CN107341781A - Based on the SAR image correcting methods for improving the matching of phase equalization characteristic vector base map - Google Patents
Based on the SAR image correcting methods for improving the matching of phase equalization characteristic vector base map Download PDFInfo
- Publication number
- CN107341781A CN107341781A CN201710594476.6A CN201710594476A CN107341781A CN 107341781 A CN107341781 A CN 107341781A CN 201710594476 A CN201710594476 A CN 201710594476A CN 107341781 A CN107341781 A CN 107341781A
- Authority
- CN
- China
- Prior art keywords
- sar
- image
- base map
- images
- vector base
- 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
- 239000013598 vector Substances 0.000 title claims abstract description 69
- 238000000034 method Methods 0.000 title claims abstract description 60
- 238000000605 extraction Methods 0.000 claims abstract description 20
- 238000012545 processing Methods 0.000 claims abstract description 12
- 230000006872 improvement Effects 0.000 claims abstract description 11
- 230000008569 process Effects 0.000 claims abstract description 10
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 claims abstract description 5
- 238000003702 image correction Methods 0.000 claims abstract description 5
- 238000004364 calculation method Methods 0.000 claims description 10
- 238000002372 labelling Methods 0.000 claims description 6
- 238000005259 measurement Methods 0.000 claims description 6
- 230000004044 response Effects 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims 1
- 238000012937 correction Methods 0.000 description 31
- 238000010586 diagram Methods 0.000 description 12
- 230000003287 optical effect Effects 0.000 description 7
- 238000001514 detection method Methods 0.000 description 6
- 238000013461 design Methods 0.000 description 5
- 230000008901 benefit Effects 0.000 description 4
- 238000003708 edge detection Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000012360 testing method Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 235000005956 Cosmos caudatus Nutrition 0.000 description 1
- 244000207740 Lemna minor Species 0.000 description 1
- 235000006439 Lemna minor Nutrition 0.000 description 1
- 240000007594 Oryza sativa Species 0.000 description 1
- 235000007164 Oryza sativa Nutrition 0.000 description 1
- 235000001855 Portulaca oleracea Nutrition 0.000 description 1
- 238000012952 Resampling Methods 0.000 description 1
- 241001189642 Theroa Species 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 239000000654 additive Substances 0.000 description 1
- 230000000996 additive effect Effects 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 239000004744 fabric Substances 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 239000003595 mist Substances 0.000 description 1
- 230000002085 persistent effect Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- VMXUWOKSQNHOCA-UKTHLTGXSA-N ranitidine Chemical compound [O-][N+](=O)\C=C(/NC)NCCSCC1=CC=C(CN(C)C)O1 VMXUWOKSQNHOCA-UKTHLTGXSA-N 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 235000009566 rice Nutrition 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000000758 substrate Substances 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000009966 trimming Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/80—Geometric correction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/02—Affine transformations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30181—Earth observation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Radar Systems Or Details Thereof (AREA)
- Image Processing (AREA)
Abstract
Present disclose provides a kind of based on the SAR image correcting methods for improving the matching of phase equalization characteristic vector base map, including：Vector base map is converted into binaryzation grating image by given resolution；Original SAR images to be corrected are subjected to geocoding according to the given resolution；The common region between two images is determined according to the geography information of the SAR images after the vector base map and geocoding of rasterizing；Based on the SAR image line feature figures after improvement phase equalization operator extraction geocoding, for carrying out Auto-matching with the vector base map of rasterizing；Image slice carries out template matches processing after cutting into slices and extracting line feature to vector base map；Matching double points are screened using Ransac methods, Mismatching point is removed, obtains the control point information of original SAR images, image correction process is carried out using image space affine Transform Model, obtains the accurate SAR images of geometry location.Disclosure bearing calibration improves matching result accuracy.
Description
Technical field
This disclosure relates to remote sensing technology field, more particularly to it is a kind of based on the base map matching of improvement phase equalization characteristic vector
SAR image correcting methods.
Background technology
Synthetic aperture radar (SAR) is as a kind of round-the-clock, round-the-clock earth observation means, in disaster region, cloudy
There is obvious advantage when mist, SAR remote sensing satellite quantity in orbit is continuously increased in recent years, including PalSAR,
Cosmo, RadarSat, TerraSAR, high score three etc., the SAR remote sensing image data amounts of acquisition are increasing.Remote sensing image
Be accurately positioned be remotely-sensed data quantification application important prerequisite, conventional Image correction in remote sensing processing need to be based on field
Control point is carried out, and this method needs to expend substantial amounts of human and financial resources, and can not obtain the letter of region-of-interest in time
Breath, a kind of ageing poor [satellite-borne SAR geometric correction methods based on persistent movement model of Chou Xiaolan, Han Chuanzhao, Liu Jiayin
[J] radar journals, 2013,2 (1):54-59].The SAR images school based on optics control base map Auto-matching is proposed in recent years
Correction method, but the optics of high accuracy positioning control image limited coverage area, and the renewal of optics control data is more tired
Difficulty, the automatic alignment requirements for constantly updating SAR images can not be met.
At present, satellite-borne SAR remote sensing image automatic geometric correction generally by with optics control base map Auto-matching complete,
Solve the problems, such as that the artificial geometric correction method low precision of remote sensing image, efficiency are low and [permitted golden duckweed etc., the satellite-borne SAR figure based on GCP storehouses
As automatic fine correction [J] Surveying and mappings, 2009,34 (5):107-109], it is necessary to choose the image slice structure control of texture-rich
System point storehouse, store the space attribute information and image information at control point；In SAR image and the matching treatment side of optics GCP sections
Face, first in thick matching stage, applied satellite orbit parameter calculates the corner location of image to be corrected, then using affine transformation
Obtain that yardstick is unified, thick matching area of irrotationality gyration with method for resampling, in smart matching stage, using normalized crosscorrelation
Coefficient carries out being accurately positioned for same place, obtains accurate matching result, calculation formula is as follows：
Wherein, (x, y) be (u, v) neighborhood point, E (I), E (T) be reference picture I and template image T gray scale it is equal
Value.
In order to realize the matching treatment of grating image and polar plot, it is necessary to carry out linear feature extraction, commonly use in this respect
Feature extraction algorithm include Sobel operators, Canny operators etc..Canny edge detection operators [Canny J.A
computational approach to edge detection[J].IEEE Trans on Pattern Analysis
Machine Intelligence 1986,8(10):679-698.] utilize the first differential of Gaussian function, in noise suppressed and
Seek preferably to balance between rim detection, its expression formula is similar to the first derivative of Gaussian function, if G (x, y) is high for two dimension
This function, to image I (x, y), edge strength is obtained by Canny operators and edge direction is as follows：
Edge strength：
Edge direction：
Carry out non-extreme value to suppress to handle with Edge track, remove weak marginal point, obtain continuous boundary.
For SAR image multiplicative noise feature, edge feature detection is realized using ratio test method, such as ROA (Ratio
Of Average) operator [R.Touzi, A.Lopes, and P.Bousquet.A statistical and geometrical
edge detectors for SAR images[J].IEEE Transaction on Geosciences and Remote
Sensing,1988,26(3):764-773.].The operator determines the picture by calculating the drench with rain ratio in upper four direction of pixel
The Grad of vegetarian refreshments, it can preferably overcome SAR image speckle noise.Ratio gradient operator calculates two nothings relative on neighborhood
The ratio (Fig. 1) of the pixel average of overlapping region.The grey blockage of Fig. 1 centers represents current calculating pixel, ROA operator meters
The grey scale pixel value mean μ of darker regions and light areas is calculated1,μ2Make ri=max { μ1/μ2,μ2/μ1, work as riMore level off to 1,
Illustrate that the average in two regions is closer, the possibility that they belong to same type region is also bigger；Then two differentiation in different regions are bigger, treat
Test point may be on two interregional borders.Moved towards in view of the difference at edge, can be each according to 4 directions shown in Fig. 1
Detection once, according to constant false alarm rate (CFAR), and takes its maximum r=max { r1,r2,r3,r4As the point ROA gradients it is strong
Degree.
As can be seen here, correction is based primarily upon the Auto-matching reality that optics controls base map to existing satellite-borne SAR remote sensing image automatically
It is existing, it there is no the research for being matched based on vector base map and realizing that satellite-borne SAR remote sensing image corrects automatically.With SAR remotely-sensed data amounts
Increase, how using further types of spatial geographic information carry out SAR should the automatic correction of image be to need to solve the problems, such as.
SAR image linear feature extraction method be mostly based on image area gray difference realize, extraction be image edge feature, pin
The characteristics of being matched to SAR image and vector image needs to propose new linear feature extraction method.
Image is controlled to pass through the matching of optical imagery and SAR image as the control base map of SAR adjustment of image by the use of optics
Processing carries out the automatic correction of satellite-borne SAR remote sensing image, the shortcomings that following be present：
1st, the existing automatic correction process of satellite-borne SAR remote sensing image is mostly based on optics control base map and realizes control be present
The problem of data coverage is limited, data can not obtain upgrading in time.The making of optics control base map data needs to pass through people
Work is surveyed and drawn on the spot determines that ground control point is realized, it is necessary to expend substantial amounts of human and material resources, and have the data of high confidence
Limited coverage area, still there are a large amount of regions can not provide optics control data；Moreover, the optics control data used now is mostly
Historical accumulation data, in the case where actual substrate changes, control information can not upgrade in time.
2nd, with the continuous development of remote sensing technology, what the resolution ratio and positioning accuracy request of SAR image were constantly lifted, and base
In the control base map of raster data, its image resolution cell and positioning precision are fixed, therefore, the control base map based on raster data
It can not meet to generate different resolution and positioning precision SAR image requirements.
3rd, raster data needs to occupy substantial amounts of memory space, it is necessary to which exclusive hard disc apparatus stores control as control base map
Base map processed, expend and calculate the hardware resources such as equipment, be not easy to deployment of the SAR automatic correction systems in different operating place.
4th, the abundant texture information of grating image record, and vector image only records the location information of interesting target, it is existing
Some matching algorithms may not apply in the Auto-matching processing of SAR image and vector image, it is necessary to design suitable for SAR remote sensing
The algorithm of image and vector controlled base map Auto-matching.
5th, calculated because SAR image has the interference of multiplicative noise, conventional Line feature operator, including Sobel in itself
The Detection results such as son, Canny operators are bad；On the other hand, vector base map (such as road, river polar plot etc.) mark is mesh
The position of center line is marked, therefore, common SAR edge detection operators, such as ROA etc. are not applied to yet.
In a word, the existing automatic correction processing method of SAR remote sensing images is realized based on optical grating control image, and optics grid
Lattice control image limited coverage area, are only capable of storing fixed resolution image, and take substantial amounts of memory space.In order to overcome
Above-mentioned deficiency is, it is necessary to realize the SAR image automatic correction methods based on vector base map, and be directed to SAR remote sensing images and vector bottom
The characteristics of figure, designs corresponding automatic matching method, ensures the stability and accuracy of automatic trimming process.
The content of the invention
(1) technical problems to be solved
In view of above-mentioned technical problem, vector base map is carried out present disclose provides one kind based on phase equalization feature is improved
The SAR image automatic correction methods matched somebody with somebody, conventional grid data are this method solve without traditional optical control image overlay area
Star-loaded optical remote sensing image automatic correction problem, SAR image and vector base map Auto-matching are realized, improve matching result standard
True property.
(2) technical scheme
According to an aspect of this disclosure, there is provided it is a kind of based on improve phase equalization characteristic vector base map matching
SAR image correcting methods, comprise the following steps：
Vector base map is converted into binaryzation grating image by given resolution；
Original SAR images to be corrected are subjected to geocoding according to the given resolution；
Being total between two images is determined according to the geography information of the SAR images after the vector base map and geocoding of rasterizing
Same region；
Based on the SAR image line feature figures after phase equalization feature operator extraction geocoding are improved, it is used for and grid
The vector base map formatted carries out Auto-matching；
Image slice carries out template matches processing after cutting into slices and extracting line feature to vector base map；
Matching double points are screened using Ransac methods, Mismatching point is removed, obtains the control point information of original SAR images,
Image correction process is carried out using image space affine Transform Model, obtains the accurate SAR images of geometry location.
In some embodiments of the present disclosure, vector base map is converted into binaryzation grid by the given resolution described
In the step of table images, the pixel value of vector labeling position in vector base map is set to 255, the pixel of scalar potential labeling position is put
For 0, to obtain a binaryzation grid map with original SAR images to be matched with resolution ratio.
In some embodiments of the present disclosure, two images are determined according to the geography information of image to be corrected and vector base map
Between common region, calculation formula is as follows：
P=FShp∩FSAR
Wherein, FShpThe geographical range information of vector base map, F are treated in expressionSARRepresent the geographic range letter of SAR images to be corrected
Breath, P are to search obtained common region information.
In some embodiments of the present disclosure, it is described to vector controlled cut into slices and extract line feature after image slice enter
In the step of row template matches processing, using the similitude of similarity measurement comparison between the standards template section, so as to obtain section
Corresponding relation.
In some embodiments of the present disclosure, the similarity measurement criterion includes：Normalized crosscorrelation, mutual information and
Phase is related.
It is described based on improvement phase equalization operator extraction original SAR shadows to be corrected in some embodiments of the present disclosure
As line feature figure, including：
For the multiplicative noise feature of SAR images, improved using the SAR Local Energy Models for meeting multiplicative noise feature former
Local Energy Model, obtain the improvement phase equalization operator suitable for SAR feature extractions；
Noise threshold estimation is carried out for the SAR Local Energy Models feature；
SAR-PC models suitable for Multiplicative noise model are obtained according to the SAR Local Energy Models and noise threshold,
Extract original SAR images line feature figure to be corrected.
In some embodiments of the present disclosure, using Gabor filter as orthogonal filter group obtain it is described meet multiply
The SAR Local Energy Models of property noise.
In some embodiments of the present disclosure, the Gabor filter is divided into an odd symmetry component and a couple weighs；
The Gabor couples weigh to be included three subwindows, utilizes odd symmetry window including two subwindows, Gabor odd symmetrys component
Image and the convolution of orthogonal filter group in proper energy amount PC models are substituted with the ratio of even symmetry window, obtains meeting multiplicative noise
SAR Local Energy Models.
It is described to be based on estimating threshold value of making an uproar in the SAR Local Energy Models in some embodiments of the present disclosure, including：
Noise response is eliminated based on the ratio calculation of odd symmetry window and even symmetry window, the estimation of noise is changed into ratio operator
Noise threshold is estimated, automatic by variation coefficient there is inverse relation between noise judgment threshold and the variation coefficient of image
Calculate noise threshold.
In some embodiments of the present disclosure, in addition to：According to the size of actual ground object target and image resolution information
Choose the scale parameter for improving SAR phase equalization detective operators.
(3) beneficial effect
It can be seen from the above technical proposal that the disclosure carries out vector base map matching based on phase equalization feature is improved
SAR image automatic correction methods at least have the advantages that one of them：
(1) for the disclosure using vector base map as control base map, having expanded the automatic correction process of satellite-borne SAR remote sensing image can
The control base map type used, the satellite-borne SAR remote sensing image for solving no traditional optical control image overlay area correct automatically
Problem.
(2) method of disclosure is applied to the automatic matching method of satellite-borne SAR remote sensing image and vector base map, is divided by specifying
Resolution geocoding solves the yardstick and rotational differential between image to be matched；Determined with reference to atural object target size and image resolution ratio
The size of SAR phase equalization Line feature operators is improved, the object edge extracted different from common edge detection algorithm,
Disclosure extraction is target's center's line, the information recorded in corresponding vector base map, improves matching result accuracy.
(3) method of disclosure is adapted to conform with the SAR image linear feature extraction of Multiplicative noise model, be SAR image and
Vector base map Auto-matching proposes important foundation.
Brief description of the drawings
By the way that shown in accompanying drawing, above and other purpose, the feature and advantage of the disclosure will become apparent from.In whole accompanying drawings
Identical reference instruction identical part, does not deliberately draw accompanying drawing, it is preferred that emphasis is show by actual size equal proportion scaling
Go out the purport of the disclosure.
Fig. 1 is the ROA template operator schematic diagrames in the direction of prior art four.
Fig. 2 is according to embodiment of the present disclosure two-dimensional Gabor filter schematic diagram.
Fig. 3 is the satellite-borne SAR remote sensing image auto-correction method skill matched according to the embodiment of the present disclosure based on vector base map
Art flow chart.
Fig. 4 is according to disclosure satellite-borne SAR remote sensing images and the schematic diagram of vector section Auto-matching result example 1.
Fig. 5 is according to disclosure satellite-borne SAR remote sensing images and the schematic diagram of vector section Auto-matching result example 2.
Fig. 6 is according to disclosure satellite-borne SAR remote sensing images and the schematic diagram of vector section Auto-matching result example 3.
Fig. 7 is TerraSAR remote sensing images correction accuracies checkpoint distribution schematic diagram.
Fig. 8 a are that the original positioning precisions of TerraSAR check schematic diagram.
Fig. 8 b are the schematic diagram of positioning precision test case 1 after disclosure correction.
Fig. 9 a are that the original positioning precisions of TerraSAR check schematic diagram.
Fig. 9 b are the schematic diagram of positioning precision test case 2 after disclosure correction.
Figure 10 a are that the original positioning precisions of TerraSAR check schematic diagram.
Figure 10 b are the schematic diagram of positioning precision test case 3 after disclosure correction.
Embodiment
For the purpose, technical scheme and advantage of the disclosure are more clearly understood, below in conjunction with specific embodiment, and reference
Accompanying drawing, the disclosure is further described.
It should be noted that in accompanying drawing or specification description, similar or identical part all uses identical figure number.It is attached
The implementation for not illustrating or describing in figure, it is form known to a person of ordinary skill in the art in art.In addition, though this
Text can provide the demonstration of the parameter comprising particular value, it is to be understood that parameter is worth accordingly without being definitely equal to, but be able to can connect
The error margin received is similar to be worth accordingly in design constraint.The direction term mentioned in embodiment, such as " on ", " under ",
"front", "rear", "left", "right" etc., only it is the direction of refer to the attached drawing.Therefore, the direction term used is for illustrating not to use
To limit the protection domain of the disclosure.
The disclosure provides a kind of SAR image correcting methods based on improvement phase equalization characteristic vector base map matching, its
For based on the SAR image automatic correction methods for improving the progress vector base map matching of phase equalization feature.Fig. 3 is based on for the disclosure
Improve the SAR image automatic correction method flow charts that phase equalization feature carries out vector base map matching.It please coordinate reference picture 3,
The SAR image automatic correction methods that vector base map matching is carried out based on improvement phase equalization feature, including：
S1, polar plot (namely vector base map, control base map) is converted into binaryzation grating image by given resolution.Deposit
255 are set in the pixel value of vector labeling position, the pixel value of scalar potential labeling position is set to 0, can obtain a width and treat
Binaryzation grid map with original SAR images with resolution ratio.
S2, by original SAR images to be corrected according to given resolution carry out geocoding.Specific coding mode can basis
The RPC files that initial data provides use universal calibration mode, or are adopted based on parameters such as the attitude of satellite, speed, orbit informations
With RD model correcting modes.
S3, according to the geography information of the SAR images after geocoding in the vector base map and S2 of rasterizing in S1 determine two
Common region between width image, calculation formula are as follows：
P=FShp∩FSAR
Wherein, FShpThe geographical range information of vector base map, F are treated in expressionSARRepresent the geographic range letter of SAR images to be corrected
Breath, P are to search obtained common region information.
S4, based on improve phase equalization operator extraction S2 in geocoding processing after SAR image line features figure (namely
Improve SAR phase equalization operator Line features algorithms), preferably extract SAR images after geocoding processing and vector
The common region part of base map, for carrying out Auto-matching, the preferably arrow with rasterizing with the vector base map of rasterizing in S1
Measure the common region part with SAR images of base map.
S41, for SAR image multiplicative noise features, utilize the SAR Local Energy Models for meeting multiplicative noise feature to improve
Former Local Energy Model, the improvement phase equalization operator suitable for SAR feature extractions is obtained, and be applied to SAR image structure
Feature extraction, it can more stably extract the linear character of SAR image.Conventional phase uniformity operator is based on Additive noise model
Design, structural information is extracted suitable for optical imagery, its calculation formula is as follows：
Wherein, W is that frequency propagates weighted volumes, and θ represents that towards angle span is in [0, π], An(x, θ) and
It is that picture point is n in Log Gabor filters yardstick, amplitude and phase when direction is θ,It is average phase, ε is one
The signal of the constant of very little, only energy value more than noise threshold T is just counted into result.
The noise model that SAR sensors obtain image is different from the noise model of optical imagery, therefore meets, it is necessary to design
The SAR Local Energy Models of multiplicative noise feature.Here using Gabor filter as orthogonal filter group, Gabor filter
An odd symmetry component can be decomposed into and a couple weighs：
By Fig. 2 it can be found that Gabor couples weigh includes three sons including two subwindows, Gabor odd symmetrys component
Window, remember that the average of each subwindow in Gabor windows is designated asWithAs shown in Figure 2, Gabor couples weigh including
The average of two subwindows is designated as respectivelyWithThe average that Gabor odd symmetrys component includes three subwindows is designated as respectivelyWithThe characteristics of for SAR image, the ratio drawn with odd symmetry window and even symmetry window calculation substitute former phase
Image and the convolution of orthogonal filter group, are shown below in bit integrity operator PC energy models：
Then corresponding Local Energy Model and range value are obtained by following methods：
When above formula is extended into two-dimensional case, it is necessary to consider directionality, it is contemplated that accuracy of detection and computation complexity
Problem, calculated frequently with six direction.
S42, the progress noise threshold estimation of noise threshold method of estimation is designed for the SAR Local Energy Models feature；
Specifically, SAR Local Energy Models theoretically solve the problems, such as speckle noise, based on the ratio in SAR Local Energy Models
Calculate to eliminate noise response, the estimation of noise is changed into the noise judgment threshold estimation of ratio operator.It is found through experiments that, makes an uproar
There is inverse relation between sound judgment threshold and the variation coefficient of image, so as to calculate noise automatically by variation coefficient
Threshold value, it is shown below：
T=0.6+alog (1/Cv)
Cv=σ/μ
Wherein, a is a constant, for adjusting the shape of threshold curve；μ and σ represents average and side in neighborhood respectively
Difference.
S43, SAR Local Energy Models and noise threshold be updated to former phase equalization (Phase Congruency,
PC) in model, the SAR-PC models suitable for Multiplicative noise model can be obtained：
Wherein, θ represents the angle of Gabor windows, Wθ(x, y) is that frequency propagates weighted volumes, and ε is small constant, Tsar-θ(x,y)
It is noise suppressed coefficient, only energy value is just credited in result more than this threshold value.
S44, chosen according to the size and image resolution information of actual ground object target and improve the detection calculation of SAR phase equalizations
The scale parameter of son.By taking road vectors figure as an example, vector mark ground object target is urban road, and road width is ten meters or so,
Terra-SAR image SpotLight model resolutions are 2 meters, then road width is 5 pixels, is chosen in phase equalization
Between template size be 5.
S5, vector controlled is cut into slices and extracted image slice progress template matches processing (piecemeal template after line feature
With), the match point relation between being cut into slices.Wherein, the similitude cut into slices using similarity measurement comparison between the standards template, so as to obtain
The corresponding relation that must be cut into slices.Similarity measurement criterion can be chosen herein includes normalized crosscorrelation, mutual information and phase phase
The methods of pass.Conventional normalized crosscorrelation method calculation formula is as follows：
Wherein, W is template image, and E (W) is the average of gray value in template image window, and I is image to be searched.Image
The center pixel (i, j) of search window is constantly mobile, and cross-correlation coefficient is obtained in the position correspondence of maximum image to be searched
With template image similitude highest region.
S6, matching double points are screened using Ransac methods, removal Mismatching point, obtain the control point letter of original SAR images
Breath, image correction process is carried out using image space affine Transform Model, obtains geometry location SAR images exactly.
Further illustrate that the disclosure carries out vector base map matching based on phase equalization feature is improved below by example
SAR image automatic correction methods.
The L1 level satellite-borne SAR audio and video products of a width TerraSAR Beijing areas are chosen first, are disclosed with Beijing in 2013
For road vectors figure as control base map, the method then provided using the disclosure carries out automatic correction process.Grating image and arrow
Amount base map is uniformly converted to the geocoding image of 2 meters of resolution ratio, and Auto-matching section is made according to 800m × 800m sizes.
The initial alignment deviation of TerraSAR images is within 50 meters, and template matches hunting zone is arranged to 50/2=25 pixels, and Fig. 4 is extremely
The result of part TerraSAR images and vector section Auto-matching is shown in Fig. 6, wherein, it is in a figures in Fig. 4 to Fig. 6
Vector base map is cut into slices and center position, and b figures are original SAR images and Auto-matching position, and c figures are SAR image extraction wire
Feature and Auto-matching position.
It is 23 that Auto-matching, which obtains number of control points, and satellite-borne SAR remote sensing image is carried out based on obtained control information
Automatic correction process.To carrying out positioning precision inspection using the front and rear image of method of disclosure correction, altogether using 18 uniform point
The checkpoint of cloth, checkpoint distribution situation in image range are as shown in Figure 7.The nothing of TerraSAR images is obtained by checkpoint
Control geometric positioning accuracy be 21.1 meters, use the disclosure be corrected processing after image geometric positioning accuracy for 2.5 meters, by 2
Rice resolution ratio calculates, and deviations are about 1.25 pixels.
On the basis of vector base map, TerraSAR remote sensing images are carried out respectively without control correction and based on this patent method
Correction process, two kinds of different disposal methods obtain the part accuracy checking result of image as shown in Fig. 8 to Figure 10, wherein, in vain
What color dotted line marked is the road position of vector base map mark, and Fig. 8 a, 9a, 10a are that original SAR image is superimposed with polar plot
The result of display, Fig. 8 b, 9b, 10b are SAR image and polar plot Overlapping display result after being corrected using the disclosure, and contrast corrects
The positioning relation of front and rear SAR images and vector base map, the disclosure effectively improve the geometric positioning accuracy of SAR image.
One of ordinary skill in the art is it should be appreciated that above implementation is intended merely to illustrate the disclosure, and is not
As the restriction to the disclosure, as long as in the scope of the present disclosure, change, modification to above example will all fall in the disclosure
Within protection domain
The foregoing description of the disclosed embodiments, professional and technical personnel in the field are enable to realize or using the application.
A variety of modifications to these embodiments will be apparent for those skilled in the art, as defined herein
General Principle can be realized in other embodiments in the case where not departing from spirit herein or scope.Therefore, the application
The embodiments shown herein is not intended to be limited to, and is to fit to and principles disclosed herein and features of novelty phase one
The most wide scope caused.
It should be noted that in accompanying drawing or specification text, the implementation that does not illustrate or describe is affiliated technology
Form known to a person of ordinary skill in the art, is not described in detail in field.In addition, the above-mentioned definition to each element and method is simultaneously
Various concrete structures, shape or the mode mentioned in embodiment are not limited only to, those of ordinary skill in the art can be carried out more to it
Change or replace：
(1) although in above-described embodiment being the energy model that phase equalization operator is improved using neighborhood ratio method, carry out
SAR grid maps target's center linear feature extraction, but the ratio method of other forms can also be used to improve phase equalization and calculated
Son, belong within the scope of the present disclosure；
(2) as being based on road vectors figure as control base map, river, coastline isovector base map can also be used to carry out certainly
The method of dynamic correction, is belonged within the category of the disclosure.
Particular embodiments described above, the purpose, technical scheme and beneficial effect of the disclosure are carried out further in detail
Describe in detail bright, should be understood that the specific embodiment that the foregoing is only the disclosure, be not limited to the disclosure, it is all
Within the spirit and principle of the disclosure, any modification, equivalent substitution and improvements done etc., the guarantor of the disclosure should be included in
Within the scope of shield.
Claims (10)
1. it is a kind of based on the SAR image correcting methods for improving the matching of phase equalization characteristic vector base map, comprise the following steps：
Vector base map is converted into binaryzation grating image by given resolution；
Original SAR images to be corrected are subjected to geocoding according to the given resolution；
Common area between two images is determined according to the geography information of the SAR images after the vector base map and geocoding of rasterizing
Domain；
Based on the SAR image line feature figures after improvement phase equalization operator extraction geocoding, for the arrow with rasterizing
Measure base map and carry out Auto-matching；
Image slice carries out template matches processing after cutting into slices and extracting line feature to vector base map；
Matching double points are screened using Ransac methods, Mismatching point is removed, obtains the control point information of original SAR images, are used
Image space affine Transform Model carries out image correction process, obtains the accurate SAR images of geometry location.
2. SAR image correcting methods according to claim 1, wherein, vector base map is pressed into the specified resolution described
Rate was converted in the step of binaryzation grating image, the pixel value of vector labeling position in vector base map was set into 255, scalar potential
The pixel of labeling position is set to 0, to obtain a binaryzation grid map with original SAR images to be matched with resolution ratio.
3. SAR image correcting methods according to claim 1, wherein, according to image to be corrected and the geography of vector base map
Information determines the common region between two images, and calculation formula is as follows：
P=FShp∩FSAR
Wherein, FShpThe geographical range information of vector base map, F are treated in expressionSARRepresent the geographical range information of SAR images to be corrected, P
To search obtained common region information.
4. SAR image correcting methods according to claim 1, wherein, wire is cut into slices and extracted to vector controlled described
Image slice was carried out in the step of template matches processing after feature, utilized the similar of similarity measurement comparison between the standards template section
Property, so as to obtain the corresponding relation of section.
5. SAR image correcting methods according to claim 4, wherein, the similarity measurement criterion includes：Normalization is mutual
Related, mutual information and phase are related.
6. SAR image correcting methods according to claim 1, wherein, it is described based on improvement phase equalization operator extraction
Original SAR images line feature figure to be corrected, including：
For the multiplicative noise feature of SAR images, improved using the SAR Local Energy Models for meeting multiplicative noise feature former local
Energy model, obtain the improvement phase equalization operator suitable for SAR feature extractions；
Noise threshold estimation is carried out for the SAR Local Energy Models feature；
SAR-PC models suitable for Multiplicative noise model, extraction are obtained according to the SAR Local Energy Models and noise threshold
Original SAR images line feature figure to be corrected.
7. SAR image correcting methods according to claim 6, wherein,
The SAR Local Energy Models for meeting multiplicative noise are obtained as orthogonal filter group using Gabor filter.
8. SAR image correcting methods according to claim 7, wherein,
The Gabor filter is divided into an odd symmetry component and a couple weighs；The Gabor couples weigh including two sons
Window, Gabor odd symmetrys component include three subwindows, and proper energy amount is substituted using the ratio of odd symmetry window and even symmetry window
Image and the convolution of orthogonal filter group, obtain the SAR Local Energy Models for meeting multiplicative noise in PC models.
9. SAR image correcting methods according to claim 6, wherein,
It is described to be based on estimating threshold value of making an uproar in the SAR Local Energy Models, including：Based on odd symmetry window and even symmetry window
Ratio calculation eliminates noise response, and the estimation of noise is changed into the noise threshold estimation of ratio operator, noise judgment threshold with
There is inverse relation between the variation coefficient of image, noise threshold is calculated by variation coefficient automatically.
10. SAR image correcting methods according to claim 6, in addition to：According to the size and image of actual ground object target
Resolution information chooses the scale parameter for improving phase equalization operator.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710594476.6A CN107341781B (en) | 2017-07-20 | 2017-07-20 | SAR image correction method based on improved phase consistency feature vector base map matching |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710594476.6A CN107341781B (en) | 2017-07-20 | 2017-07-20 | SAR image correction method based on improved phase consistency feature vector base map matching |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107341781A true CN107341781A (en) | 2017-11-10 |
| CN107341781B CN107341781B (en) | 2020-06-09 |
Family
ID=60216178
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710594476.6A Active CN107341781B (en) | 2017-07-20 | 2017-07-20 | SAR image correction method based on improved phase consistency feature vector base map matching |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107341781B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110555804A (en) * | 2018-05-31 | 2019-12-10 | 清华大学 | Correction method and device for high-resolution remote sensing data, computer equipment and readable storage medium |
| CN111062972A (en) * | 2019-12-25 | 2020-04-24 | 泉州装备制造研究所 | Image tracking method based on image frequency domain conversion |
| CN111832635A (en) * | 2020-06-28 | 2020-10-27 | 北方工业大学 | Matching method and device for ground-based SAR image and laser point cloud topographic data |
| CN113643369A (en) * | 2021-08-16 | 2021-11-12 | 中国科学院空天信息创新研究院 | Positioning information correction method for remote sensing image |
| CN114581498A (en) * | 2022-05-05 | 2022-06-03 | 环球数科集团有限公司 | Iterative model registration system combining vector data and raster image |
| CN114638766A (en) * | 2022-04-08 | 2022-06-17 | 中国科学院空天信息创新研究院 | Method for correcting luminous remote sensing image |
| CN116109848A (en) * | 2022-12-30 | 2023-05-12 | 中国人民解放军61646部队 | SAR image feature matching method and device with unchanged radiation geometry |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070122025A1 (en) * | 2002-05-30 | 2007-05-31 | The Regents Of The University Of California | Automated macromolecular crystal detection system and method |
| CN103914847A (en) * | 2014-04-10 | 2014-07-09 | 西安电子科技大学 | SAR image registration method based on phase congruency and SIFT |
| CN106920235A (en) * | 2017-02-28 | 2017-07-04 | 中国科学院电子学研究所 | Star-loaded optical remote sensing image automatic correction method based on the matching of vector base map |
-
2017
- 2017-07-20 CN CN201710594476.6A patent/CN107341781B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070122025A1 (en) * | 2002-05-30 | 2007-05-31 | The Regents Of The University Of California | Automated macromolecular crystal detection system and method |
| CN103914847A (en) * | 2014-04-10 | 2014-07-09 | 西安电子科技大学 | SAR image registration method based on phase congruency and SIFT |
| CN106920235A (en) * | 2017-02-28 | 2017-07-04 | 中国科学院电子学研究所 | Star-loaded optical remote sensing image automatic correction method based on the matching of vector base map |
Non-Patent Citations (1)
| Title |
|---|
| YUMING XIANG ETC.: ""SAR-PC:Edge Detection in SAR Images via an Advanced Phase Congruency Model"", 《REMOTE SENSING》 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110555804B (en) * | 2018-05-31 | 2022-04-15 | 清华大学 | Correction method and device for high-resolution remote sensing data, computer equipment and readable storage medium |
| CN110555804A (en) * | 2018-05-31 | 2019-12-10 | 清华大学 | Correction method and device for high-resolution remote sensing data, computer equipment and readable storage medium |
| CN111062972B (en) * | 2019-12-25 | 2023-11-21 | 泉州装备制造研究所 | Image tracking method based on image frequency domain conversion |
| CN111062972A (en) * | 2019-12-25 | 2020-04-24 | 泉州装备制造研究所 | Image tracking method based on image frequency domain conversion |
| CN111832635A (en) * | 2020-06-28 | 2020-10-27 | 北方工业大学 | Matching method and device for ground-based SAR image and laser point cloud topographic data |
| CN111832635B (en) * | 2020-06-28 | 2023-12-05 | 北方工业大学 | Matching method and device for ground-based SAR image and laser point cloud topographic data |
| CN113643369A (en) * | 2021-08-16 | 2021-11-12 | 中国科学院空天信息创新研究院 | Positioning information correction method for remote sensing image |
| CN114638766A (en) * | 2022-04-08 | 2022-06-17 | 中国科学院空天信息创新研究院 | Method for correcting luminous remote sensing image |
| CN114638766B (en) * | 2022-04-08 | 2022-11-22 | 中国科学院空天信息创新研究院 | Method for correcting luminous remote sensing image |
| CN114581498B (en) * | 2022-05-05 | 2022-07-29 | 环球数科集团有限公司 | Iterative model registration system combining vector data and raster image |
| CN114581498A (en) * | 2022-05-05 | 2022-06-03 | 环球数科集团有限公司 | Iterative model registration system combining vector data and raster image |
| CN116109848A (en) * | 2022-12-30 | 2023-05-12 | 中国人民解放军61646部队 | SAR image feature matching method and device with unchanged radiation geometry |
| CN116109848B (en) * | 2022-12-30 | 2023-11-24 | 中国人民解放军61646部队 | SAR image feature matching method and device with unchanged radiation geometry |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107341781B (en) | 2020-06-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107341781A (en) | Based on the SAR image correcting methods for improving the matching of phase equalization characteristic vector base map | |
| CN104574347B (en) | Satellite in orbit image geometry positioning accuracy evaluation method based on multi- source Remote Sensing Data data | |
| CN104361590B (en) | High-resolution remote sensing image registration method with control points distributed in adaptive manner | |
| CN104637073B (en) | It is a kind of based on the banding underground structure detection method for shining upon shadow compensation | |
| CN104867126B (en) | Based on point to constraint and the diameter radar image method for registering for changing region of network of triangle | |
| CN104536009B (en) | Above ground structure identification that a kind of laser infrared is compound and air navigation aid | |
| CN103383773B (en) | The remote sensing satellite image of a kind of dynamic extraction Image Control Point is the most just penetrating framework and the method for correction | |
| CN103337052B (en) | Automatic geometric correcting method towards wide cut remote sensing image | |
| CN105957054B (en) | A kind of image change detection method | |
| CN112712535B (en) | Mask-RCNN landslide segmentation method based on simulation difficult sample | |
| CN109544612A (en) | Point cloud registration method based on the description of characteristic point geometric jacquard patterning unit surface | |
| CN101826157B (en) | Ground static target real-time identifying and tracking method | |
| CN106920235A (en) | Star-loaded optical remote sensing image automatic correction method based on the matching of vector base map | |
| Hong et al. | A robust technique for precise registration of radar and optical satellite images | |
| CN102073990A (en) | System framework and method for automatic geometric correction of remote sensing images | |
| CN104236498A (en) | Method for measuring end size of switch rail | |
| CN103136525A (en) | Hetero-type expanded goal high-accuracy positioning method with generalized Hough transposition | |
| CN108053416A (en) | A kind of maximum oil reserve extraction system based on single width satellite image | |
| CN109100719A (en) | Combine plotting method with the topographic map of optical image based on satellite-borne SAR image | |
| CN107765257A (en) | A kind of laser acquisition and measuring method based on the calibration of reflected intensity accessory external | |
| Yoo et al. | True orthoimage generation by mutual recovery of occlusion areas | |
| Korpela et al. | The performance of a local maxima method for detecting individual tree tops in aerial photographs | |
| CN105631849A (en) | Polygon object change detection method and device | |
| Anders et al. | Influence of spatial and temporal resolution on time series-based coastal surface change analysis using hourly terrestrial laser scans | |
| CN107886505B (en) | A kind of synthetic aperture radar airfield detection method based on line primitives polymerization |
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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20211020 Address after: 100089 No.19, Beisihuan West Road, Haidian District, Beijing Patentee after: Research Institute of aerospace information innovation, Chinese Academy of Sciences Address before: 100190 No. 19 West North Fourth Ring Road, Haidian District, Beijing Patentee before: Institute of Electronics, Chinese Academy of Sciences Effective date of registration: 20211020 Address after: 250000 floor 24, block B, No. 188, Xiaoqinghe North Road, Tianqiao District, Jinan City, Shandong Province Patentee after: Zhongke satellite (Shandong) Technology Group Co.,Ltd. Address before: 100089 No.19, Beisihuan West Road, Haidian District, Beijing Patentee before: Research Institute of aerospace information innovation, Chinese Academy of Sciences |