CN105372663A - Resampling method facing images of sidescan sonar - Google Patents

Abstract

The present invention relates to a resampling method facing images of a sidescan sonar. The resampling method is configured to refer to geospatial coordinates and establish the calculation relation of pixels in resampling images and data of the sidescan sonar through analysis of the space geometry relation between scanning lines of the sidescan sonar and the resampling points to realize the geocoding of data of the sidescan sonar through indirect resampling and eliminate various gaps, of the data of the sidescan sonar, generated in the geocoding process, so that a way is provided for the acquisition of clear and continuous undersea images; and moreover, in a concrete realization mechanism, the resampling method is able to support reading in batches, local processing and vector calculation so as to facilitate simplifying the indirect resampling of the data of the sidescan sonar to a linear operation, therefore the treatment efficiency of the geocoding of the data of the sidescan sonar is improved.

Description

A kind of method for resampling towards side-scanning sonar image

Technical field

The present invention relates to a kind of side-scanning sonar image Geocoding, be specifically related to a kind of method for resampling towards side-scanning sonar image.

Background technology

Side-scan sonar is the equipment using echo depth sounding principle detection sea-bed topography and immersed body, it utilizes transducer to water-bed Firing Sonar, received formation sonar video by towfish after reflection and find immersed body, the display processing unit that the signal received is passed on deck by towing cable carries out subsequent treatment, it can provide the acoustic imaging of seabed form intuitively, because it has the advantages such as high precision, high resolving power and cost are lower, be widely used in the fields such as marine charting, submarine geology exploration, oceanographic engineering.Original side-scanning sonar image is by the tactic sequence image of sweep trace, does not possess the geographic orientation of globality, therefore needs to carry out geocoding to it so that further analysis and application.Resampling is the important step in the geocoding process of side-scan sonar data, and it utilizes the geometric relationship between side-scan sonar sweep trace and towfish flight path, realizes primary side sweep sonar data to geocoding target image pixel-map by a series of computing.Because side-scan sonar is when carrying out data acquisition, towfish is because being subject to the impacts such as ocean current, tide and hull turning, it is abstract that its ship trajectory is difficult to use accurate function, and then cause resampling to become an extremely complicated nonlinear computation process, directly cannot utilize the method for resampling in satellite remote sensing field.At present, the geocoding that offside sweeps sonar data adopts the mode of direct resampling usually, namely from side-scan sonar data, utilize the geometric relationship between towfish flight path and sonar sweep trace, sweep trace is directly projected in geocoding target image, and then generation sweeps image with the side of geospatial reference.After carrying out resampling based on this mode; usually can there is gap in geocoding target image, namely adjacent scanning lines is after projection is resolved, and can produce white space therebetween; affect the display effect of the actual image in seabed, this causes due to ship's speed inequality or hull going direction changing.About the elimination in gap, usually have two kinds of solutions, one is weaken gap by the resolution of reduction resampling image, but this method cannot process the larger situation in gap, easily causes the loss of encoded images readability simultaneously; Another kind eliminates gap by filtering process, but this method needs to carry out repeatedly filtering operation according to image effect to view picture image, and calculated amount is comparatively large, and needs manual intervention, and processing procedure efficiency is lower.

Summary of the invention

Technical matters to be solved by this invention provides a kind of method for resampling towards side-scanning sonar image for above-mentioned prior art deficiency, this method for resampling can generate the side-scan sonar geocoding image with high spatial resolution, and has higher treatment effeciency.

The present invention solves the problems of the technologies described above adopted technical scheme: a kind of method for resampling towards side-scanning sonar image, is characterized in that: comprise the steps:

Step one: the File header information of resolving side-scanning sonar image file, obtains at least following two property parameters in side-scanning sonar image file: oblique distance range R and passage image width ppc;

Step 2: the scan-line data in order reading side-scanning sonar image file and the towfish elevation information of correspondence, receiving time information and sequence number of scanning lines, utilize oblique distance calibration model to carry out oblique distance correction process to the scan-line data obtained, and result is stored to oblique distance correction file;

Step 3: order reads the track line file corresponding with side-scanning sonar image file, extract the track points coordinate sequence of towfish, time of reception sequence and Case Number sequence, utilize side-scan sonar towfish flight path processing method, simulate the ship trajectory of towfish reality, and calculate the towfish geographic coordinate residing when receiving every bar scan-line data;

Step 4: on the basis of step 2 and step 3, utilize side-scanning sonar image geo-coding techniques, resolve the geographic coordinate of each sweep trace end points, the each sweep trace end points geographic coordinate calculated is stored to sweep trace extreme coordinates temporary file, calculate the minimum enclosed rectangle of this side-scanning sonar image file institute covering space simultaneously, record space four to the scope that this side-scanning sonar image file covers;

Step 5: the resampling resolution of space four to the scope covered according to the side-scanning sonar image file calculated in step 4 and needs creates the resampling image of side-scanning sonar image file, and each pixel value in resampling image is initialized as zero;

Step 6: sequentially read scan-line data corresponding in the oblique distance correction file created by step 2, get adjacent two scan-line datas, in the sweep trace extreme coordinates temporary file simultaneously created in step 4, order reads the sweep trace extreme coordinates corresponding with these two sweep traces, build the minimum enclosed rectangle comprising these two sweep traces, then this minimum enclosed rectangle pixel coordinate corresponding in resampling image is calculated according to this rectangle upper left angle point and the geographic coordinate of bottom right angle point and the resampling resolution of needs, obtain the image block of current pending re-sampling operations by this,

Step 7: conjugate points is carried out to each pixel obtaining image block in step 6 and resolves, and according to the pixel value of bilinear interpolation determination conjugate points;

Step 8: return step 6, until recording processing all in sweep trace extreme coordinates temporary file is complete.

As improvement, in described step 2, the detailed process utilizing the scan-line data of oblique distance calibration model to acquisition to carry out oblique distance correction process is:

First, after oblique distance being corrected, on side-scanning sonar image, the pixel value of all pixels is initialized as zero;

Then, find out oblique distance and correct the conjugate points of some pixels on side-scanning sonar image file on rear side-scanning sonar image, note P _p(x _p, y _p) be certain pixel after oblique distance corrects on side-scanning sonar image, the sweep trace of its correspondence is L, and when receiving this sweep trace, the resolution of corresponding towfish height to be towfishAlt, sidescanRes be original side-scanning sonar image file, utilizes formula 1 to calculate P _pconjugate points Ps (xs, ys) on side-scanning sonar image file;

formula 1

Finally, the pixel P after the pixel value assignment of conjugate points Ps (xs, ys) in side-scanning sonar image file being corrected to oblique distance on side-scanning sonar image _p(x _p, y _p).

Improve, in described step 4, the end points geographic coordinate of note sweep trace L1 ~ Ln is respectively P again _s1(x _s1, y _s1), P _e1(x _e1, y _e1), P _s2(x _s2, y _s2), P _e2(x _e2, y _e2) ... P _sn(x _sn, y _sn), P _en(x _en, y _en), if x _min=min{x _s1, x _e1, x _s2, x _e2x _sn, x _en, x _max=max{x _s1, x _e1, x _s2, x _e2x _sn, x _en, y _min=min{y _s1, y _e1, y _s2, y _e2y _sn, y _en, y _max=max{y _s1, y _e1, y _s2, y _e2y _sn, y _en, then four angular coordinates of the minimum enclosed rectangle of side-scanning sonar image file institute covering space are respectively P _tl(x _min, y _max), P _tr(x _max, y _max), P _bl(x _min, y _min), P _br(x _max, y _min), in described step 5, if resResolution is the resolution of the resampling image needed, with P _tl, P _tr, P _bl, P _brbe four to border, create the resampling image of side-scanning sonar image file, and the pixel value of each pixel in resampling image is set to zero.

Improve again, in described step 6, if to get adjacent sweep trace be L1, L2, remember that the line number in its place oblique distance correction file is respectively Row1, Row2, the sweep trace end points geographic coordinate of its correspondence is P _s1(x _s1, y _s1), P _e1(x _e1, y _e1), P _s2(x _s2, y _s2) and P _e2(x _e2, y _e2), if slx _min=min{x _s1, x _e1, x _s2, x _e2, slx _max=max{x _s1, x _e1, x _s2, x _e2, sly _min=min{y _s1, y _e1, y _s2, y _e2, sly _max=max{y _s1, y _e1, y _s2, y _e2, then the geographic coordinate comprising minimum enclosed rectangle four angle points of L1, L2 sweep trace is respectively slP _tl(slx _min, sly _max), slP _tr(slx _max, sly _max), slP _bl(slx _min, sly _min), slP _br(slx _max, sly _min), the pixel coordinate P of height H eight, width W idth that this boundary rectangle is corresponding in resampling image and upper left angle point thereof _rs(x _start, y _start) calculate acquisition by formula 2;

$\left\{\begin{array}{c}{x}_{start}=ceil(\left({\mathrm{slx}}_{min}-x_{min}\right)/res\mathrm{Re}solution)\\ y_{start}=ceil(\left(y_{max}-{\mathrm{sly}}_{max}\right)/res\mathrm{Re}solution)\\ Width=round(\left({\mathrm{slx}}_{\max }-{\mathrm{slx}}_{\min }+0.5\right)/res\mathrm{Re}solution)\\ Height=round(\left({\mathrm{sly}}_{max}-{\mathrm{sly}}_{min}+0.5\right)/res\mathrm{Re}solution)\end{array}\right.$ Formula 2.

Improve again, in described step 7, with P _s1(x _s1, y _s1), P _e1(x _e1, y _e1), P _s2(x _s2, y _s2) and P _e2(x _e2, y _e2) be angle point, build convex quadrangle P _s1p _e1p _e2p _s2, the concrete steps that conjugate points resolves are carried out one by one to the pixel in image block as follows: when any pixel in image block, remember that its pixel coordinate in image block is P _d( _xd, _yd), first utilize formula 3 by P _dpoint pixel coordinate transfers corresponding geographic coordinate P to _dg(x _dg, y _dg), utilize phasor difference area method to judge the spatial relation of this pixel and convex quadrangle, as P _dg(x _dg, y _dg) not in convex quadrangle or on four edges line, then continue next pixel of process, otherwise this pixel P is obtained in calculating _dgwith the bee-line of scan lines L1, L2, if P _dgbe D1 with the bee-line of L1, calculating the intersection point obtaining D1 place straight line and L1 is P _cg1(x _cg1, y _cg1), if P _dgbe D2 with the bee-line of L2, calculating the intersection point obtaining D2 place straight line and L2 is P _cg2(x _cg2, y _cg2), utilize formula 4 to calculate pixel P _dfour neighborhood pixel coordinate P of conjugate points in oblique distance correction file _nc1(x _nc1, Row1), P _nc2(x _nc2, Row1), P _nc3(x _nc3, Row2) and P _nc4(x _nc4, Row2), finally utilize bilinear interpolation method to determine pixel P _dpixel value.(note: the weight in bilinearity difference process on y direction is determined by D1, D2);

$\left\{\begin{array}{c}{x}_{dg}=(x_d+0.5)*res\mathrm{Re}solution+x\min \\ y_{dg}=y_{\max }-(y_d+0.5)*res\mathrm{Re}solution\end{array}\right.$ Formula 3

$\left\{\begin{array}{c}{x}_{nc1}=ceil(\frac{x_{cg1}-x_{s1}}{R}\×ppc)\\ x_{nc2}=round(\frac{x_{cg1}-x_{s1}}{R}\×ppc)\\ x_{nc3}=ceil(\frac{x_{cg2}-x_{s2}}{R}\×ppc)\\ x_{nc4}=round(\frac{x_{cg2}-x_{s2}}{R}\×ppc)\end{array}\right.$ Formula 4.

Compared with prior art, the invention has the advantages that: take geospatial coordinates as reference, by analyzing the space geometry relation between side-scan sonar sweep trace and resample points, that sets up pixel in resampling image and side-scan sonar data resolves relation, the geocoding of side-scan sonar data can be realized by indirect resampling, and effectively can eliminate all kinds of gaps that side-scan sonar data produce in geocoding process, supply approach for obtaining clear continuous print seabed image; Simultaneously the method is in concrete realization mechanism, can support to read, Local treatment and Vector operation in batches, contributes to the indirect resampling of side-scan sonar data to be reduced to linear operation, and then improves the treatment effeciency of side-scan sonar data geographic coding.

Accompanying drawing explanation

Fig. 1 is the process flow diagram towards the rate method for resampling of side-scanning sonar image in the embodiment of the present invention.

Embodiment

Below in conjunction with accompanying drawing embodiment, the present invention is described in further detail.

Towards a method for resampling for side-scanning sonar image, it is characterized in that: comprise the steps:

Step one: the File header information of resolving side-scanning sonar image file, obtains at least following two property parameters in side-scanning sonar image file: oblique distance range R and passage image width ppc;

Step 2: the scan-line data in order reading side-scanning sonar image file and the towfish elevation information of correspondence, receiving time information and sequence number of scanning lines, utilize oblique distance calibration model to carry out oblique distance correction process to the scan-line data obtained, and result is stored to oblique distance correction file;

In this step, the detailed process utilizing the scan-line data of oblique distance calibration model to acquisition to carry out oblique distance correction process is:

First, after oblique distance being corrected, on side-scanning sonar image, the pixel value of all pixels is initialized as zero;

Then, find out oblique distance and correct the conjugate points of some pixels on side-scanning sonar image file on rear side-scanning sonar image, note P _p(x _p, y _p) be certain pixel after oblique distance corrects on side-scanning sonar image, the sweep trace of its correspondence is L, and when receiving this sweep trace, the resolution of corresponding towfish height to be towfishAlt, sidescanRes be original side-scanning sonar image file, utilizes formula 1 to calculate P _pconjugate points Ps (xs, ys) on side-scanning sonar image file,

formula 1

Finally, the pixel P after the pixel value assignment of conjugate points Ps (xs, ys) in side-scanning sonar image file being corrected to oblique distance on side-scanning sonar image _p(x _p, y _p)

Step 3: order reads the track line file corresponding with side-scanning sonar image file, extract the track points coordinate sequence of towfish, time of reception sequence and Case Number sequence, utilize side-scan sonar towfish flight path processing method, simulate the ship trajectory of towfish reality, and calculate the towfish geographic coordinate residing when receiving every bar scan-line data;

Step 4: on the basis of step 2 and step 3, utilize side-scanning sonar image geo-coding techniques, resolve the geographic coordinate of each sweep trace end points, the each sweep trace end points geographic coordinate calculated is stored to sweep trace extreme coordinates temporary file, calculate the minimum enclosed rectangle of this side-scanning sonar image file institute covering space simultaneously, record space four to the scope that this side-scanning sonar image file covers;

The end points geographic coordinate of note sweep trace L1 ~ Ln is respectively P _s1(x _s1, y _s1), P _e1(x _e1, y _e1), P _s2(x _s2, y _s2), P _e2(x _e2, y _e2) ... P _sn(x _sn, y _sn), P _en(x _en, y _en), if x _min=min{x _s1, x _e1, x _s2, x _e2x _sn, x _en, x _max=max{x _s1, x _e1, x _s2, x _e2x _sn, x _en, y _min=min{y _s1, y _e1, y _s2, y _e2y _sn, y _en, y _max=max{y _s1, y _e1, y _s2, y _e2y _sn, y _en, then four angular coordinates of the minimum enclosed rectangle of side-scanning sonar image file institute covering space are respectively P _tl(x _min, y _max), P _tr(x _max, y _max), P _bl(x _min, y _min), P _br(x _max, y _min);

Step 5: the resampling resolution of space four to the scope covered according to the side-scanning sonar image file calculated in step 4 and needs creates the resampling image of side-scanning sonar image file, and each pixel value in resampling image is initialized as zero;

Step 6: sequentially read scan-line data corresponding in the oblique distance correction file created by step 2, get adjacent two scan-line datas, in the sweep trace extreme coordinates temporary file simultaneously created in step 4, order reads the sweep trace extreme coordinates corresponding with these two sweep traces, build the minimum enclosed rectangle comprising these two sweep traces, then this minimum enclosed rectangle pixel coordinate corresponding in resampling image is calculated according to this rectangle upper left angle point and the geographic coordinate of bottom right angle point and the resampling resolution of needs, obtain the image block of current pending re-sampling operations by this, concrete account form is:

If to get adjacent sweep trace be L1, L2, remember that the line number in its place oblique distance correction file is respectively Row1, Row2, the sweep trace end points geographic coordinate of its correspondence is P _s1(x _s1, y _s1), P _e1(x _e1, y _e1), P _s2(x _s2, y _s2) and P _e2(x _e2, y _e2), if slx _min=min{x _s1, x _e1, x _s2, x _e2, slx _max=max{x _s1, x _e1, x _s2, x _e2, sly _min=min{y _s1, y _e1, y _s2, y _e2, sly _max=max{y _s1, y _e1, y _s2, y _e2, then the geographic coordinate comprising minimum enclosed rectangle four angle points of L1, L2 sweep trace is respectively slP _tl(slx _min, sly _max), slP _tr(slx _max, sly _max), slP _bl(slx _min, sly _min), slP _br(slx _max, sly _min), the pixel coordinate P of height H eight, width W idth that this boundary rectangle is corresponding in resampling image and upper left angle point thereof _rs(x _start, y _start) calculate acquisition by formula 2;

$\left\{\begin{array}{c}{x}_{start}=ceil(\left({\mathrm{slx}}_{min}-x_{min}\right)/res\mathrm{Re}solution)\\ y_{start}=ceil(\left(y_{max}-{\mathrm{sly}}_{max}\right)/res\mathrm{Re}solution)\\ Width=round(\left({\mathrm{slx}}_{\max }-{\mathrm{slx}}_{\min }+0.5\right)/res\mathrm{Re}solution)\\ Height=round(\left({\mathrm{sly}}_{max}-{\mathrm{sly}}_{min}+0.5\right)/res\mathrm{Re}solution)\end{array}\right.$ Formula 2;

Comprise height H eight, the width W idth of minimum enclosed rectangle corresponding in resampling image and the pixel coordinate P of upper left angle point thereof of L1, L2 sweep trace _rs(x _start, y _start) know after, this minimum enclosed rectangle comprising L1, L2 sweep trace is the image block of current pending re-sampling operations;

Step 7: conjugate points is carried out to each pixel obtaining image block in step 6 and resolves, and according to the pixel value of bilinear interpolation determination conjugate points, concrete account form is:

With P _s1(x _s1, y _s1), P _e1(x _e1, y _e1), P _s2(x _s2, y _s2) and P _e2(x _e2, y _e2) be angle point, build convex quadrangle P _s1p _e1p _e2p _s2, the concrete steps that conjugate points resolves are carried out one by one to the pixel in image block as follows: when any pixel in image block, remember that its pixel coordinate in image block is P _d( _xd, _yd), first utilize formula 3 by P _dpoint pixel coordinate transfers corresponding geographic coordinate P to _dg(x _dg, y _dg), utilize phasor difference area method to judge the spatial relation of this pixel and convex quadrangle, as P _dg(x _dg, y _dg) not in convex quadrangle or on four edges line, then continue next pixel of process, otherwise this pixel P is obtained in calculating _dgwith the bee-line of scan lines L1, L2, if P _dgbe D1 with the bee-line of L1, calculating the intersection point obtaining D1 place straight line and L1 is P _cg1(x _cg1, y _cg1), if P _dgbe D2 with the bee-line of L2, calculating the intersection point obtaining D2 place straight line and L2 is P _cg2(x _cg2, y _cg2), utilize formula 4 to calculate pixel P _dfour neighborhood pixel coordinate P of conjugate points in oblique distance correction file _nc1(x _nc1, Row1), P _nc2(x _nc2, Row1), P _nc3(x _nc3, Row2) and P _nc4(x _nc4, Row2), finally according to the pixel value of four neighborhood pixel coordinate place pixels of conjugate points, utilize bilinear interpolation method to determine pixel P _dpixel value, the weight in bilinearity difference process on y direction is determined by D1, D2.

$\left\{\begin{array}{c}{x}_{dg}=(x_d+0.5)*res\mathrm{Re}solution+x\min \\ y_{dg}=y_{\max }-(y_d+0.5)*res\mathrm{Re}solution\end{array}\right.$ Formula 3

$\left\{\begin{array}{c}{x}_{nc1}=ceil(\frac{x_{cg1}-x_{s1}}{R}\×ppc)\\ x_{nc2}=round(\frac{x_{cg1}-x_{s1}}{R}\×ppc)\\ x_{nc3}=ceil(\frac{x_{cg2}-x_{s2}}{R}\×ppc)\\ x_{nc4}=round(\frac{x_{cg2}-x_{s2}}{R}\×ppc)\end{array}\right.$ Formula 4;

Step 8: return step 6, until recording processing all in sweep trace extreme coordinates temporary file is complete.

Below with the side-scan sonar data instance in Hangzhou Wan marine site, above-described embodiment is remarked additionally, the acquisition time of side-scan sonar data is in April, 2008, storage format is XTFRev.18, file is called 200804P1.xtf, towards the method for resampling of side-scanning sonar image in the present embodiment, comprise the steps:

Step one: the File header information of resolving side-scanning sonar image file, reads the basic parameter information of side-scan sonar data, and the oblique distance range R obtaining side-scan sonar is 100m, passage image width ppc is 1024 pixels;

Step 2: order reads the scan-line data in side-scanning sonar image file, scan-line data also claims Ping data, resolve the height t owfishHeight in towfish distance seabed when the scanning sequence number corresponding to every bar sweep trace, time of reception and every bar sweep trace are received, then oblique distance calibration model and formula 1 is utilized to carry out oblique distance correction process to the scan-line data obtained, and correction result is write in temporary file Cor200804P1.dat, this temporary file Cor200804P1.dat is oblique distance correction file;

Step 3: read the track line file CS200804P1.txt corresponding with side-scan sonar data, parse the track points coordinate sequence of towfish, time of reception sequence and Case Number sequence, utilization is disclosed as the side-scan sonar towfish flight path processing method described in the Chinese patent " a kind of side-scan sonar towfish flight path processing method based on track line file " of CN101592731A (patent No. is ZL200910100493.5), smoothing and the interpolation to the flight path of towfish, simulate the ship trajectory of towfish reality, and calculate the towfish geographic coordinate residing when receiving every bar scanning-line signal,

Step 4: on the basis of step 2 and step 3, side-scanning sonar image geo-coding techniques in interim " research of sound sodium image geo-coding techniques is swept in side " of being delivered by Deng Xueqing, Gong Danchao, Luo Rui of utilization is published in " marine charting " the 22nd volume the 4th, resolve the geographic coordinate of each sweep trace end points, calculate the minimum enclosed rectangle of this side-scan sonar data institute covering space; In the present embodiment, the spatial dimension four most P that covers of side-scan sonar data _tl(510463.90,3117182.61), P _tr(510820.70,3117182.61), P _bl(510463.90,3116289.01), P _br(510820.70,3116289.01);

Step 5: establish the image file after resampling to be called G200804P1.GIF, its resolution resResolution is 0.4m, figure image width after then can calculating resampling and height are respectively 892 pixels and 2234 pixels, utilize the raster spatial data transformation warehouse GDAL increased income to create G200804P1.GIF, and its pixel is all initialized as zero;

Step 6: order reads adjacent two scan-line datas stored in oblique distance correction file Cor200804P1.dat, two scan-line datas read first by oblique distance correction file are set to Article 1 sweep trace and Article 2 sweep trace, called after L1, L2 respectively, the end points geographic coordinate that L1 and L2 is corresponding is respectively P _s1(510639.15,3117180.11), P _e1(510812.59,3117121.99), P _s2(510637.65,3117179.05) and P _e2(510811.75,3117121.16), calculate the geographic coordinate slP respectively obtaining and comprise this two sweep trace minimum enclosed rectangle four angle points _tl(510637.65,3117180.11), slP _tr(510812.59,3117180.11), slPbl (510637.65,3117121.16), slP _br(510812.59,3117121.16), utilize formula 2 can obtain the image block treating resampling, remember that this image block is R herein, the pixel coordinate of the upper left angle point of this image block is Prs (434,6), the width of resampling scope is 438, is highly 148;

Step 7: with P _s1, P _e1, P _s2, P _e2for angle point, build convex quadrangle image block R, carry out conjugate points one by one to the pixel in image block R and resolve, concrete steps are as follows: a pixel in optional image block, remember that its pixel coordinate is in the picture P _d(x _d, y _d), as P _dpoint not in convex quadrangle, then continues next pixel of process; When pixel Pd point is in the scope of convex quadrangle, make x in the present embodiment _dvalue be 581, y _dvalue be 56, the image coordinate namely corresponding to pixel Pd is (581,56), utilizes formula 3 to obtain P _dthe geographic coordinate that point is corresponding is P _dg(510696.5,3117160.21), utilize phasor difference area method to judge a pixel P _dgbe in convex quadrangle image block R inner, if P _dgbe D1 with the bee-line of line segment L1, calculating the intersection point obtaining D1 place straight line and L1 is P _cg1, be D2 with the bee-line of line segment L2, calculating the intersection point obtaining D2 place straight line and L2 is P _cg2, have D1=0.65 as calculated, P _cg1(510696.71,3117160.82), D2=0.69, P _cg2(510696.28,3117159.55), the four neighborhood point coordinate utilizing formula 4 to calculate pixel Pd corresponding conjugate points in oblique distance rectification file are respectively P _nc1(589,1), P _nc2(590,1), P _nc3(600,2) and P _nc4(601,2), finally according to the pixel value of four neighborhood pixel coordinate place pixels of conjugate points, utilize bilinear interpolation method to determine pixel P _dpixel value be 198;

Step 8: after the pixel in image block R is all disposed, with Article 2 and Article 3 sweep trace for object, continues to return step 6 and processes, until all sweep trace is disposed.

Claims (5)

1. towards a method for resampling for side-scanning sonar image, it is characterized in that: comprise the steps:

Step one: the File header information of resolving side-scanning sonar image file, obtains at least following two property parameters in side-scanning sonar image file: oblique distance range R and passage image width ppc;

Step 2: the scan-line data in order reading side-scanning sonar image file and the towfish elevation information of correspondence, receiving time information and sequence number of scanning lines, utilize oblique distance calibration model to carry out oblique distance correction process to the scan-line data obtained, and result is stored to oblique distance correction file;

Step 3: order reads the track line file corresponding with side-scanning sonar image file, extract the track points coordinate sequence of towfish, time of reception sequence and Case Number sequence, utilize side-scan sonar towfish flight path processing method, simulate the ship trajectory of towfish reality, and calculate the towfish geographic coordinate residing when receiving every bar scan-line data;

Step 4: on the basis of step 2 and step 3, utilize side-scanning sonar image geo-coding techniques, resolve the geographic coordinate of each sweep trace end points, the each sweep trace end points geographic coordinate calculated is stored to sweep trace extreme coordinates temporary file, calculate the minimum enclosed rectangle of this side-scanning sonar image file institute covering space simultaneously, record space four to the scope that this side-scanning sonar image file covers;

Step 5: the resampling resolution of space four to the scope covered according to the side-scanning sonar image file calculated in step 4 and needs creates the resampling image of side-scanning sonar image file, and each pixel value in resampling image is initialized as zero;

Step 6: sequentially read scan-line data corresponding in the oblique distance correction file created by step 2, get adjacent two scan-line datas, in the sweep trace extreme coordinates temporary file simultaneously created in step 4, order reads the sweep trace extreme coordinates corresponding with these two sweep traces, build the minimum enclosed rectangle comprising these two sweep traces, then this minimum enclosed rectangle pixel coordinate corresponding in resampling image is calculated according to this rectangle upper left angle point and the geographic coordinate of bottom right angle point and the resampling resolution of needs, obtain the image block of current pending re-sampling operations by this,

Step 7: conjugate points is carried out to each pixel obtaining image block in step 6 and resolves, and according to the pixel value of bilinear interpolation determination conjugate points;

Step 8: return step 6, until recording processing all in sweep trace extreme coordinates temporary file is complete.

2. the method for resampling towards side-scanning sonar image according to claim 1, is characterized in that: in described step 2, and the detailed process utilizing the scan-line data of oblique distance calibration model to acquisition to carry out oblique distance correction process is:

First, after oblique distance being corrected, on side-scanning sonar image, the pixel value of all pixels is initialized as zero;

Then, find out oblique distance and correct the conjugate points of some pixels on side-scanning sonar image file on rear side-scanning sonar image, note P _p(x _p, y _p) be certain pixel after oblique distance corrects on side-scanning sonar image, the sweep trace of its correspondence is L, and when receiving this sweep trace, the resolution of corresponding towfish height to be towfishAlt, sidescanRes be original side-scanning sonar image file, utilizes formula 1 to calculate P _pconjugate points Ps (xs, ys) on side-scanning sonar image file:

Finally, the pixel P after the pixel value assignment of conjugate points Ps (xs, ys) in side-scanning sonar image file being corrected to oblique distance on side-scanning sonar image _p(x _p, y _p).

3. the method for resampling towards side-scanning sonar image according to claim 2, is characterized in that: in described step 4, and the end points geographic coordinate of note sweep trace L1 ~ Ln is respectively P _s1(x _s1, y _s1), P _e1(x _e1, y _e1), P _s2(x _s2, y _s2), P _e2(x _e2, y _e2) ... P _sn(x _sn, y _sn), P _en(x _en, y _en), if x _min=min{x _s1, x _e1, x _s2, x _e2x _sn, x _en, x _max=max{x _s1, x _e1, x _s2, x _e2x _sn, x _en, y _min=min{y _s1, y _e1, y _s2, y _e2y _sn, y _en, y _max=max{y _s1, y _e1, y _s2, y _e2y _sn, y _en, then four angular coordinates of the minimum enclosed rectangle of side-scanning sonar image file institute covering space are respectively P _tl(x _min, y _max), P _tr(x _max, y _max), P _bl(x _min, y _min), P _br(x _max, y _min), in described step 5, if resResolution is the resolution of the resampling image needed, with P _tl, P _tr, P _bl, P _brbe four to border, create the resampling image of side-scanning sonar image file, and the pixel value of each pixel in resampling image is set to zero.

4. the method for resampling towards side-scanning sonar image according to claim 3, is characterized in that: in described step 6, if to get adjacent sweep trace be L1, L2, remember that the line number in its place oblique distance correction file is respectively Row1, Row2, the sweep trace end points geographic coordinate of its correspondence is P _s1(x _s1, y _s1), P _e1(x _e1, y _e1), P _s2(x _s2, y _s2) and P _e2(x _e2, y _e2), if slx _min=min{x _s1, x _e1, x _s2, x _e2, slx _max=max{x _s1, x _e1, x _s2, x _e2, sly _min=min{y _s1, y _e1, y _s2, y _e2, sly _max=max{y _s1, y _e1, y _s2, y _e2, then the geographic coordinate comprising minimum enclosed rectangle four angle points of L1, L2 sweep trace is respectively slP _tl(slx _min, sly _max), slP _tr(slx _max, sly _max), slP _bl(slx _min, sly _min), slP _br(slx _max, sly _min), the pixel coordinate P of height H eight, width W idth that this boundary rectangle is corresponding in resampling image and upper left angle point thereof _rs(x _start, y _start) calculate acquisition by formula 2;

$\left\{\begin{array}{c}{x}_{start}=ceil(\left({\mathrm{slx}}_{min}-x_{min}\right)/res\mathrm{Re}solution)\\ y_{start}=ceil(\left(y_{max}-{\mathrm{sly}}_{max}\right)/res\mathrm{Re}solution)\\ Width=round(\left({\mathrm{slx}}_{\max }-{\mathrm{slx}}_{\min }+0.5\right)/res\mathrm{Re}solution)\\ Height=round(\left({\mathrm{sly}}_{max}-{\mathrm{sly}}_{min}+0.5\right)/res\mathrm{Re}solution)\end{array}\right.$ Formula 2.

5. the method for resampling towards side-scanning sonar image according to claim 4, is characterized in that: in described step 7, with P _s1(x _s1, y _s1), P _e1(x _e1, y _e1), P _s2(x _s2, y _s2) and P _e2(x _e2, y _e2) be angle point, build convex quadrangle P _s1p _e1p _e2p _s2, the concrete steps that conjugate points resolves are carried out one by one to the pixel in image block as follows: when any pixel in image block, remember that its pixel coordinate in image block is P _d( _xd, _yd), first utilize formula 3 by P _dpoint pixel coordinate transfers corresponding geographic coordinate P to _dg(x _dg, y _dg), utilize phasor difference area method to judge the spatial relation of this pixel and convex quadrangle, as P _dg(x _dg, y _dg) not in convex quadrangle or on four edges line, then continue next pixel of process, otherwise this pixel P is obtained in calculating _dgwith the bee-line of scan lines L1, L2, if P _dgbe D1 with the bee-line of L1, calculating the intersection point obtaining D1 place straight line and L1 is P _cg1(x _cg1, y _cg1), if P _dgbe D2 with the bee-line of L2, calculating the intersection point obtaining D2 place straight line and L2 is P _cg2(x _cg2, y _cg2), utilize formula 4 to calculate pixel P _dfour neighborhood pixel coordinate P of conjugate points in oblique distance correction file _nc1(x _nc1, Row1), P _nc2(x _nc2, Row1), P _nc3(x _nc3, Row2) and P _nc4(x _nc4, Row2), finally utilize bilinear interpolation method to determine pixel P _dpixel value;

$\left\{\begin{array}{c}{x}_{dg}=(x_d+0.5)*res\mathrm{Re}solution+x\min \\ y_{dg}=y_{\max }-(y_d+0.5)*res\mathrm{Re}solution\end{array}\right.$ Formula 3

$\left\{\begin{array}{c}{x}_{nc1}=ceil(\frac{x_{cg1}-x_{s1}}{R}\×ppc)\\ x_{nc2}=round(\frac{x_{cg1}-x_{s1}}{R}\×ppc)\\ x_{nc3}=ceil(\frac{x_{cg2}-x_{s2}}{R}\×ppc)\\ x_{nc4}=round(\frac{x_{cg2}-x_{s2}}{R}\×ppc)\end{array}\right.$ Formula 4.