CN113759330B - SAR cross calibration reference target selection method - Google Patents
SAR cross calibration reference target selection method Download PDFInfo
- Publication number
- CN113759330B CN113759330B CN202110964448.5A CN202110964448A CN113759330B CN 113759330 B CN113759330 B CN 113759330B CN 202110964448 A CN202110964448 A CN 202110964448A CN 113759330 B CN113759330 B CN 113759330B
- Authority
- CN
- China
- Prior art keywords
- data
- reference target
- backscattering coefficient
- longitude
- latitude
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000010187 selection method Methods 0.000 title claims abstract description 9
- 238000012937 correction Methods 0.000 claims abstract description 16
- 238000004458 analytical method Methods 0.000 claims abstract description 13
- 239000002689 soil Substances 0.000 claims description 35
- 238000000034 method Methods 0.000 claims description 29
- 238000007781 pre-processing Methods 0.000 claims description 22
- 238000006243 chemical reaction Methods 0.000 claims description 11
- 238000004519 manufacturing process Methods 0.000 claims description 5
- 238000005286 illumination Methods 0.000 claims description 4
- 230000001678 irradiating effect Effects 0.000 claims description 3
- 238000007670 refining Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 238000013519 translation Methods 0.000 claims description 3
- 238000005520 cutting process Methods 0.000 claims description 2
- 239000003513 alkali Substances 0.000 description 28
- 230000005855 radiation Effects 0.000 description 7
- 238000009826 distribution Methods 0.000 description 6
- 238000012216 screening Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000005484 gravity Effects 0.000 description 3
- 239000002585 base Substances 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000001556 precipitation Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 239000003305 oil spill Substances 0.000 description 1
- 230000000149 penetrating effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000000007 visual 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
- 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
- G01S7/40—Means for monitoring or calibrating
Abstract
The invention discloses a SAR cross calibration reference target selection method, which aims at the problem of limited number of reference targets in the field-free calibration of a synthetic aperture radar, and obtains the reference targets which have relatively stable scattering characteristics and are distributed in various parts of the world on a large scale by analyzing the scattering characteristics of regions with high and low backscattering coefficients. Firstly, carrying out scattering characteristic analysis on a high backscattering coefficient value region, and selecting a high backscattering coefficient region with stable time sequence scattering characteristic and scattering characteristic along with an incident angle as a reference target. And then, carrying out scattering characteristic analysis on the low backscattering coefficient value region, and selecting a time sequence scattering characteristic and a low backscattering coefficient ground object region which is relatively stable along with the incident angle after model correction as a reference target. And finally, adding the latitude and longitude ranges of the high and low backscattering coefficient reference targets meeting the scattering characteristic requirement into a reference target database to obtain a large number of reference targets meeting the SAR cross calibration requirement. The main implementation object of the invention is a scaled spaceborne synthetic aperture radar image, and the main work is to acquire more reference targets which can be used for SAR cross scaling work.
Description
Technical Field
The invention relates to a Synthetic Aperture Radar (SAR) cross calibration reference target selection method, and belongs to the field of radar calibration.
Background
Synthetic Aperture Radar (SAR) is an active remote sensing system, and adopts a detection mode of actively transmitting microwaves to acquire scattering characteristic information of a target. The microwaves have strong cloud and rain penetrating capacity and can be observed all day by day, so that the SAR technology is vigorously developed from the transmission of the first SAR satellite of the world Seasat-A to the present, is widely applied in the fields of agricultural remote sensing, military analysis, mapping and the like, and gradually goes to quantitative remote sensing applications such as physical quantity estimation, geometric parameter measurement and the like by the traditional qualitative applications such as visual interpretation, morphological identification and the like.
SAR radiation calibration is an important link of remote sensing data processing and application links, can eliminate errors of a system and monitor changes of load performance, ensures accuracy, reliability and consistency of remote sensing data, and is a precondition of quantitative remote sensing application. The uncalibrated SAR data can only extract target primary, qualitative information, and can result in erroneous information being extracted. The SAR radiometric calibration method is characterized in that a radiometric calibration model is constructed through reference target 'true value' and corresponding DN value for resolving, a manual scaler with known Radar Cross Section (RCS) is generally used as a reference target, DN-RCS data pairs are constructed, and calibration parameters are obtained
The radiation calibration mainly comprises the following steps: 1) Standard point target measurement, by placing a series of standard point targets of known radar cross sectional area (RCS) over the swath, SAR radiometric calibration operations are performed with the point targets. And carrying out radiation calibration by combining the information of the direction angle, the geographic position, the radar cross-sectional area and the like of the known point targets and the corresponding response energy of the point targets in the SAR image. 2) And carrying out radiation calibration on SAR satellite data to be calibrated through the calibrated SAR satellite data by a cross calibration method. The backscattering coefficient of the accurate calibrated other side SAR satellite image is known, the backscattering coefficient value of the SAR satellite image to be calibrated is obtained through the backscattering coefficient of the other side SAR satellite image, and the calibration coefficient is calculated. The radiation calibration is carried out by the cross calibration method, so that a large number of point targets can be prevented from being paved, and the satellite revisit period is not limited.
The accuracy of the cross-scaling method, however, is largely dependent on the "true" accuracy of the reference target, and therefore the choice of the reference target in the cross-scaling method is extremely important. The cross-scaled reference target needs to meet the following requirements: the scattering characteristics are relatively stable compared with the calibrated SAR reference satellite and the SAR satellite to be calibrated, the reference target is widely distributed, and the high and low backscattering intensity values are covered.
Currently, students perform radiometric calibration by using the center of gravity of the urban area as a reference target. And the center of gravity of the median of the urban area is directly used as a reference target, and the stability of the backscattering coefficient of the median of the urban area along with the change of time is greatly different. Therefore, a learner carries out fine screening on a large number of urban section data through the two classification models, and the screening threshold value needs to be estimated according to different urban section data. This method requires the acquisition of a large amount of data for the scaled city, which is difficult to implement in practice. There are also scholars to radiometric scale through ocean wind farms. Because the backscattering coefficient of the ocean is influenced by the sea surface wind field, when the ocean reference target is selected, the sea surface wind speed of the selected reference target needs to be stable. When the actual reference target is selected, real-time sea surface wind field data are required to be acquired by a certain method, and the stability of the wind speed is judged. The actual operation of acquiring the sea surface wind field data can introduce certain errors, and the operation implementation is difficult. There is an urgent need for a method for selecting a SAR cross calibration reference target that is widely available, relatively stable in scattering characteristics, and easier in reference target data acquisition.
Disclosure of Invention
The main object of the invention is to provide a ground reference target selection method oriented to SAR cross scaling.
The invention provides a method for selecting a SAR cross-scaling reference target after fully researching the SAR cross-scaling correlation direction. The reference targets of the traditional SAR field-free scaling method are limited to Amazon rainforest areas, the selection of the reference targets is less, the scaling frequency is limited (more reference targets mean higher satellite imaging scaling frequency), and therefore other reference targets are required to be explored urgently. The accuracy of SAR cross scaling depends on the backscatter characteristics of the reference target. The specific invention is as follows: 1) And taking the ground object type with the backscattering coefficient value larger than-8 dB as a high backscattering coefficient reference target, and removing the extreme point by screening the high-frequency mean value of the backscattering coefficient. And judging the time sequence stability and the stability of the backscattering coefficient along with the change of the incident angle. 2) And selecting a reference target with a low backscattering coefficient value, selecting a ground object type with a backscattering coefficient smaller than-15 dB, such as saline-alkali soil, desert, river and the like, as the reference target, and correcting the incident angle difference through a bare land scattering model to obtain the actual backscattering coefficient value. 3) The method is characterized in that the selection range of the reference target is expanded by analyzing the backward scattering characteristics of different ground objects (respectively having high and low backward scattering coefficient values), and a reference target library with relatively stable scattering characteristics and wide distribution is constructed.
The technical scheme of the invention specifically mainly comprises the following technical contents:
and step A, selecting a reference target with a high backscattering coefficient, wherein the reference target is used for selecting the reference target with the high backscattering coefficient value.
A1, preparing a high backscattering coefficient data set.
1) Data set 1 (for timing stability discrimination): and selecting a proper area with high backscattering coefficient as a reference target with high backscattering coefficient value, such as an oil spill platform, an urban area and the like. The backscattering coefficient of the reference target is greater than-8 dB to be used as a high backscattering coefficient reference target. For each selected reference target, 12 sets of SAR data for the same targeted SAR satellite, illuminating the same region with high backscatter coefficient value, and different months of the same year are selected. These 12 sets of SAR data require the same elevation track, incidence angle, illumination of the same area at different months of the year.
2) Data set 2 (for angle of incidence stability discrimination): a set of scaled SAR satellite data is selected that irradiates the same region with high backscatter coefficient value as in data set 1, and another set of scaled SAR satellite data that irradiates the same region within 20 days of the transit time of the set of data, at a different angle of incidence, at the same elevation orbit.
A2, preprocessing data.
Data preprocessing is performed on the data set 1 and the data set 2. Since the initial SAR data provides SAR image digital values (DNs) instead of backscatter coefficient values, the backscatter coefficient values need to be obtained by a preprocessing step to facilitate subsequent backscatter coefficient stability analysis.
The pretreatment comprises the following steps: track correction, radiometric calibration, terrain correction. The data backscattering coefficient value is obtained by preprocessing and refining the orbit state information, improving the distance distortion, establishing the relation between the image digital value and the backscattering coefficient value.
A3, selecting a reasonable regional reference target longitude and latitude range with high backscattering coefficient value.
The rough range of the selected area is determined through google earth, large-area air-ground areas such as parks, courts and the like are avoided as much as possible, the area with dense building groups in the center of the urban area can be selected as the initially selected range of the urban area reference target, and specific longitude and latitude are recorded.
A4, judging time sequence stability.
1) A slice database is made.
12 scene-targeted SAR data in dataset 1 are truncated against the latitude and longitude range of the reference target with high backscatter coefficient value as initially selected by A3. Because the 12-view SAR data are acquired by the same satellite and the same orbit, the data set can be directly intercepted by referring to longitude and latitude, and the corresponding urban positions are sliced according to the proper scale size to form a slice database.
2) And analyzing the time sequence stability of the urban backscattering coefficient.
And performing backward scattering coefficient stability analysis on database data consisting of 12 scene data slices. Extremum points in the data need to be removed, for example, urban areas are selected to be used as high backscattering coefficient reference targets, and a large number of roads, trees, bridges and the like also influence the distribution situation of the backscattering coefficient values. In order to remove extreme points, frequency statistics is carried out on the backscattering coefficient values in each slice, and the frequency of the backscattering coefficient in each group of slices in the interval with the group distance of 0.4 and the statistical range of 0-4 is counted. Calculating a mean value of data in a statistical range with the occurrence frequency being more than ten percent of the total number of the statistical pixel points, and carrying out unit conversion on the mean value as shown in the following formula (1):
in the middle ofIs the average value of the backscattering coefficients and sigma of each 12 scene slices i Is the mean value (unit dB) of the backscattering coefficient after unit conversion, and the statistical mean value (sigma) of the backscattering coefficient corresponding to the 12 scene slice is counted i ). Calculating the statistical mean value (sigma) of the backscattering coefficients of 12 scene slices with the same longitude and latitude i ) Mean square error between them. If the mean square error is higher than 0.8dB, the set of longitude and latitude slices is removed from the slice database. The latitude and longitude range with the mean square error of the back scattering coefficient of the slice lower than 0.8dB is recorded.
And A5, judging that the backscattering coefficient changes along with the incident angle.
1) A slice database is made.
And (3) based on the latitude and longitude range of the high backscattering coefficient reference target finally selected in the step (A4), carrying out database 2 slice manufacturing on the backscattering coefficient value of the registered data set 2. Due to the different incidence angles, the two sets of data have a certain difference in longitude and latitude. Registration directly from longitude and latitude is difficult when making data slices. A landmark ground truth determination is introduced herein to register the two sets of data, assisting in registration of the two sets of scaled SAR satellite data with different angles of incidence. And carrying out left-right translation on one group of scaled SAR satellite data according to the registration result, and matching the two groups of data to be consistent with the pixel points.
2) The stability of the backscattering coefficient as a function of the angle of incidence was analyzed.
In order to remove extreme points, a method for calculating a high-frequency mean value is adopted. Frequency statistics were performed on the backscatter coefficient values for each slice in database 2, counting the frequency of backscatter coefficient values in each group of slices that occur in the range of 0-4 with a group spacing of 0.4. And calculating the mean value of the statistical range with the frequency being more than ten percent and performing unit conversion. And (3) eliminating the longitude and latitude range of the slice group if the difference value between the back scattering coefficient mean values (unit dB) of the two slices with different incidence angles is larger than 0.8 dB.
And A6, establishing a high backscattering coefficient database.
The backscattering coefficient of the calibrated SAR satellite in the longitude and latitude range meets the time sequence stability and does not change along with the incident angle, meets the selection condition of a cross calibration reference target, and can be used as an alternative high backscattering coefficient reference target. And adding the longitude and latitude ranges of the slices meeting the conditions A4 and A5 into a high backscattering coefficient reference target database.
And B, selecting a low-backscattering coefficient reference target, wherein the low-backscattering coefficient reference target is used for selecting the reference target with a low backscattering coefficient value.
Common low backscattering coefficient ground object scenes are: the fields such as deserts, saline-alkali lands, lakes and the like have wide scene distribution of the ground features, the backscattering coefficient is lower than-15 dB, and the method is suitable for serving as a reference target with low backscattering coefficient.
B1, preparing a low backscattering coefficient reference target data set.
When data is selected, the scaled SAR satellite data of countries and regions with a large range of low backscattering coefficient ground object scenes are selected, and whether the selected scaled SAR satellite data irradiation region contains the ground object scene region is determined by referring to a world soil information base (HWSD). And recording the longitude and latitude of the selected reference target. When the calibrated SAR satellite data are selected, 6-8 months of data are excluded, and because 6-8 months of precipitation is abundant, the soil humidity is greatly influenced by precipitation, and errors are easily caused to the backward scattering coefficient analysis.
1) Data set 1 (for timing stability discrimination): for the selected low backscatter coefficient reference target area, 9 sets of SAR data for the same targeted SAR satellite are selected for different months (1-5 months and 9-12 months) in the same year, and the target area is illuminated. These 9 sets of SAR data require the same elevation track, incidence angle, illumination of the same area at different months of the year.
2) Data set 2 (for angle of incidence stability discrimination): a group of scaled SAR satellite data in the data set 1 and another group of scaled SAR satellites which are different from the data set by 20 days in transit time, different in incidence angle, same in lifting orbit and irradiating the same area are selected.
And B2, preprocessing data.
Data preprocessing is performed on the data set 1 and the data set 2. Since the initial SAR data provides SAR image digital values (DNs) rather than backscatter coefficient values. Therefore, the backscatter coefficient value needs to be obtained through a pretreatment step, so that the subsequent backscatter coefficient stability analysis is facilitated.
The pretreatment comprises the following steps: track correction, radiometric calibration, terrain correction. And obtaining the data backscattering coefficient value by preprocessing and refining the track state information, improving the relationship between the distance distortion, the resume image digital value and the backscattering coefficient value.
B3, selecting a reasonable regional reference target longitude and latitude range with low backscattering coefficient value.
The selected data set has a wide range of irradiation of the scaled SAR satellite data, and the data set needs to be intercepted when the low backscatter coefficient region is not fully irradiated. And judging the surface information by referring to the surface soil information provided by a world soil information library (HWSD), and selecting an area with a large range where the low back scattering ground object type is located as much as possible. Checking whether the selected area is covered by vegetation through the google map, and selecting the area without vegetation as much as possible in order to avoid the influence of vegetation on the backscattering coefficient. And recording the latitude and longitude range of the finally selected reference target.
And B4, judging time sequence stability.
1) A slice database is made.
And intercepting 9-scene scaled SAR data in the data set 1 according to the longitude and latitude range of the low backscatter coefficient reference target preliminarily selected by the comparison B3. Because the 9-view SAR data are acquired by the same satellite and the same orbit, the longitude and latitude intercepting data set can be directly referred, and the corresponding reference target position slicing is carried out on the maximum reference target range according to the proper scale. And (3) slicing the 9-scene calibrated SAR data in the data set 1 corresponding to the reasonable reference target longitude and latitude range selected in the step (B3) to form a slice database.
2) And analyzing the time sequence stability of the urban backscattering coefficient.
Slicing 9-view dataThe composition database was analyzed for backscatter coefficient stability. Because the selected reference target area is the bare earth surface (without other vegetation and building influence), the backscattering coefficient of the reference target area slice database is directly averaged, errors caused by data clipping are reduced, and the average is subjected to unit transformation. Calculating the statistical mean value (sigma) of the backscattering coefficients of 9 scene slices with the same longitude and latitude i ) Mean square error between them. If the mean square error is higher than 0.8dB, the set of longitude and latitude slices is removed from the slice database. Recording longitude and latitude range with mean square error of slice backscattering coefficient lower than 0.8dB
And B5, judging that the backscattering coefficient changes along with the incident angle.
1) A slice database is made.
Due to the different incidence angles, the two sets of data have a certain difference in longitude and latitude. Registration directly from longitude and latitude is difficult when making data slices. And introducing a marked ground feature to judge to register the two groups of data, translating one group of calibrated SAR satellite data left and right according to a registration result, and matching the two groups of data to be consistent with the pixel points. And (3) based on the low backscattering coefficient reference target latitude and longitude range with stable time sequence finally selected in the step (B4), carrying out database 2 slice manufacturing on the backscattering coefficient value of the registered data set 2.
2) The stability of the backscattering coefficient as a function of the angle of incidence was analyzed.
To reduce errors due to registration, extreme points, the mean value of the backscatter coefficients of each slice in database 2 is calculated and unit conversion is performed. And correcting the differences of the backscattering coefficients caused by different incident angles based on a bare earth scattering model, such as an Oh model:
middle sigma vv1 Is the backscattering coefficient, θ, of each data block of the corrected low incidence angle data 1 Is the angle of incidence, sigma, of each data block of the low angle of incidence data vv2 Is the backscattering system of each data block of high incidence angle dataNumber, θ 2 Is the angle of incidence of each data block of the high angle of incidence data.
Finally to sigma vv1 、σ vv2 And carrying out unit conversion, and calculating the mean square error between the mean value of the backscattering coefficients of the corrected low-incidence-angle data and the mean value of the backscattering coefficients of the corrected high-incidence-angle data. Because the selected reference target is a reference target with a low backscattering coefficient, the backscattering coefficient is greatly changed by environment, incident angle, system performance and the like, the stability is a certain difference from that of the reference target with a high backscattering coefficient, the requirement on the reference target with the low backscattering coefficient is lower than that on the reference target with the high backscattering coefficient when judging the stability, and the reference difference is larger than 1dB, and the longitude and latitude range of the slice group is eliminated.
And B6, establishing a low backscattering coefficient reference target database.
The backscattering coefficient of the calibrated SAR satellite in the longitude and latitude range meets the time sequence stability and is difficult to change along with the incident angle, meets the selection condition of a cross calibration reference target, and can be used as an alternative low backscattering coefficient reference target. And adding the range of the longitude and latitude of the slice meeting the conditions B4 and B5 into a low backscattering coefficient reference target database.
C. A reference target of high and low backscattering coefficients is determined.
And D, adding the longitude and latitude of the reference target with the high backscattering coefficient and the low backscattering coefficient selected in the step A and the step B into a reference target database, so as to facilitate the selection of the reference target later.
Drawings
FIG. 1 is a flow chart of the present invention.
Fig. 2 is a flow chart of high backscatter coefficient reference object selection.
Fig. 3 is a flow chart of low backscatter coefficient reference object selection.
FIG. 4 is a schematic diagram of urban time series scattering stability implementation.
Fig. 5 is a schematic diagram of implementation of time-series scattering stability in saline-alkali soil.
Fig. 6 is a schematic diagram of implementation of the effect of correcting the scattering characteristics of the incident angle of the saline-alkali soil.
Detailed Description
The invention mainly aims at solving the problems of too small number of existing reference targets and too high limitation in the SAR cross scaling process. Currently, there is a lack of a ubiquitous reference target with stable scattering properties. Two reference object selection methods with relatively stable scattering characteristics are provided, and a reference object database is enriched.
The invention provides a method for selecting a SAR cross-scaling reference target after fully researching the SAR cross-scaling correlation direction. The reference targets of the traditional SAR field-free scaling method are limited to Amazon rainforest areas, the selection of the reference targets is less, the scaling frequency is limited (more reference targets mean higher satellite imaging scaling frequency), and therefore other reference targets are required to be explored urgently. According to the invention, a plurality of ubiquitous high and low backscattering coefficient reference targets with stable scattering characteristics are used, and the reference target areas with stable scattering characteristics are obtained through a high-frequency mean value and a scattering model correction method, so that the calibration precision is improved, and the reference target database is expanded. Firstly, judging the time sequence scattering stability of a reference target, and selecting a region with stable time sequence scattering characteristics. And secondly, judging whether the scattering characteristics of the two reference targets are stable along with the change of the incident angle, and correcting the incident angle error of the reference target with low backscattering coefficient through a scattering model. And finally, acquiring a region with relatively stable scattering characteristics in time sequence and along with the change of the incident angle, and adding the region into a final reference target database.
The technical scheme of the invention specifically mainly comprises the following innovation points:
1) And taking the ground object type with the backscattering coefficient value larger than-8 dB as a high backscattering coefficient reference target, and removing the extreme point by screening the high-frequency mean value of the backscattering coefficient. And judging the time sequence stability and the stability of the backscattering coefficient along with the change of the incident angle. The conventional analysis method of the ground object scattering characteristics with high backscattering coefficient is based on screening the median center of gravity of urban ground objects by a neural network. In the method, a large number of data slices are needed to train a network model for acquiring the numerical value of the scattering region in the stable urban area, and the method only judges the time sequence scattering stability and does not analyze the change of the incident angle.
2) The reference targets with low backscattering coefficient values are selected, ground object types with backscattering coefficients smaller than-15 dB, such as saline-alkali soil, desert and the like, can be selected as the reference targets, the incidence angle difference is corrected through the bare land scattering model, the calibration precision is improved, and the existing field-free calibration method is insufficient in consideration of the incidence angle influence and is greatly influenced by the incidence angle change as in the existing method for carrying out radiation calibration through the desert. In actual radiation calibration, correction for angle of incidence variations is required. The method fully considers the difference of incidence angles, effectively improves the calibration precision through model correction, and more accurately acquires the actual backscattering coefficient value.
3) The method is characterized in that the selection range of the reference target is expanded by analyzing the backward scattering characteristics of different ground objects (respectively having high and low backward scattering coefficient values), and a reference target library with relatively stable scattering characteristics and wide distribution is constructed.
The basic flow of the synthetic aperture radar cross calibration reference target selection method based on the cross calibration thought is shown in figure 1, and the method specifically comprises the following steps:
and step A, selecting a reference target with a high backscattering coefficient, wherein the reference target is used for selecting the reference target with the high backscattering coefficient value.
A1, preparing a high backscattering coefficient data set.
The data is selected from Sentinel-1A data, the ground object type of the reference target with high backscattering coefficient is selected from Beijing urban area, and the actual data is selected as shown in Table 2:
TABLE 1 urban data set specific information
| Data set | Satellite | Incidence angle | Exploration city | Acquisition time |
| Data set 1 | Sentinel-1A | Identical to | Beijing | 2019/01/10 |
| Data set 1 | Sentinel-1A | Identical to | Beijing | 2019/02/03 |
| Data set 1 | Sentinel-1A | Identical to | Beijing | 2019/03/11 |
| Data set 1 | Sentinel-1A | Identical to | Beijing | 2019/04/16 |
| Data set 1 | Sentinel-1A | Identical to | Beijing | 2019/05/10 |
| Data set 1 | Sentinel-1A | Identical to | Beijing | 2019/06/15 |
| Data set 1 | Sentinel-1A | Identical to | Beijing | 2019/07/09 |
| Data set 1 | Sentinel-1A | Identical to | Beijing | 2019/08/14 |
| Data set 1 | Sentinel-1A | Identical to | Beijing | 2019/09/07 |
| Data set 1 | Sentinel-1A | Identical to | Beijing | 2019/10/13 |
| Data set 1 | Sentinel-1A | Identical to | Beijing | 2019/11/18 |
| Data set 1 | Sentinel-1A | Identical to | Beijing | 2019/12/12 |
| Data set 2 | Sentinel-1A | Different from | Beijing | 2020/10/02 |
| Data set 2 | Sentinel-1A | Different from | Beijing | 2020/10/19 |
A2, preprocessing data.
And carrying out data preprocessing on the data set 1 and the data set 2 to obtain the backscattering coefficient values. The preprocessing of the Sentinel-1 satellite data is very mature, and the preprocessing of orbit correction, radiometric calibration, terrain correction and the like can be directly performed through remote sensing software such as SNAP, ENVI and the like. In the scheme, SNAP software is directly adopted for preprocessing operation, and the backscattering coefficient value is obtained.
A3, selecting a reasonable regional reference target longitude and latitude range with high backscattering coefficient value.
The Beijing urban area is selected as a reference target with high backscattering coefficient, the total area of the Beijing urban area reaches 16310.54 square kilometers, the Beijing urban area is a world first-class city, and the mature city construction provides a lot of choices for urban data selection. The rough range of the selected Beijing urban area is determined through google earth, large-area air-ground areas such as parks and courts are avoided, the central zone of the urban area is selected, a plurality of areas are selected at different positions in the Beijing urban area to serve as the preliminary selection range of urban area reference targets, and specific longitude and latitude are recorded.
A4, judging time sequence stability.
1) A slice database is made.
And (5) clipping the backscattering coefficient data of the 12-scene scaled SAR data in the data set 1 according to the longitude and latitude of the selected Beijing urban reference target. The selected scale is 100×100, and data slices are made to form a slice database.
2) And analyzing the time sequence stability of the backscattering coefficient.
And (5) performing backward scattering coefficient stability analysis on a database formed by 12 scene data slices. Because of the distribution of the backscattering coefficient values affected by a large number of roads, trees, bridges, etc. in urban areas. And (5) eliminating extreme points through frequency statistics, and calculating a mean value of the backscattering coefficient. Mean square error was calculated for the mean value of the backscatter coefficients of the 12 sets of slices, and the effect of the time series stability of the data of the sets of slices is shown in fig. 4. The final mean square error is 0.26dB, the selection condition is met, the time sequence stability is good, and the urban area in the longitude and latitude range meets the time sequence requirement and enters the next step of discrimination.
And A5, judging that the backscattering coefficient changes along with the incident angle.
1) A slice database is made.
And C, based on the urban longitude and latitude range with stable time sequence finally selected in the step A4, carrying out database 2 slice manufacturing on the backscattering coefficient value of the data set 2.
Due to the different incidence angles, the two sets of data have a certain difference in longitude and latitude. And registering the two groups of data through the landmark ground feature judgment, translating one group of calibrated SAR satellite data left and right according to the registration result, and matching the two groups of data to be consistent with the pixel points. And then slicing the corresponding longitude and latitude range data.
2) The stability of the backscattering coefficient as a function of the angle of incidence was analyzed.
And (3) carrying out frequency statistics on the backscattering coefficient value of each slice in the database 2, removing extreme points, calculating the backscattering coefficient mean value and carrying out unit conversion. And calculating errors of the mean value of the backscattering coefficients of the two slices, wherein the final errors are 0.4dB, and the requirements of the invention are met.
A6, establishing a reference target database
The backscattering coefficient of the urban scaled SAR satellite in the longitude and latitude range meets the time sequence stability and is difficult to change along with the incident angle, meets the selection condition of the cross scaling reference target, and can be used as an alternative urban reference target. And adding the range of the longitude and latitude of the slice meeting the conditions A4 and A5 into an urban reference target database.
And B, selecting a low-backscattering coefficient reference target, wherein the low-backscattering coefficient reference target is used for selecting the reference target with a low backscattering coefficient value.
B1, preparing a low backscattering coefficient data set.
And selecting the saline-alkali soil in the Dunhuang area as a low backscattering coefficient reference target selection area, wherein Dunhuang city is located at the most west section of a Hexi corridor in Gansu province, and the Hexi corridor is a large-area distribution area of the famous saline-alkali soil in China. The final selected saline-alkali soil range is determined with reference to the world soil information library (HWSD). The selected data set is shown in the following table.
Table 2 saline-alkali soil dataset specific information
| Data set | Satellite | Incidence angle | Exploration area | Acquisition time |
| Data set 1 | Sentinel-1A | Identical to | Dunhuang teaSaline-alkali soil | 2019/01/17 |
| Data set 1 | Sentinel-1A | Identical to | Dunhuang saline-alkali soil | 2019/02/22 |
| Data set 1 | Sentinel-1A | Identical to | Dunhuang saline-alkali soil | 2019/03/18 |
| Data set 1 | Sentinel-1A | Identical to | Dunhuang saline-alkali soil | 2019/04/11 |
| Data set 1 | Sentinel-1A | Identical to | Dunhuang saline-alkali soil | 2019/05/29 |
| Data set 1 | Sentinel-1A | Identical to | Dunhuang saline-alkali soil | 2019/09/14 |
| Data set 1 | Sentinel-1A | Identical to | Dunhuang saline-alkali soil | 2019/10/20 |
| Data set 1 | Sentinel-1A | Identical to | Dunhuang saline-alkali soil | 2019/11/01 |
| Data set 1 | Sentinel-1A | Identical to | Dunhuang saline-alkali soil | 2019/12/19 |
| Data set 2 | Sentinel-1A | Different from | Dunhuang saline-alkali soil | 2020/12/24 |
| Data set 2 | Sentinel-1A | Different from | Dunhuang saline-alkali soil | 2020/12/19 |
And B2, preprocessing data.
Data preprocessing is performed on the data set 1 and the data set 2. And carrying out data preprocessing by SNAP software, improving distortion, constructing the relationship between the image digital value and the backscattering coefficient value, and obtaining the final backscattering coefficient value.
B3, selecting a reasonable regional reference target longitude and latitude range with low backscattering coefficient value.
The selected data set has a wide irradiation range of the scaled SAR satellite data, and the data set is required to be intercepted when the saline-alkali soil area is not completely irradiated. The latitude and longitude are finally selected in the east longitude by referring to the surface soil information provided by the world soil information base (HWSD): 94 ° 38', north latitude: the region around 40 ° 48' serves as the reference target range.
And B4, judging time sequence stability.
1) A slice database is made.
And (3) intercepting 9-scene calibrated SAR data in the data set 1 according to the longitude and latitude range of the reference target of the saline-alkali soil preliminarily selected by the B3. And intercepting the back scattering coefficient data with the same longitude and latitude range and the size of 100 multiplied by 100 to form a slice database.
2) And analyzing the time sequence stability of the backscattering coefficient.
And (5) performing backward scattering coefficient stability analysis on a database formed by 9 scene data slices. And (3) taking an average value of the backscattering coefficient of the slicing database in the saline-alkali soil region, reducing errors caused by data cutting, and carrying out unit transformation on the average value. Calculating the statistical mean value (sigma) of the backscattering coefficients of 9 scene slices with the same longitude and latitude i ) Mean square error between them. The actual timing stability effect is shown in fig. 5. The mean square error is 0.62dB, the stability requirement of the reference target time sequence is met, and the longitude and latitude range of which the mean square error of the slice back scattering coefficient is lower than 0.8dB is recorded.
And B5, judging that the backscattering coefficient changes along with the incident angle.
1) A slice database is made.
Due to the different incidence angles, the two sets of data have a certain difference in longitude and latitude. And registering through the landmark, carrying out left-right translation on one group of calibrated SAR satellite data according to a registration result, and matching the two groups of data to be consistent with the pixel points. And (3) based on the saline-alkali soil longitude and latitude range with stable time sequence finally selected in the step (B4), making a database 2 slice for the backscattering coefficient value of the registered data set 2. The final slice size was 100×100.
2) The backscattering coefficient is stable with the change of the incident angle.
In order to reduce errors caused by registration and extreme points, the mean value of the backscattering coefficient of each slice in the database 2 is calculated, and backscattering coefficient correction is performed through an oh model, and a comparison chart before and after correction is shown in fig. 6. And calculating an error between the backscattering coefficient of the corrected low-incidence SAR satellite and the average value of the backscattering coefficients of the actual low-incidence SAR satellite, wherein the final error is 0.78dB, and the selection requirement of a reference target is met.
And B6, establishing a low backscattering coefficient reference target database.
The backscattering coefficient of the calibrated SAR satellite data in the longitude and latitude range of the saline-alkali soil meets the time sequence stability and is difficult to change along with the incident angle, meets the selection condition of the cross calibration reference target, and can be used as an alternative saline-alkali soil reference target. And adding the longitude and latitude ranges of the slices meeting the conditions B4 and B5 into a saline-alkali soil reference target database.
C. A reference target of high and low backscattering coefficients is determined.
And C, adding the longitude and latitude of the urban area and the saline-alkali soil reference target selected in the step A and the step B into a reference target database, so as to facilitate the selection of the reference target later.
Claims (2)
1. A SAR cross-scaling reference target selection method, characterized by: the method comprises the following steps:
step A, selecting a reference target with a high backscattering coefficient, wherein the reference target is used for selecting the reference target with the high backscattering coefficient value;
step B, selecting a low backscatter coefficient reference target, which is used for selecting a reference target with a low backscatter coefficient value;
C. determining a reference target with high and low backscattering coefficients;
c, adding the longitude and latitude of the reference target with the high backscattering coefficient and the low backscattering coefficient selected in the step A and the step B into a reference target database, so that the subsequent reference target selection is facilitated;
the implementation process of the step A is as follows, A1, preparing a high backscattering coefficient data set;
1) Data set 1 is used for timing stability discrimination: selecting a high backscatter coefficient region as a reference target having a high backscatter coefficient value; the backscattering coefficient of the reference target is larger than-8 dB and can be used as a high backscattering coefficient reference target; selecting 12 sets of SAR data of the same calibrated SAR satellite, irradiating the same region with high backscattering coefficient value and different months in the same year for each selected reference target; these 12 sets of SAR data require the same elevation track, incidence angle, illumination of the same area at different months of the year;
2) Data set 2 was used for angle of incidence stability discrimination: selecting one set of scaled SAR satellite data illuminating the same region with high backscatter coefficient value as in data set 1, and another set of scaled SAR satellite data illuminating the same region within 20 days of the time of transit of the set of data, at different angles of incidence, with the same lifting orbit;
a2, preprocessing data;
carrying out data preprocessing on the data set 1 and the data set 2, and obtaining a backscattering coefficient value through a preprocessing step;
the pretreatment comprises the following steps: track correction, radiometric calibration, terrain correction; the track state information is refined through preprocessing, the distance distortion is improved, the relation between an image digital value and a backscattering coefficient value is established, and the data backscattering coefficient value is obtained;
a3, selecting a longitude and latitude range of a regional reference target with a high backscattering coefficient value;
selecting a central zone of an urban area and an area with dense building groups as a range preliminarily selected by an urban area reference target, and recording specific longitude and latitude;
a4, judging time sequence stability;
1) Making a slice database;
intercepting 12 scene scaled SAR data in the data set 1 according to the longitude and latitude range of the reference target with high backscatter coefficient value which is preliminarily selected by the A3; because the 12-view SAR data are obtained by the same satellite and the same orbit, the data set is directly intercepted by referring to longitude and latitude, and the corresponding urban positions are sliced according to the size of a scale to form a slice database;
2) Analyzing time sequence stability of a backward scattering coefficient of an urban area;
performing backward scattering coefficient stability analysis on database data formed by 12 scene data slices, and eliminating extreme points in the data; in order to remove extreme points, counting the frequency of the backscattering coefficient value in each slice, and counting the frequency of the backscattering coefficient in each group of slices in the interval with the group distance of 0.4 and the counting range of 0-4; calculating a mean value of data in a statistical range with the occurrence frequency being more than ten percent of the total number of the statistical pixel points, and carrying out unit conversion on the mean value as shown in the following formula (1):
in the middle ofIs the average value of the backscattering coefficients and sigma of each 12 scene slices i Is the mean value of the backscattering coefficient after unit conversion, and counts the statistical mean value sigma of the backscattering coefficient corresponding to 12 scene slices i The method comprises the steps of carrying out a first treatment on the surface of the Calculating the statistical mean sigma of the backscattering coefficients of 12 scene slices with the same longitude and latitude i Mean square error between; if the mean square error is higher than 0.8dB, the group of longitude and latitude slices are removed from the slice database; recording longitude and latitude ranges of which the mean square error of the back scattering coefficient of the slice is lower than 0.8 dB;
a5, judging that the backscattering coefficient changes along with the incident angle;
1) Making a slice database;
c, based on the latitude and longitude range of the high backscattering coefficient reference target finally selected in the step A4, carrying out database 2 slice manufacturing on the backscattering coefficient value of the registered data set 2; due to different incidence angles, the two groups of data have certain differences in longitude and latitude; registration by longitude and latitude is difficult to directly perform when making a data slice; introducing a marked ground object to judge to register the two groups of data, and assisting the registration of the two groups of calibrated SAR satellite data with different incidence angles; performing left-right translation on one group of calibrated SAR satellite data according to the registration result, and matching the two groups of data to be consistent with the pixel points;
2) Analyzing the stability of the backscattering coefficient along with the change of the incident angle;
in order to remove extreme points, a method for calculating a high-frequency mean value is adopted; carrying out frequency statistics on the backscattering coefficient value of each slice in the database 2, and counting the frequency of the backscattering coefficient value in each group of slices within the range of 0-4 within the range of 0.4; calculating the mean value of the statistical range with the frequency more than ten percent and carrying out unit conversion; for the difference value between the back scattering coefficient mean values of two groups of slices with different incidence angles, if the difference value is greater than 0.8dB, rejecting the longitude and latitude range of the group of slices;
a6, establishing a high backscattering coefficient database;
the backscattering coefficient of the calibrated SAR satellite in the longitude and latitude range meets the time sequence stability and does not change along with the incident angle, meets the selection condition of a cross calibration reference target, and is used as an alternative high backscattering coefficient reference target; and adding the longitude and latitude ranges of the slices meeting the conditions A4 and A5 into a high backscattering coefficient reference target database.
2. A SAR cross-scaling reference target selection method according to claim 1, wherein: the implementation method of the step B is as follows, B1, preparing a low backscattering coefficient reference target data set;
when data is selected, selecting scaled SAR satellite data of countries and regions with a large range of low backscattering coefficient ground object scenes, and referring to a world soil information base to determine whether the selected scaled SAR satellite data irradiation region contains such ground object scene regions;
1) Data set 1 is used for timing stability discrimination: selecting 9 groups of SAR data of the same calibrated SAR satellite which are different from each other in the same year and irradiate the target area for the selected low backscattering coefficient reference target area; these 9 sets of SAR data require the same elevation track, incidence angle, illumination of the same area at different months of the year;
2) Data set 2 was used for angle of incidence stability discrimination: selecting one group of calibrated SAR satellite data in the data set 1 and another group of calibrated SAR satellites which are different from the data set in transit time by 20 days, different in incidence angle, same in lifting orbit and irradiating the same area;
b2, data preprocessing;
carrying out data preprocessing on the data set 1 and the data set 2; the back scattering coefficient value is obtained through a pretreatment step, so that the subsequent stability analysis of the back scattering coefficient is facilitated;
the pretreatment comprises the following steps: track correction, radiometric calibration, terrain correction; acquiring a data backscattering coefficient value by preprocessing and refining track state information, improving the relation between distance distortion, resume image digital values and backscattering coefficient values;
b3, selecting a reasonable regional reference target longitude and latitude range with low backscattering coefficient value; selecting a vegetation-free area, and recording the latitude and longitude range of a finally selected reference target;
b4, judging time sequence stability;
1) Making a slice database;
intercepting 9-scene calibrated SAR data in the data set 1 according to the longitude and latitude range of the low backscatter coefficient reference target preliminarily selected by the B3; slicing 9-scene calibrated SAR data in the data set 1 corresponding to the reasonable reference target longitude and latitude range selected in the step B3 to form a slice database;
2) Analyzing time sequence stability of a backward scattering coefficient of an urban area;
performing backward scattering coefficient stability analysis on a database formed by 9 scene data slices; because the selected reference target area is the naked earth surface, the back scattering coefficient of the reference target area slice database is directly averaged, the error caused by data cutting is reduced, and the average is subjected to unit transformation; calculating the statistical mean sigma of the backscattering coefficients of 9 scene slices with the same longitude and latitude i Mean square error between; if the mean square error is higher than 0.8dB, the group of longitude and latitude slices are removed from the slice database; recording longitude and latitude ranges of which the mean square error of the back scattering coefficient of the slice is lower than 0.8 dB;
b5, judging that the backscattering coefficient changes along with the incident angle;
1) Making a slice database;
due to different incidence angles, the two groups of data have certain differences in longitude and latitude; registration by longitude and latitude is difficult to directly perform when making a data slice; introducing a marked ground feature to judge to register the two groups of data, translating one group of calibrated SAR satellite data left and right according to a registration result, and matching the two groups of data to be consistent with pixel points; b4, based on the low backscattering coefficient reference target latitude and longitude range with stable time sequence finally selected in the step B, carrying out database 2 slice manufacturing on the backscattering coefficient value of the registered data set 2;
2) Analyzing the stability of the backscattering coefficient along with the change of the incident angle;
in order to reduce errors caused by registration and extreme points, calculating the mean value of the backscattering coefficient of each slice in the database 2 and performing unit conversion; and correcting the differences of the backscattering coefficients caused by different incident angles based on a bare earth scattering model, such as an Oh model:
middle sigma vv1 Is the backscattering coefficient, θ, of each data block of the corrected low incidence angle data 1 Is the angle of incidence, sigma, of each data block of the low angle of incidence data vv2 Is the backscattering coefficient, θ, of each data block of high incidence angle data 2 Is the angle of incidence of each data block of high angle of incidence data;
finally to sigma vv1 、σ vv2 Performing unit conversion, and calculating a mean square error between a mean value of the backscattering coefficients of the corrected low-incidence-angle data and a mean value of the backscattering coefficients of the corrected high-incidence-angle data; because the selected reference target is a reference target with a low backscattering coefficient, the requirement on the reference target with the low backscattering coefficient is lower than the requirement on the reference target with the high backscattering coefficient when judging stability, and the reference difference value is greater than 1dB, rejecting the longitude and latitude range of the group of slices;
b6, establishing a low backscattering coefficient reference target database;
the backscattering coefficient of the calibrated SAR satellite in the longitude and latitude range meets the time sequence stability and changes along with the incidence angle, meets the selection condition of a cross calibration reference target, and is used as an alternative low backscattering coefficient reference target; and adding the range of the longitude and latitude of the slice meeting the conditions B4 and B5 into a low backscattering coefficient reference target database.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110964448.5A CN113759330B (en) | 2021-08-22 | 2021-08-22 | SAR cross calibration reference target selection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110964448.5A CN113759330B (en) | 2021-08-22 | 2021-08-22 | SAR cross calibration reference target selection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113759330A CN113759330A (en) | 2021-12-07 |
| CN113759330B true CN113759330B (en) | 2023-12-15 |
Family
ID=78790725
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110964448.5A Active CN113759330B (en) | 2021-08-22 | 2021-08-22 | SAR cross calibration reference target selection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113759330B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108562882A (en) * | 2018-06-21 | 2018-09-21 | 武汉大学 | A kind of satellite-borne SAR image geometry intersects calibrating method and system |
| CN112444783A (en) * | 2019-09-05 | 2021-03-05 | 中国科学院光电研究院 | Synthetic aperture radar on-orbit rapid calibration method based on natural uniform target |
| CN113204023A (en) * | 2021-05-10 | 2021-08-03 | 中国地质大学（武汉） | Dual-polarization phase optimization earth surface deformation monitoring method combining PS target and DS target |
-
2021
- 2021-08-22 CN CN202110964448.5A patent/CN113759330B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108562882A (en) * | 2018-06-21 | 2018-09-21 | 武汉大学 | A kind of satellite-borne SAR image geometry intersects calibrating method and system |
| CN112444783A (en) * | 2019-09-05 | 2021-03-05 | 中国科学院光电研究院 | Synthetic aperture radar on-orbit rapid calibration method based on natural uniform target |
| CN113204023A (en) * | 2021-05-10 | 2021-08-03 | 中国地质大学（武汉） | Dual-polarization phase optimization earth surface deformation monitoring method combining PS target and DS target |
Non-Patent Citations (5)
| Title |
|---|
| InSAR时序分析高相干目标选取方法比较研究;范锐彦;焦健;高胜;曾琪明;;地球信息科学学报(第06期);全文 * |
| SAR校准常用参考目标分析和比较;莫锦军,袁乃昌;航天返回与遥感(第02期);全文 * |
| Sentinel-1双极化数据舰船目标几何特性提取;李博颖;柳彬;郭炜炜;张增辉;郁文贤;;科技导报(第20期);全文 * |
| 时序InSAR同质样本选取算法研究;蒋弥;丁晓利;李志伟;;地球物理学报(第12期);全文 * |
| 高分辨率SAR定标参考目标辐射特性的校正方法;洪峻;雷大力;王宇;费春娇;;电子与信息学报(第02期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113759330A (en) | 2021-12-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108921885B (en) | Method for jointly inverting forest aboveground biomass by integrating three types of data sources | |
| CN111078678B (en) | Satellite precipitation data correction method based on multi-source information fusion and scale reduction | |
| CN106780091B (en) | Agricultural disaster information remote sensing extraction method based on vegetation index time-space statistical characteristics | |
| CN108732129B (en) | System and method for representing farmland soil components by images | |
| CN110136194B (en) | Snow coverage measuring and calculating method based on satellite-borne multispectral remote sensing data | |
| CN106951836B (en) | crop coverage extraction method based on prior threshold optimization convolutional neural network | |
| CN105004320A (en) | Gaofen satellite data land surface vegetation coverage retrieval method and system | |
| CN112836575B (en) | Multi-time-sequence image rice yield estimation method based on crop weather period | |
| CN102194127A (en) | Multi-frequency synthetic aperture radar (SAR) data crop sensing classification method | |
| CN111091079A (en) | TLS-based method for measuring dominant single plant structural parameters of vegetation in alpine and fragile regions | |
| Dutta et al. | Characterizing vegetation canopy structure using airborne remote sensing data | |
| CN113920438A (en) | Method for checking hidden danger of trees near power transmission line by combining ICESat-2 and Jilin image I | |
| CN110988909A (en) | TLS-based vegetation coverage determination method for sandy land vegetation in alpine and fragile areas | |
| Yamase et al. | Ground-penetrating radar estimates of tree root diameter and distribution under field conditions | |
| CN112329265A (en) | Satellite remote sensing rainfall refinement space estimation method and system | |
| CN115984778A (en) | Multi-feature optimization based method for rapidly and dynamically monitoring Sentinel-1 data in flood | |
| CN110428013B (en) | Crop remote sensing classification method and system | |
| Mokarram et al. | RELATIONSHIP BETWEEN LAND COVER AND VEGETATION INDICES. CASE STUDY: EGHLID PLAIN, FARS PROVINCE, IRAN. | |
| CN113759330B (en) | SAR cross calibration reference target selection method | |
| CN112924967A (en) | Remote sensing monitoring method for crop lodging based on radar and optical data combination characteristics and application | |
| CN117197668A (en) | Crop lodging level prediction method and system based on deep learning | |
| CN113406584B (en) | Cross calibration method for synthetic aperture radar | |
| CN108594335B (en) | Global precipitation detection method | |
| Qin et al. | Inundation impact on croplands of 2020 flood event in three Provinces of China | |
| CN115294183A (en) | Disc-shaped sub-lake water body time sequence extraction method based on multi-source remote sensing data |
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 |