CN105741331A - SAR (Synthetic Aperture Radar) image geocoding method based on triangulated irregular network lookup table - Google Patents

Abstract

The invention discloses a SAR (Synthetic Aperture Radar) image geocoding method based on a triangulated irregular network lookup table. The SAR image geocoding method comprises the following steps: converting the Grid DEM (Digital Elevation Model) data of an SAR image coverage area to generate TIN (Triangulated Irregular Network) DEM data; according to SAR image information, calculating to obtain a mapping relationship between each node of TIN DEM and a SAR image radar coordinate, and storing the mapping relationship into the TIN lookup table; and according to the mapping relationship between the SAR image information and the lookup table, carrying out resampling to obtain the SAR image subjected to geocoding. On the basis of the TIN lookup table, only the TIN node is calculated to obtain the mapping relationship, other grid points utilize interpolation to obtain the mapping relationship, storage and calculated amounts are reduced, the deficiencies of data redundancy, low efficiency and the like of a traditional geocoding method which utilizes a Grid lookup table can be overcome, and the SAR image geocoding method is suitable for carrying out quick and efficient geocoding on a great quantity of SAR images in the same area.

Description

SAR image Geocoding based on TIN look-up table

Technical field

The present invention relates to synthetic aperture radar (SyntheticApertureRadar, SAR) image geocoding, particularly relate to a kind of SAR image Geocoding based on TIN (TriangulatedIrregularNetwork, TIN) look-up table.

Background technology

Diameter radar image is widely used in the fields such as mapping, geological mapping, environmental monitoring and Disaster Assessment.SAR image stores with radar fix system, be unfavorable for special topic geography information extraction and with the relative analysis of other geography information；It addition, SAR image is because of the feature of its side-looking imaging, it is subject to the impacts such as conformation geometry, hypsography and produces geometric distortion, for instance forward sight is shunk, folded and cover and shade etc..

The purpose of SAR image geocoding is to set up the mapping relations between radar fix system and the geographic coordinate system of image, eliminates geometric distortion, it is simple to match with other geography information, and look-up table is a kind of conventional data structure expressed and store this mapping relations.Traditional look-up table has consistent geographic coordinate system with DEM, stores coordinate mapping relations with the row, column of regular grid (Grid) order by pixel.And Grid look-up table does not account for the feature of topography variation, thus there is substantial amounts of data redundancy.Such as, relative to steep hills district, use in smooth Plain less grid points just can express mapping relations, and geography encoding precision will not be caused and significantly affect.It addition, each grid points of Grid look-up table is required for calculating mapping relations, therefore memory space and calculating time overhead are all bigger.Along with SAR technical development, SAR image quantity is continuously increased, areal would be likely to occur different a large amount of SAR image such as SAR platform, polarization, phase, observation angle, and utilize the existing Geocoding based on Grid look-up table to process these SAR image, it is required for different SAR image and calculates mapping relations by pixel every time, it is difficult to meet and process requirement fast and efficiently.

Summary of the invention

In order to overcome above-mentioned the deficiencies in the prior art, the present invention provides a kind of based on TIN (TriangulatedIrregularNetwork, TIN) Geocoding of the SAR image of look-up table, the deficiencies such as data redundancy, inefficiency that the existing Geocoding utilizing Grid look-up table exists can be overcome, it is adaptable to carry out geocoding fast and efficiently for a large amount of SAR image in areal.

Present invention provide the technical scheme that

The Geocoding of a kind of SAR image based on TIN (TriangulatedIrregularNetwork, TIN) look-up table, comprises the following steps:

1) the GridDEM data of SAR image overlay area are changed, generate TINDEM data；

2) according to SAR image information, calculate the mapping relations of each node of TINDEM and SAR image radar fix, be stored in TIN look-up table；

3) according to the mapping relations in SAR image information and look-up table, carry out resampling, obtain the SAR image after geocoding.

In resampling steps, the present invention uses TIN look-up table, only TIN node is calculated and obtains mapping relations, to the not grid points of other on node, then utilizes interpolation to obtain mapping relations.Specifically, if (x, y) not on TIN look-up table node, then (i j) is obtained this image picture elements by the node index value of this pixel place triangular facet of linear interpolation.Owing to interpolation calculation is very simple, comparing and strictly calculate mapping relations according to SAR conformation model, interpolation calculation expense can be ignored substantially.Therefore, method provided by the invention can reduce storage, reduces amount of calculation.

For the Geocoding of the above-mentioned SAR image based on TIN look-up table, further, step 1) first described conversion set an elevation tolerance limit, and described elevation tolerance limits the use of in the degree of approximation of the structure and TINDEM and GridDEM controlling TINDEM；Carrying out geocoding based on TIN look-up table, can reduce the memory space that look-up table takies, and elevation tolerance limit is more big, its memory space is more little；It addition, when processing several SAR image of identical area, step 1) described conversion only needs to carry out once；

Geocoding for the above-mentioned SAR image based on TIN look-up table, further, step 1) described generate TINDEM data structure in, all nodes are all under geographic coordinate system, each node data field is the SAR image radar fix that this node is corresponding, and described radar fix includes distance to index and orientation to index.

For the Geocoding of the above-mentioned SAR image based on TIN look-up table, further, step 2) described calculating with specific reference to SAR image information, be calculated by range Doppler model；In embodiments of the present invention, described range Doppler model is made up of ellipsoid equation or plane equation (formula 1), oblique distance equation (formula 2), Doppler equation (formula 3):

(formula 1)

$R=|{\stackrel{\→}{P}}_S-\stackrel{\→}{P}|$ (formula 2)

$f_d=-\frac{2}{\mathrm{\λ}}\frac{({\stackrel{\→}{P}}_S-\stackrel{\→}{P})({\stackrel{\→}{V}}_S-\stackrel{\→}{V})}{|{\stackrel{\→}{P}}_S-\stackrel{\→}{P}|}$ (formula 3)

In formula 1～formula 3,For ground point location vector；R_eFor earth ellipsoid major semiaxis；R_pFor earth ellipsoid semi-minor axis；H is ground point geodetic height；For SAR platform position vector；R is SAR platform and ground point oblique distance length；SAR platform velocity；For ground point velocity；λ is radar wavelength；f_dFor doppler centroid.When the arbitrary pixel of SAR image can obtain its imaging according to SAR image information, unique SAR platform position vector and oblique distance, utilize range Doppler model and TINDEM elevation information to calculate and obtain its geographical location information.SAR image information describes SAR image parameter, mainly includes SAR platform positional information, velocity information, oblique distance information, imaging time information, imaging parameters information etc..

Above-mentioned steps 2) method therefor is effectively reduced the operation time compared to traditional method.When the image processed increases, during such as SAR image geocoding a large amount of for areal, due to step 1) only need to carry out once when processing several images of identical area；Therefore, when step 2) time of reducing is more than step 1) required time time, the calculating time decreased of the inventive method, method efficiency is high.

Geocoding for the above-mentioned SAR image based on TIN look-up table, further, step 2) the TIN topological structure of described TIN look-up table and step 1) the TIN topological structure of TINDEM that generates is identical, and the distance that TIN data field is SAR image of TIN look-up table is to index and orientation to index.

Geocoding for the above-mentioned SAR image based on TIN look-up table, further, step 3) described resampling first determine output image range and output pixel size, then according to the mapping relations in TIN look-up table, the SAR image of pending geocoding is carried out resampling, obtains the SAR image after geocoding.

For the Geocoding of the above-mentioned SAR image based on TIN look-up table, further, the SAR image of described pending geocoding is original SAR image or the original SAR image result under same radar fix system.When the geographical coordinate of SAR image pixel point of pending geocoding is not on TIN node, obtained the index value of this pixel point by the index value of the node of this pixel point place triangular facet of linear interpolation, thus obtain image radar fix value.

Compared with prior art, the invention has the beneficial effects as follows:

The present invention provides a kind of new for TIN (TriangulatedIrregularNetwork, TIN) Geocoding of the SAR image of look-up table, the deficiency of the existing method based on Grid look-up table can be overcome, it is suitable for a large amount of SAR image in areal and carries out geocoding fast and efficiently, be not only suitable for spaceborne being equally applicable to on-board SAR image geocoding.Technical solution of the present invention has the advantage that

(1) in memory space, relative to Grid look-up table, the present invention can reduce the memory space that look-up table takies, and the elevation tolerance limit set is more big, and its memory space is more little；

(2) aspect at runtime, although step 1) need to necessarily run the time, but for the multiple image of areal, step 1) only need to carry out once.It addition, step 2) effectively reduce compared to the method operation time based on Grid look-up table, therefore for a large amount of SAR image geocoding in areal, the present invention can reduce its operation time.

Accompanying drawing explanation

Fig. 1 is the FB(flow block) of the SAR image Geocoding based on TIN look-up table provided by the invention.

Fig. 2 represents in the embodiment of the present invention that TIN look-up table describes the schematic diagram of mapping relations between geographic coordinate system and SAR image radar fix system；

Wherein, (a) is TIN look-up table configuration；B () is SAR image radar fix system；TIN look-up table has consistent geographic coordinate system with DEM, X represent the north under geographic coordinate system to；Y represent east under geographic coordinate system to；(i j) represents that the distance of pixel correspondence SAR image radar fix system indexes to orientation to pixel.

Detailed description of the invention

Below in conjunction with accompanying drawing, further describe the present invention by embodiment, but the scope not limited the present invention in any way.

The present invention provides a kind of new for TIN (TriangulatedIrregularNetwork, TIN) Geocoding of the SAR image of look-up table, the deficiency of the existing method based on Grid look-up table can be overcome, it is adaptable to carry out geocoding fast and efficiently for a large amount of SAR image in areal.Fig. 1 is the FB(flow block) of the SAR image Geocoding based on TIN look-up table, and in the present embodiment, the inventive method is implemented specifically to perform following steps:

1) according to elevation tolerance limit set in advance, SAR image overlay area GridDEM is converted to TINDEM.The process of its conversion is summarized as follows: first input the geographical coordinate of tetra-boundary points of GridDEM, DEM is divided into two triangular facets, then one of them triangular facet is selected, calculate and fall in this triangular facet DEM possessive case site to the depth displacement of this triangular facet, judge whether highest elevation difference is limit more than elevation tolerance, if limitting more than elevation tolerance, insert DEM grid points, by some for triangular facet subdivision little triangular facets, successively newly-generated triangular facet is judged, until grid points elevation is limit less than elevation tolerance with the depth displacement of this triangular facet in all triangular facets, stop recurrence.When processing several SAR image of areal, this step only needs to carry out once.

2) according to SAR image information and TINDEM, use range Doppler model, calculate TINDEM each node correspondence SAR image radar fix (distance is to indexing with orientation to indexing), and result of calculation is stored in TIN look-up table.Fig. 2 illustrates TIN look-up table and describes the mapping relations between geographic coordinate system and SAR image radar fix system.The TIN topological structure of the TIN topological structure of TIN look-up table and the corresponding TINDEM of generation is identical, and the distance that TIN data field is SAR image of TIN look-up table is to index and orientation to index.It is, TIN look-up table and corresponding both TINDEM have the node of equal number and position, triangular facet is also identical；Its all nodes process all under geographic coordinate system, and each node data field is SAR image radar fix that this node is corresponding (distance to orientation to indexing).

Wherein, range Doppler model is made up of ellipsoid equation or plane equation (formula 1), oblique distance equation (formula 2), Doppler equation (formula 3):

(formula 1)

$R=|{\stackrel{\→}{P}}_S-\stackrel{\→}{P}|$ (formula 2)

$f_d=-\frac{2}{\mathrm{\λ}}\frac{({\stackrel{\→}{P}}_S-\stackrel{\→}{P})({\stackrel{\→}{V}}_S-\stackrel{\→}{V})}{|{\stackrel{\→}{P}}_S-\stackrel{\→}{P}|}$ (formula 3)

In formula 1～formula 3,For ground point location vector；R_eFor earth ellipsoid major semiaxis；R_pFor earth ellipsoid semi-minor axis；H is ground point geodetic height；For SAR platform position vector；R is SAR platform and ground point oblique distance length；SAR platform velocity；For ground point velocity；λ is radar wavelength；f_dFor doppler centroid.When the arbitrary pixel of SAR image can obtain its imaging according to SAR image information, unique SAR platform position vector and oblique distance, utilize range Doppler model and TINDEM elevation information to calculate and obtain its geographical location information.

(3) resampling

According to the mapping relations in SAR image information and look-up table, carry out resampling；Wherein, SAR image information includes SAR image distance to pixel quantity and orientation to pixel quantity.When carrying out resampling, when the index of look-up table more than distance to, orientation to pixel quantity or less than 0 time, this pixel is set to null value.

The process carrying out resampling is as follows:

First, determine geographic range and Pixel domain size (can be consistent with initial GridDEM) of image after the geocoding of output, then geographical encoded images pixel (x is selected one by one, y), according to TIN look-up table, search the distance of this pixel correspondence SAR image to orientation to pixel index (i, j).If (x, y) not on TIN look-up table node, then (i j) is obtained this image picture elements by the node index value of this pixel place triangular facet of linear interpolation.Here SAR image is original SAR image or other SAR image result under same radar fix system.By SAR image (i, j) pixel gray value give image picture elements after geocoding (x, y).If (i j) does not index for integer pixel, then obtains this pixel gray value by interpolation；If (i, j) more than SAR image scope, then by image picture elements after geocoding, (x y) is set to null value.

It should be noted that the purpose publicizing and implementing example is in that help is further appreciated by the present invention, but it will be appreciated by those skilled in the art that: in the spirit and scope without departing from the present invention and claims, various substitutions and modifications are all possible.Therefore, the present invention should not be limited to embodiment disclosure of that, and the scope that the scope of protection of present invention defines with claims is as the criterion.

Claims (9)

1., based on a Geocoding for the SAR image of TIN look-up table, comprise the following steps:

1) the GridDEM data of SAR image overlay area are changed, generate TINDEM data；

2) according to described SAR image information, by calculating the mapping relations obtaining each node of TINDEM with SAR image radar fix, it is stored in TIN look-up table；

3) according to the mapping relations in described SAR image information and described TIN look-up table, carry out resampling, obtain the SAR image after geocoding.

2. as claimed in claim 1 based on the Geocoding of the SAR image of TIN look-up table, it is characterized in that, step 1) first described conversion set elevation tolerance limit, and described elevation tolerance limits the use of in the degree of approximation of the structure and TINDEM and GridDEM controlling TINDEM；When processing several SAR image of identical area, step 1) described conversion only needs to carry out once.

3. as claimed in claim 1 based on the Geocoding of the SAR image of TIN look-up table, it is characterized in that, step 1) described generate TINDEM data structure in, all nodes are all under geographic coordinate system, each node data field is the SAR image radar fix that this node is corresponding, and described radar fix includes distance to index and orientation to index.

4., as claimed in claim 1 based on the Geocoding of the SAR image of TIN look-up table, it is characterized in that, step 2) described calculating is calculated especially by range Doppler model；Described range Doppler model is made up of ellipsoid equation or plane equation (formula 1), oblique distance equation (formula 2), Doppler equation (formula 3):

$R=|{\stackrel{\→}{P}}_S-\stackrel{\→}{P}|$ (formula 2)

$f_d=-\frac{2}{\mathrm{\λ}}\frac{\left({\stackrel{\→}{P}}_S-\stackrel{\→}{P}\right)\left({\stackrel{\→}{V}}_S-\stackrel{\→}{V}\right)}{|{\stackrel{\→}{P}}_S-\stackrel{\→}{P}|}$ (formula 3)

In formula 1～formula 3,For ground point location vector；R_eFor earth ellipsoid major semiaxis；R_pFor earth ellipsoid semi-minor axis；H is ground point geodetic height；For SAR platform position vector；R is SAR platform and ground point oblique distance length；Platform speed vector；For ground point velocity；λ is radar wavelength；f_dFor doppler centroid.When the arbitrary pixel of SAR image can obtain its imaging according to SAR image information, unique SAR platform position vector and oblique distance, utilize range Doppler model and TINDEM elevation information to calculate and obtain its geographical location information.

5., as claimed in claim 1 based on the Geocoding of the SAR image of TIN look-up table, it is characterized in that, step 2) data structure of described TIN look-up table and step 1) structure of TINDEM data that generates is identical.

6. as claimed in claim 1 based on the Geocoding of the SAR image of TIN look-up table, it is characterized in that, step 3) described resampling first determine output image range and output pixel size, then according to the mapping relations in TIN look-up table, the SAR image of pending geocoding is carried out resampling, obtains the SAR image after geocoding.

7. as claimed in claim 6 based on the Geocoding of the SAR image of TIN look-up table, it is characterized in that, the SAR image of described pending geocoding is original SAR image or the original SAR image result under same radar fix system.

8. as claimed in claim 7 based on the Geocoding of the SAR image of TIN look-up table, it is characterized in that, SAR image for described pending geocoding carries out resampling, when the geographical coordinate of SAR image pixel point of pending geocoding is not on TIN node, obtained the index value of this pixel point by the index value of the node of this pixel point place triangular facet of linear interpolation, thus obtain image radar fix value.

9. as claimed in claim 1 based on the Geocoding of the SAR image of TIN look-up table, it is characterized in that, described SAR image information describes SAR image parameter；It is one or more that described SAR image parameter includes in SAR platform position, SAR platform speed, oblique distance, imaging time and imaging parameters.