1GRD (Sentinel-1 Ground Range Detected, Sentinel No. 1 Ground distance multiple Look) image, so that VV and VH polarization backscattering coefficients VV before lodging in an operation area can be obtainednl、VHnlAnd after lodging VV and VH polarization backscattering coefficients VVl、VHl。
Here, the SLC is a first-order product in the sentinel No. 1 satellite product, and can obtain phase and amplitude information, the phase information is a function of time, and distance measurement can be realized according to the phase information and speed, and the SLC can be used for distance measurement and deformation observation; GRD is also the first-class product in the sentinel No. 1 satellite product and has multi-view intensity data, the intensity data is related to the backscattering coefficient, and the GRD can be used for soil moisture inversion.
Based on any of the above embodiments, the method step S2 includes:
dividing the VH polarization backscattering coefficient before the operation area is lodged by the VV polarization backscattering coefficient before the operation area is lodged to obtain a coefficient ratio before lodging;
dividing the VH polarization backscattering coefficient after the operation area is lodged by the VV polarization backscattering coefficient after the operation area is lodged to obtain a coefficient ratio after lodging;
respectively making a difference between the VH polarization backscattering coefficient before the operation area is lodged and the VH polarization backscattering coefficient after the operation area is lodged, making a difference between the VV polarization backscattering coefficient before the operation area is lodged and the VV polarization backscattering coefficient after the operation area is lodged, and making a difference between the ratio of the coefficient before the lodging and the coefficient after the lodging, so as to obtain the backscattering variation characteristic;
constructing an average texture feature based on second-order probability statistics based on the backscattering variation feature;
and carrying out interference processing on the Sentinel-1SLC image to obtain a VV polarization interference pattern and a VH polarization interference pattern, and estimating and obtaining the coherence coefficient based on the VV polarization interference pattern and the VH polarization interference pattern.
Wherein, the construction of the mean texture feature based on the second-order probability statistics based on the backscatter variation feature comprises:
determining a two-dimensional matrix of the Sentinel-1SLC image, and obtaining a normalized SAR image value based on the maximum value and the minimum value in the two-dimensional matrix;
based on the backscattering variation characteristics, counting a gray level co-occurrence matrix of the normalized SAR image value according to a preset direction and a preset displacement;
and calculating the average value texture characteristics of the gray level co-occurrence matrix based on the optimized preset neighborhood window.
Wherein, the optimized preset neighborhood window comprises:
acquiring the average texture feature of the crop sample before lodging, the average texture feature of the crop sample after lodging, the covariance matrix of the crop sample before lodging and the covariance matrix of the crop sample after lodging;
determining the optimized preset neighborhood window by adopting J-M distance based on the average texture feature of the crop sample before lodging, the average texture feature of the crop sample after lodging, the covariance matrix of the crop sample before lodging and the covariance matrix of the crop sample after lodging;
the J-M distance comprises a preset feasible region, and the preset feasible region is sequentially divided into a first feasible region interval, a second feasible region interval and a third feasible region interval based on the separability degree of the textural features.
Wherein, the interference processing is performed on the Sentinel-1SLC image to obtain a VV polarization coherence coefficient and a VH polarization coherence coefficient, and the interference processing includes:
respectively taking phase information provided by the Sentinel-1SLC images before and after lodging as a main phase and an auxiliary phase for calculating a coherence coefficient;
calculating a coordinate mapping relation between the main phase and the auxiliary phase homonymy points, sampling the auxiliary phase image into a pixel grid of the main phase image according to the coordinate mapping relation, and enabling the homonymy points of the main phase image and the auxiliary phase image to correspond to the same ground resolution unit;
marking main and auxiliary phase values of the same spatial position in VV polarization as a pixel pair, carrying out complex conjugate multiplication on the pixel pair by pixel pair to obtain an interference complex value, and resolving a primary differential interference phase pixel by pixel based on a real part and an imaginary part of the interference complex value;
phase unwrapping is carried out on the primary differential interference phase of the VV polarization, the fuzzy whole-cycle phase 2k pi in the phase main value is recovered, and the VV polarization coherence coefficient is obtained;
marking main and auxiliary phase values at the same spatial position in VH polarization as a pixel pair, carrying out complex conjugate multiplication on the pixel pair by pixel pair to obtain an interference complex value, and resolving a primary differential interference phase pixel by pixel based on a real part and an imaginary part of the interference complex value;
and performing phase unwrapping on the primary differential interference phase of the VH polarization, and recovering the blurred whole-cycle phase 2k pi in the phase main value to obtain the coherence coefficient of the VH polarization.
Specifically, VV and VH polarization backscattering coefficients VV before obtaining the lodging of the operation regionnl、VHnlAnd backscattering coefficient VV after lodgingl、VHlThen, respectively carrying out quotient calculation on the two polarization data to obtain the backscattering dual polarization Ratio before and after lodgingnl、RatiolAnd (5) characterizing.
Obtaining the Ratio of the VV polarization backscattering coefficients and the VH polarization backscattering coefficients of the Sentinel-1GRD images before and after lodging respectivelynl、RatiolIs characterized in that:
polarization backscattering coefficient VV of image before lodgingnl、VHnl、RationlPolarization features (including VV) respectively corresponding to the images after lodgingl、VHl、Ratiol) Calculating difference, and constructing the backscattering variation characteristics (VV) before and after crop lodgingvar、VHvar、Ratiovar) The method comprises the following steps:
VHvar＝VHnl-VHl
VVvar＝VVnl-VVl
Ratiovar＝Rationl-Ratiol
wherein, VVvar、VHvar、RatiovarCharacterised by the variation of the backscattering coefficient, VVnlAnd VHnlVV and VH polarization backscattering coefficients, Ratio, of the pre-lodging Sentinel-1GRD data, respectivelynlIs VVnlAnd VHnlThe ratio of (A) to (B); VVlAnd VHlVV and VH polarization backscattering coefficients, Ratio, of the post-lodging Sentinel-1GRD data, respectivelylIs VVlAnd VHlThe ratio of.
Further, according to the three characteristic maps (VV)var、VHvar、Ratiovar) Constructing GLCM _ Mean (Gray-level Co-occurrrence Matrix _ Mean, Gray level Co-occurrence Matrix average) based on second-order probability statistics:
three backscatter varying features s for any position (x, y) in the imagexyGray scale domain mapping g ofxy：
Wherein s isminAnd sminFor the minimum and maximum values, g, in the two-dimensional matrix of the entire SAR imagexyIs the value of the normalized SAR image.
Here, it is necessary to count a probability matrix from an i-value pixel position to a j-value pixel position of the image in a certain calculation preset direction θ and a preset displacement d based on the mapped backscatter change characteristics, that is, a gray level co-occurrence matrix p (i, j, d, θ) in a certain preset neighborhood window:
wherein, # denotes the number of occurrences of an image value within a certain window, and the pixel intensity distribution range is [ i, j]，(x1,y1) And (x)2,y2) Respectively, first and second pixel locations adjacent in the direction theta.
According to the gray level co-occurrence matrix, calculating GLCM _ Mean characteristics of the gray level co-occurrence matrix under a certain neighborhood space window:
where N is the gray level of the image.
It should be noted that the present invention optimizes the direction angle and window size parameters required for calculating GLCM _ Mean characteristics, and adjusts the individual parameters one by one while keeping other parameters constant, according to the J-M (Jeffries-matrix) distance JM in the characteristic diagram of the samples of the lodging crops and the unbolting cropsijDetermining the optimal value of the parameter, specifically:
JMij＝2*(1-e-B)
wherein m isnlAnd mlAverage, Σ X, of textural features of unbolted and lodging crop samples, respectivelynlAnd ∑ XlCovariance matrices of texture features for samples of unbolted and lodging crops, respectively.
The JM mentioned aboveijHas a feasible region interval of [0, 2 ]]The present invention is defined as follows: when JMij<1, indicating that the separability of the texture features under the parameter configuration is poor and the parameter search needs to be continued; when 1 is not more than JMij<1.8, the texture features obtained by the current parameters have certain discrimination, but the parameter search still needs to be continued; when 1.8 is not more than JMijAnd when the current parameter is less than or equal to 2, the texture feature configured by the current parameter is proved to have higher discrimination, and the current parameter can be determined to be optimal.
In addition, the invention also carries out interference processing based on the Sentinel-1SLC data to respectively form interference patterns of VV polarization and VH polarization so as to estimate the coherence coefficient, and the method comprises the following steps:
by estimating the spatial average of all pixel numbers L of a single SAR image, the coherence coefficient gamma of VV and VH polarization of an operation area is obtainedvv、γvh：
Wherein i is the number of picture elements in the image, VVnlAnd VHnlVV, VH polarization, VV, respectively, of the pre-lodging Sentinel-1SLC imagelAnd VHlVV and VH polarization of the post-lodging Sentinel-1SLC image, respectively.
Since the coherence coefficient provides amplitude information and phase information of the SAR signal for lodging crop monitoring, and simultaneously brings a great amount of interference phase noise, similar to random noise in backscattering coefficients, spatial continuity of lodging crop monitoring results is reduced. Therefore, the invention integrates the second-order probability texture characteristics based on the spatial adjacent characteristic statistics on the basis of comprehensively considering the backscattering change information and the coherence coefficient, thereby achieving the purposes of inhibiting noise interference and improving the accuracy and spatial continuity of the lodging monitoring result.
Based on any of the above embodiments, the method step S3 includes:
superposing the pre-lodging crop sample and the post-lodging crop sample to a Google Earth remote sensing image, and obtaining multi-sample data through visual interpretation;
extracting characteristic values of the multi-sample data based on the backscattering variation characteristic, the coherence coefficient and the average value texture characteristic to obtain a sample point characteristic data set;
and inputting the sample point characteristic data set and the corresponding sample label into a random forest-recursive characteristic elimination filter to obtain the lodging crop monitoring sensitive characteristic.
Specifically, the collected lodging crop sample points and non-lodging crop sample points are superposed into the Google Earth remote sensing image with high spatial resolution, and more similar samples are obtained through visual interpretation.
Extracting characteristic values of the plurality of samples to obtain VV of the space position of the sample pointvar、VHvar、Ratiovar、GLCM_Mean、γvv、γvhAnd (3) a characteristic data set, inputting sample labels, generally 0 and 1, and the characteristic data set into a random forest-recursive characteristic elimination filter, gradually eliminating the characteristics which are not suitable for monitoring the lodging crops, and keeping lodging crop monitoring sensitive characteristics.
Based on any of the above embodiments, the method step S4 includes:
performing centralized processing on the matrix of the lodging crop monitoring sensitive characteristics and the corresponding sample label set matrix to respectively obtain a sensitive characteristic principal component score matrix, a sensitive characteristic principal component decomposition load matrix and a sensitive characteristic principal component error matrix, and a sample label set principal component score matrix, a sample label set principal component decomposition load matrix and a sample label set principal component error matrix;
performing multiple linear regression on the sensitive feature principal component score matrix and the sample label set principal component score matrix to obtain an unknown pixel sensitive feature score matrix, and obtaining an unknown pixel label based on the unknown pixel sensitive feature score matrix;
and dividing the matrix of the lodging crop monitoring sensitive characteristics and the corresponding sample label set matrix into a training set and a testing set according to a preset proportion, inputting the training set and the testing set to a partial least square regression classification model, and performing model optimization to obtain lodging crop distribution information of the operation area.
Specifically, the invention encodes a sample label set and sensitive features into a matrix, inputs an optimal partial least square classification model, and discriminates labels of unknown pixels in a feature map pixel by pixel, thereby obtaining a monitoring result of lodging crops in an operation area, wherein the calculation process comprises the following steps:
after the sensitive characteristic matrix X and the sample label set matrix Y are subjected to centralization processing, decomposing the X and the Y to obtain principal component score matrixes T and U:
X＝TP+E
Y＝UQ+F
performing multiple linear regression on T and U:
U＝TB
B＝(XTX)-1XTY
calculating a scoring matrix T ' of the sensitive feature matrix X ' of the unknown pixel from the matrix P, thereby predicting the label Y ' of the unknown pixel:
Y′＝T′BQ
where T and U are principal component score matrices of X, Y, respectively, P and Q are principal component decomposed load matrices, respectively, and E and F are error matrices.
It should be noted that, the invention divides the sample label and the corresponding sensitive characteristic data set into a training set and a testing set according to the proportion of 3:1, inputs the partial least square regression classification model, verifies the precision of the classification model in the training set and the testing set, and repeatedly adjusts the classification model parameters until the precision in the training set and the testing set reaches the optimum.
According to the method, the coherence coefficient of the lodging crops in the operation area is constructed through coherence analysis of the double-time-phase SAR remote sensing data, the lodging crops are monitored by combining the backscattering variation characteristic and the texture characteristic, and the phase variation information of the SAR remote sensing data is applied to lodging crop identification to provide the deformation characteristic of the crops, so that the accuracy of the monitoring result is further improved.
The technical scheme of the invention is further illustrated by taking a county in a certain city as a remote sensing monitoring operation area of lodging crops.
The county is high in the southeast and low in the northwest, the southern part is mainly low mountain and hilly, the middle part is mainly plain, the crop planting area of the county reaches 24.2 million hectares, and the corn planting is mainly used. In 9/4/2020, the crop in the area is fallen down by typhoon. The data were acquired as follows: the method comprises the steps of obtaining Sentinel-1SLC images of 23 days 8 and 16 days 9 and 16 days 2020 from an official website of the European and aviation administration, wherein the collection time of lodging crop samples is 11 days-16 days 9 and 16 days 2020.
Preprocessing the Sentinel-1SLC image to obtain a ground distance multi-view Sentinel-1GRD image based on VV before lodging and VH polarization backscattering coefficients VVnl、VHnlAnd VV and VH polarization backscattering coefficients VV after the operation region is fallen downl、VHlDividing the two polarization data to obtain Ratiol、RationlAnd (5) characterizing. In the case, the coordinate information of the collected sample is encoded into vector points with space geographic coordinates, the vector points are loaded into ArcGIS software, and the vector points are superposed with Google Earth remote sensing images with high spatial resolution, and based on visual interpretation, more similar samples are obtained.
The polarization backscattering coefficient VV of the image before lodgingnl、VHnl、RationlPolarization features (VV) respectively corresponding to the images after the collapsel、VHl、Ratiol) Making a difference, and finally obtaining the backscattering change characteristics of 3 crops before and after lodging, wherein the characteristics comprise: VVvar、VHvar、RatiovarThe process is implemented based on the GDAL library of Python.
In this case, VV is utilizedvar、VHvar、RatiovarFeature construction is based on the Mean texture (GLCM _ Mean) feature of second-order probability statistics, and the process is realized by calling GDAL and Numpy libraries through Python. Determining optimal parameters according to J-M distances between lodging crops and non-lodging crops in texture characteristics under different window size and direction parameter settings, and finally obtaining VVvar、VHvar、RatiovarCorresponding 3 GLCM _ Mean features.
Carrying out interference processing on the Sentinel-1SLC data to construct coherence change characteristics of two polarization modes of VV and VH. The process is realized through SNAP software, and finally 2 coherence characteristics of VV and VH are obtained.
Further, the collected sample is subjected to characteristic value extraction operation to obtain VV of the spatial position of the sample pointvar、VHvar、Ratiovar、GLCM_Mean、γvv、γvhAnd (5) feature data collection. And inputting the sample label and the feature data set into a random forest-recursive feature elimination filter, gradually eliminating features which are not suitable for monitoring lodging crops, and keeping sensitive features.
Based on the python machine learning library scimit-learn library, the lodging region identification is realized by using a partial least square classification method, so that the distribution information of lodging crops in the whole operation region is obtained.
And coding the sample label set and the sensitive characteristics into a matrix, inputting an optimal least square classification model, and distinguishing the label of the unknown pixel in the characteristic diagram pixel by pixel so as to obtain the monitoring result of the lodging crops in the operation area.
The crop lodging monitoring method based on the synthetic aperture radar dual-polarization data coherence comprises 3 stages of data preparation, feature extraction and lodging region identification, converts lodging crop classification into an image identification process of SAR polarization features, and applies the relative coefficient, backscattering change information and texture features to the lodging crop classification, so that the monitoring method can fully consider crop forms and scattering signal time-varying features and simultaneously reduce noise disturbance of SAR signals, and accuracy of classification results is improved.
The crop lodging monitoring system based on the synthetic aperture radar dual-polarization data coherence provided by the invention is described below, and the crop lodging monitoring system based on the synthetic aperture radar dual-polarization data coherence described below and the crop lodging monitoring method based on the synthetic aperture radar dual-polarization data coherence described above can be referred to correspondingly.
Fig. 2 is a schematic structural diagram of a crop lodging monitoring system based on synthetic aperture radar dual-polarization data coherence, as shown in fig. 2, including: a first processing module 21, a second processing module 22, a third processing module 23 and a fourth processing module 24, wherein:
the first processing module 21 is used for collecting crop samples with different lodging degrees in an operation area, acquiring dual-polarized Synthetic Aperture Radar (SAR) remote sensing data before and after lodging, and extracting backscattering coefficients before and after the lodging in the operation area based on the dual-polarized SAR remote sensing data; the second processing module 22 is configured to determine a backscatter change characteristic, a coherence coefficient and an average texture characteristic before and after lodging based on the backscatter coefficients; the third processing module 23 is configured to filter and obtain lodging crop monitoring sensitive features in the backscatter change feature, the coherence coefficient, and the average texture feature based on a random forest-recursive feature elimination algorithm; the fourth processing module 24 is configured to identify the lodging crops by using a partial least squares classification method, so as to obtain spatial distribution information of the lodging crops in the working area.
According to the invention, the coherence coefficient provided by the double-time-phase dual-polarization SAR single-vision complex data is applied to crop lodging monitoring, morphological change characteristics before and after crop lodging are fully considered, second-order probability texture characteristics based on spatial proximity statistics are fused, backscattering change characteristics and noise interference of the coherence coefficient are inhibited to a certain extent, and the precision of a crop lodging monitoring result is further improved.
Fig. 3 illustrates a physical structure diagram of an electronic device, which may include, as shown in fig. 3: a processor (processor)310, a communication Interface (communication Interface)320, a memory (memory)330 and a communication bus 340, wherein the processor 310, the communication Interface 320 and the memory 330 communicate with each other via the communication bus 340. The processor 310 may invoke logic instructions in the memory 330 to perform a method of crop lodging monitoring based on synthetic aperture radar dual-polarization data coherency, the method comprising: collecting crop samples with different lodging degrees in an operation area, acquiring dual-polarized Synthetic Aperture Radar (SAR) remote sensing data before and after lodging, and extracting backscattering coefficients before and after the lodging in the operation area based on the dual-polarized SAR remote sensing data; determining backscattering change characteristics, coherence coefficients and average texture characteristics before and after lodging based on the backscattering coefficients; screening and obtaining lodging crop monitoring sensitive characteristics in the backscattering change characteristics, the coherence coefficient and the average value texture characteristics based on a random forest-recursive characteristic elimination algorithm; and identifying the lodging crops by adopting a partial least square classification method to obtain the spatial distribution information of the lodging crops in the operation area.
In addition, the logic instructions in the memory 330 may be implemented in the form of software functional units and stored in a computer readable storage medium when the software functional units are sold or used as independent products. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
In another aspect, the present invention also provides a computer program product, the computer program product comprising a computer program, the computer program being stored on a non-transitory computer-readable storage medium, wherein when the computer program is executed by a processor, the computer is capable of executing the crop lodging monitoring method based on the coherence of the dual-polarized data of the synthetic aperture radar, which is provided by the above methods, the method comprises: collecting crop samples with different lodging degrees in an operation area, acquiring dual-polarized Synthetic Aperture Radar (SAR) remote sensing data before and after lodging, and extracting backscattering coefficients before and after the lodging in the operation area based on the dual-polarized SAR remote sensing data; determining backscattering change characteristics, coherence coefficients and average texture characteristics before and after lodging based on the backscattering coefficients; screening and obtaining lodging crop monitoring sensitive characteristics in the backscattering change characteristics, the coherence coefficient and the average value texture characteristics based on a random forest-recursive characteristic elimination algorithm; and identifying the lodging crops by adopting a partial least square classification method to obtain the spatial distribution information of the lodging crops in the operation area.
In yet another aspect, the present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements a method for crop lodging monitoring based on synthetic aperture radar dual polarization data coherence provided by the above methods, the method comprising: collecting crop samples with different lodging degrees in an operation area, acquiring dual-polarized Synthetic Aperture Radar (SAR) remote sensing data before and after lodging, and extracting backscattering coefficients before and after the lodging in the operation area based on the dual-polarized SAR remote sensing data; determining backscattering change characteristics, coherence coefficients and average texture characteristics before and after lodging based on the backscattering coefficients; screening and obtaining lodging crop monitoring sensitive characteristics in the backscattering change characteristics, the coherence coefficient and the average value texture characteristics based on a random forest-recursive characteristic elimination algorithm; and identifying the lodging crops by adopting a partial least square classification method to obtain the spatial distribution information of the lodging crops in the operation area.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A crop lodging monitoring method based on the dual-polarized data coherence of a synthetic aperture radar is characterized by comprising the following steps:
collecting crop samples with different lodging degrees in an operation area, acquiring dual-polarized Synthetic Aperture Radar (SAR) remote sensing data before and after lodging, and extracting backscattering coefficients before and after the lodging in the operation area based on the dual-polarized SAR remote sensing data;
determining backscattering change characteristics, coherence coefficients and average texture characteristics before and after lodging based on the backscattering coefficients;
screening and obtaining lodging crop monitoring sensitive characteristics in the backscattering change characteristics, the coherence coefficient and the average value texture characteristics based on a random forest-recursive characteristic elimination algorithm;
and identifying the lodging crops by adopting a partial least square classification method to obtain the spatial distribution information of the lodging crops in the operation area.
2. The crop lodging monitoring method based on the coherence of the synthetic aperture radar dual-polarized data as claimed in claim 1, wherein the collecting of crop samples with different lodging degrees in the working area, the obtaining of dual-polarized SAR remote sensing data before and after lodging, the extraction of backscattering coefficients before and after lodging in the working area based on the dual-polarized SAR remote sensing data, comprises:
determining a pre-lodging crop sample and a post-lodging crop sample, acquiring a single-view complex Sentinel-1SLC image in the dual-polarization SAR remote sensing data, and preprocessing the Sentinel-1SLC image to obtain a ground-distance multi-view Sentinel-1GRD image;
and respectively acquiring a VV polarization backscattering coefficient before the operation area is lodged, a VH polarization backscattering coefficient before the operation area is lodged, a VV polarization backscattering coefficient after the operation area is lodged and a VH polarization backscattering coefficient after the operation area is lodged from the Sentinel-1GRD image.
3. The crop lodging monitoring method based on the dual-polarized data coherence of the synthetic aperture radar, as claimed in claim 2, wherein the determining of the backscatter variation characteristics, the coherence coefficients and the mean texture characteristics before and after lodging based on the backscatter coefficients comprises:
dividing the VH polarization backscattering coefficient before the operation area is lodged by the VV polarization backscattering coefficient before the operation area is lodged to obtain a coefficient ratio before lodging;
dividing the VH polarization backscattering coefficient after the operation area is lodged by the VV polarization backscattering coefficient after the operation area is lodged to obtain a coefficient ratio after lodging;
respectively making a difference between the VH polarization backscattering coefficient before the operation area is lodged and the VH polarization backscattering coefficient after the operation area is lodged, making a difference between the VV polarization backscattering coefficient before the operation area is lodged and the VV polarization backscattering coefficient after the operation area is lodged, and making a difference between the ratio of the coefficient before the lodging and the coefficient after the lodging, so as to obtain the backscattering variation characteristic;
constructing an average texture feature based on second-order probability statistics based on the backscattering variation feature;
and carrying out interference processing on the Sentinel-1SLC image to obtain a VV polarization coherent coefficient and a VH polarization coherent coefficient, and representing morphological changes before and after crop lodging based on phase difference information in the VV polarization coherent coefficient and the VH polarization coherent coefficient.
4. The crop lodging monitoring method based on the coherence of the synthetic aperture radar dual-polarized data according to claim 3, wherein the constructing of the mean texture feature based on the backscatter variation feature and based on the second-order probability statistics comprises:
determining a two-dimensional matrix of the Sentinel-1SLC image, and obtaining a normalized SAR image value based on the maximum value and the minimum value in the two-dimensional matrix;
based on the backscattering change characteristics, counting a gray level co-occurrence matrix of the normalized SAR image value according to a preset direction and a preset displacement;
and calculating the average value texture characteristics of the gray level co-occurrence matrix based on the optimized preset neighborhood window.
5. The crop lodging monitoring method based on the dual-polarized data coherence of the synthetic aperture radar as claimed in claim 4, wherein the optimized preset neighborhood window comprises:
acquiring the average texture feature of the crop sample before lodging, the average texture feature of the crop sample after lodging, the covariance matrix of the crop sample before lodging and the covariance matrix of the crop sample after lodging;
determining the optimized preset neighborhood window by adopting J-M distance based on the average texture feature of the crop sample before lodging, the average texture feature of the crop sample after lodging, the covariance matrix of the crop sample before lodging and the covariance matrix of the crop sample after lodging;
the J-M distance comprises a preset feasible region, and the preset feasible region is sequentially divided into a first feasible region interval, a second feasible region interval and a third feasible region interval based on the separability degree of the textural features.
6. The crop lodging monitoring method based on synthetic aperture radar dual-polarization data coherence of claim 3, wherein the interference processing of the Sentinel-1SLC image to obtain a VV polarization coherence coefficient and a VH polarization coherence coefficient comprises:
respectively taking phase information provided by the Sentinel-1SLC images before and after lodging as a main phase and an auxiliary phase for calculating a coherence coefficient;
calculating a coordinate mapping relation between the main phase and the auxiliary phase homonymy points, and sampling an auxiliary phase image into a pixel grid of the main phase image according to the coordinate mapping relation so that the homonymy points of the main phase image and the auxiliary phase image correspond to the same ground resolution unit;
marking main and auxiliary phase values of the same spatial position in VV polarization as a pixel pair, carrying out complex conjugate multiplication on the pixel pair by pixel pair to obtain an interference complex value, and resolving a primary differential interference phase pixel by pixel based on a real part and an imaginary part of the interference complex value;
phase unwrapping is carried out on the primary differential interference phase of the VV polarization, the fuzzy whole-cycle phase 2k pi in the phase main value is recovered, and the VV polarization coherence coefficient is obtained;
marking the main phase value and the auxiliary phase value of the same spatial position in VH polarization as a pixel pair, carrying out complex conjugate multiplication on the pixel pair by pixel pair to obtain an interference complex value, and calculating a primary differential interference phase by pixel based on a real part and an imaginary part of the interference complex value;
and performing phase unwrapping on the primary differential interference phase of the VH polarization, and recovering the blurred whole-cycle phase 2k pi in the phase main value to obtain the coherence coefficient of the VH polarization.
7. The crop lodging monitoring method based on the coherence of the synthetic aperture radar dual-polarized data, as claimed in claim 1, wherein the screening for lodging crop monitoring sensitive features in the backscatter variation feature, the coherence coefficient and the mean texture feature based on the random forest-recursive feature elimination algorithm comprises:
superposing the pre-lodging crop sample and the post-lodging crop sample to a Google Earth remote sensing image, and obtaining multi-sample data through visual interpretation;
extracting characteristic values of the multi-sample data based on the backscattering variation characteristic, the coherence coefficient and the average value texture characteristic to obtain a sample point characteristic data set;
and inputting the sample point characteristic data set and the corresponding sample label into a random forest-recursive characteristic elimination filter to obtain the lodging crop monitoring sensitive characteristic.
8. The crop lodging monitoring method based on the coherence of the synthetic aperture radar dual-polarization data of claim 1, wherein the identifying of the lodging crops by the partial least squares classification method to obtain the spatial distribution information of the lodging crops in the working area comprises:
performing centralized processing on the matrix of the lodging crop monitoring sensitive characteristics and the corresponding sample label set matrix to respectively obtain a sensitive characteristic principal component score matrix, a sensitive characteristic principal component decomposition load matrix and a sensitive characteristic principal component error matrix, and a sample label set principal component score matrix, a sample label set principal component decomposition load matrix and a sample label set principal component error matrix;
performing multiple linear regression on the sensitive feature principal component score matrix and the sample label set principal component score matrix to obtain an unknown pixel sensitive feature score matrix, and obtaining an unknown pixel label based on the unknown pixel sensitive feature score matrix;
and dividing the matrix of the lodging crop monitoring sensitive characteristics and the corresponding sample label set matrix into a training set and a testing set according to a preset proportion, inputting the training set and the testing set to a partial least square regression classification model, and performing model optimization to obtain lodging crop spatial distribution information of the operation area.
9. An electronic device comprising a memory, a processor and a computer program stored on said memory and executable on said processor, characterized in that said processor when executing said program performs the steps of a method for crop lodging monitoring based on the coherence of synthetic aperture radar dual polarized data according to any of the claims 1 to 8.
10. A non-transitory computer readable storage medium having stored thereon a computer program for implementing the steps of the method for crop lodging monitoring based on the coherence of the synthetic aperture radar dual polarization data according to any of the claims 1 to 8 when being executed by a processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210157010.0A CN114545410B (en) | 2022-02-21 | 2022-02-21 | Crop lodging monitoring method based on synthetic aperture radar dual-polarized data coherence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210157010.0A CN114545410B (en) | 2022-02-21 | 2022-02-21 | Crop lodging monitoring method based on synthetic aperture radar dual-polarized data coherence |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114545410A true CN114545410A (en) | 2022-05-27 |
| CN114545410B CN114545410B (en) | 2024-04-19 |
Family
ID=81675469
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210157010.0A Active CN114545410B (en) | 2022-02-21 | 2022-02-21 | Crop lodging monitoring method based on synthetic aperture radar dual-polarized data coherence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114545410B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117593542A (en) * | 2023-11-27 | 2024-02-23 | 首都师范大学 | Grain production area ground subsidence difference evolution characteristic calculation method, device and medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR100885246B1 (en) * | 2008-02-13 | 2009-02-25 | (주) 지오씨엔아이 | Forest arrangement method to use database establish method and this for forest arrangement that use gis |
| WO2011102520A1 (en) * | 2010-02-22 | 2011-08-25 | 株式会社パスコ | Method of generating paddy rice crop yield forecasting model, and method of forecasting crop yield of paddy rice |
| CN103969632A (en) * | 2014-03-26 | 2014-08-06 | 北京农业信息技术研究中心 | Device and method of using radar remote sensing data for monitoring wheat lodging |
| CN108802728A (en) * | 2018-04-28 | 2018-11-13 | 中国农业大学 | The crop irrigation guidance method of dual polarization synthetic aperture radar and crop modeling assimilation |
| CN113780174A (en) * | 2021-09-10 | 2021-12-10 | 福州大学 | High vegetation platform storm and rain type landslide identification method combined with random forest algorithm |
| CN113902580A (en) * | 2021-10-18 | 2022-01-07 | 四川农业大学 | Historical farmland distribution reconstruction method based on random forest model |
-
2022
- 2022-02-21 CN CN202210157010.0A patent/CN114545410B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR100885246B1 (en) * | 2008-02-13 | 2009-02-25 | (주) 지오씨엔아이 | Forest arrangement method to use database establish method and this for forest arrangement that use gis |
| WO2011102520A1 (en) * | 2010-02-22 | 2011-08-25 | 株式会社パスコ | Method of generating paddy rice crop yield forecasting model, and method of forecasting crop yield of paddy rice |
| CN103969632A (en) * | 2014-03-26 | 2014-08-06 | 北京农业信息技术研究中心 | Device and method of using radar remote sensing data for monitoring wheat lodging |
| CN108802728A (en) * | 2018-04-28 | 2018-11-13 | 中国农业大学 | The crop irrigation guidance method of dual polarization synthetic aperture radar and crop modeling assimilation |
| CN113780174A (en) * | 2021-09-10 | 2021-12-10 | 福州大学 | High vegetation platform storm and rain type landslide identification method combined with random forest algorithm |
| CN113902580A (en) * | 2021-10-18 | 2022-01-07 | 四川农业大学 | Historical farmland distribution reconstruction method based on random forest model |
Non-Patent Citations (1)
| Title |
|---|
| 韩东;杨浩;杨贵军;邱春霞;: "基于Sentinel-1雷达影像的玉米倒伏监测模型", 农业工程学报, no. 03, 8 February 2018 (2018-02-08) * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117593542A (en) * | 2023-11-27 | 2024-02-23 | 首都师范大学 | Grain production area ground subsidence difference evolution characteristic calculation method, device and medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114545410B (en) | 2024-04-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Halme et al. | Utility of hyperspectral compared to multispectral remote sensing data in estimating forest biomass and structure variables in Finnish boreal forest | |
| Quegan et al. | Multitemporal ERS SAR analysis applied to forest mapping | |
| CN103236063B (en) | Based on the SAR image oil spilling detection method of multiple dimensioned spectral clustering and decision level fusion | |
| CN105930772A (en) | City impervious surface extraction method based on fusion of SAR image and optical remote sensing image | |
| CN115236655B (en) | Landslide identification method, system, equipment and medium based on fully-polarized SAR | |
| Shaoqing et al. | The comparative study of three methods of remote sensing image change detection | |
| Janowski et al. | Exploration of glacial landforms by object-based image analysis and spectral parameters of digital elevation model | |
| CN108230375A (en) | Visible images and SAR image registration method based on structural similarity fast robust | |
| Chu et al. | Technical framework for shallow-water bathymetry with high reliability and no missing data based on time-series sentinel-2 images | |
| Ghasemi et al. | Urban classification using preserved information of high dimensional textural features of Sentinel-1 images in Tabriz, Iran | |
| CN115984778A (en) | Multi-feature optimization based method for rapidly and dynamically monitoring Sentinel-1 data in flood | |
| Kurosu et al. | Texture statistics for classification of land use with multitemporal JERS-1 SAR single-look imagery | |
| CN114545410B (en) | Crop lodging monitoring method based on synthetic aperture radar dual-polarized data coherence | |
| Radoux et al. | Quality assessment of segmentation results devoted to object-based classification | |
| Nguyen et al. | Mapping Land use/land cover using a combination of Radar Sentinel-1A and Sentinel-2A optical images | |
| CN114494282A (en) | Complex background downslope identification method and device fusing polarization information | |
| CN112989940B (en) | Raft culture area extraction method based on high-resolution third satellite SAR image | |
| CN115271560B (en) | Quantitative evaluation system and evaluation method for offshore oil drilling spilled oil weathering | |
| CN111368716A (en) | Geological disaster catastrophe farmland extraction method based on multi-source time-space data | |
| Leonardo et al. | Comparison of TanDEM-X InSAR data and high-density ALS for the prediction of forest inventory attributes in plantation forests with steep terrain | |
| Zhang et al. | Superpixelwise likelihood ratio test statistic for PolSAR data and its application to built-up area extraction | |
| US20220222497A1 (en) | Method for identifying dry salt flat based on sentinel-1 data | |
| Wang et al. | Remote Sensing Satellite Image-Based Monitoring of Agricultural Ecosystem | |
| Zhao | Multitemporal SAR images denoising and change detection: applications to Sentinel-1 data | |
| Karimzadeh et al. | A multisensor satellite image classification for the detection of mangrove forests in Qeshm Island (Southern Iran) |
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 |