CN108062767A - Statistics based on sequential SAR image is the same as distribution space pixel selecting method - Google Patents
Statistics based on sequential SAR image is the same as distribution space pixel selecting method Download PDFInfo
- Publication number
- CN108062767A CN108062767A CN201810141076.4A CN201810141076A CN108062767A CN 108062767 A CN108062767 A CN 108062767A CN 201810141076 A CN201810141076 A CN 201810141076A CN 108062767 A CN108062767 A CN 108062767A
- Authority
- CN
- China
- Prior art keywords
- pixel
- sar
- distribution
- image
- statistics
- 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
- 238000009826 distribution Methods 0.000 title claims abstract description 42
- 238000000034 method Methods 0.000 title claims abstract description 26
- 238000003657 Likelihood-ratio test Methods 0.000 claims abstract description 23
- 238000001914 filtration Methods 0.000 abstract description 20
- 238000012360 testing method Methods 0.000 abstract description 7
- 230000000694 effects Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 4
- 238000003018 immunoassay Methods 0.000 description 4
- 238000005191 phase separation Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 230000001427 coherent effect Effects 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000004445 quantitative analysis Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000004451 qualitative analysis 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
- 238000010187 selection method Methods 0.000 description 1
- 230000001629 suppression Effects 0.000 description 1
Classifications
-
- 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
- G06T7/38—Registration of image sequences
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
- G06T2207/10044—Radar image
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)
Abstract
The invention discloses a kind of statistics based on sequential SAR image with distribution space pixel selecting method, pretreatment is carried out to original SAR image data sequence first and obtains haplopia SAR intensity image sequences;Registration is carried out to SAR intensity image sequences;Haplopia SAR intensity images after registration are obeyed under the hypothesis of index statistical distribution, obtain the region of rejection of likelihood ratio test;Compare the statistics similarity of the time samples of the time samples and center reference pixel of each space pixel in rectangular slide window;Each space pixel in whole image is traveled through, the statistics for obtaining each space pixel is distributed sample together;SAR image is filtered using the same distribution sample of Lee filtering and each space pixel;The present invention will be come out by likelihood ratio hypothesis testing with sample selection of the reference pixel with same alike result, is more in line with the space distribution rule and backscattering characteristic of pixel in radar image, is obtained the filtering image that full resolution is more nearly with real surface.
Description
Technical field
The invention belongs to SAR data processing technology fields, and in particular to a kind of statistics based on sequential SAR image is the same as distribution
Space pixel selecting method.
Background technology
In recent years, with novel satellite borne SAR (Synthetic Aperture Radar, synthetic aperture radar) sensor
Sequentially emit, time series SAR technologies occupy more and more important position in radar remote sensing application.Due to SAR echo signal
It is the cumulative of scattering object back scattering contribution in ground resolution unit, these scattering objects show random phase, cause SAR numbers
It is believed that number formation coherent spot.The coherent speckle noise in SAR data how is effectively removed, recovers true SAR reflected values, is always
The difficult point of SAR researchs.SAR phase separation immunoassays since the eighties be suggested, such as regard filtering, Lee filtering, MAP-Sigma
Filtering, Kuan filtering, IDAN filtering and NLSAR filtering etc..But these filtering weaken noise using local spatial statistics so that
Spatial resolution is difficult to maintain, and quality margin signal is lost or the loss of signal is difficult to weigh with noise suppressed.
SAR filtering methods based on time series carry out pixel selection due to the use of time samples, are not related to spatial operation,
There is absolute advantage in solution more than problem.This kind of method compares aerial image under nonparametric or parameter hypothesis test frame
Plain similarity, but it is because it is assumed that the size for examining itself effect, so first and second class error of different hypothesis testings differs,
Cause actual filter effect ginseng layer uneven.In contrast, the effect of parameter hypothesis test is more preferable, especially likelihood ratio test, quilt
Riemann-Pearson came theorem proving is optimal hypothesis testing.However, the region of rejection of likelihood ratio test is often difficult to provide, thus
It is limited in practical application.
The content of the invention
Present invention aims to SAR filter results to reflect true Radar backscattering coefficients, when proposing that one kind is based on
The statistics of sequence SAR image realizes that SAR with samples selection is distributed, solves likelihood ratio test and exists with distribution space pixel selecting method
Compare border caused by participating in filtering due to heterogeneous pixel in limitation traditional SAR image phase separation immunoassay on the pixel similarity of space
The technical issues of fuzzy and image fault.
The present invention adopts the following technical scheme that a kind of statistics based on sequential SAR image is the same as distribution space pixel selection side
Method is as follows:
1) original SAR image data sequence is pre-processed, obtains haplopia SAR intensity image sequences;
2) registration is carried out to pretreated haplopia SAR intensity image sequences, obtains the haplopia SAR intensity images after registration
Sequence;
3) haplopia SAR intensity images after registration are obeyed under the hypothesis of index statistical distribution, obtain likelihood ratio test
Region of rejection;
4) rectangular slide window of the size as m × m is set, m is pixel number, and m is odd number, every in comparison window one by one
The statistics similarity of the time samples of a space pixel and the time samples of center reference pixel, sets null hypothesis as two times
Sample has same distribution, if null hypothesis is overthrown in likelihood ratio test under the conditions of given level of signifiance α, the space pixel is in
Heart reference pixel is heterogeneous pixel;Otherwise likelihood ratio test receives null hypothesis, i.e., the space pixel and center reference pixel are same
It is distributed pixel;
5) step 4) is repeated, travels through each space pixel in whole image size, obtains the statistics of each space pixel
With distribution sample;
6) using wave filter, the pixel in original Directional Windows mouth in wave filter is obtained instead of above-mentioned each space pixel
Statistics with being distributed sample, SAR image is filtered, exports result.
Preferably, using plural modulus square method obtain haplopia SAR intensity image sequences.
Preferably, registration uses maximum intensity cross correlation algorithm.
Preferably, obtain the region of rejection of likelihood ratio test specifically, for obey exponential distribution time samples { x1,
x2,…,xnAnd { y1,y2,…,yn, the reduced form of the likelihood ratio test under null hypothesis is obtained, when setting null hypothesis as two
Between sample have same distribution:
Wherein Λ is the equivalent-simplification statistic of likelihood ratio test,WithThe average value of two time samples is represented respectively;
F (2n, 2n) represents that obey degree of freedom is distributed for the F of (2n, 2n), and n represents time series SAR image quantity.
Preferably, in comparison window the time samples of the time samples and center reference pixel of each space pixel statistics
Similarity is specifically, in the window for being m × m in size, the time samples { y of more each space pixel i one by one1 i,y2 i,…,
yn iAnd center reference pixel { x1,x2,…,xnStatistics similarity, i=1,2 ..., m × m, in given level of signifiance α conditions
Under, when the estimated result of the equivalent-simplification statistic Λ of likelihood ratio test falls in region of rejection, overthrow null hypothesis, i.e. two spaces
Pixel is dissimilar, which is heterogeneous pixel with center reference pixel;Otherwise null hypothesis is received, two pixels are same are distributed
Pixel, discrimination formula are as follows:
Or
WhereinWithRepresent that F (2n, 2n) is distributed upper respectivelyWithDivide position
Point.
Preferably, wave filter uses Lee wave filters.
The reached advantageous effect of invention:The present invention is a kind of statistics based on sequential SAR image with distribution space pixel
Selection method realizes SAR with samples selection is distributed, and solves in traditional SAR image phase separation immunoassay since heterogeneous pixel participates in filtering
Obscurity boundary caused by ripple and image fault technical problem;On the basis of sequential SAR image, the likelihood with region of rejection is utilized
The space similar neighborhood pixel of each space pixel is picked out than Statistical Identifying Method, avoids and has in phase separation immunoassay
There is obscuring for the topographical features of different scattering mechanisms;During same distribution statistics samples selection, it is two fingers to have used for reference F distributions
Simplifying for number distribution likelihood ratio test statistic is distributed this property, gives fixed region of rejection so that this power of test
It is optimal, the same distribution pixel confidence level higher of selection;All same distributed collections of selection are filtered for Lee, it can be true
The spatial resolution and details for protecting filtering image retain, and take into account noise suppressed maximization.
Description of the drawings
Fig. 1 is the flow chart of the present invention;
Fig. 2 is the same distribution pixel sample set schematic diagram that the present invention is extracted using the likelihood ratio test with region of rejection;
Fig. 3 is raw noise image and various filtering method result schematic diagrams.
Specific embodiment
Below according to attached drawing and technical scheme is further elaborated in conjunction with the embodiments.
Fig. 1 is the flow chart of the present invention, and the statistics based on sequential SAR image is with distribution space pixel selecting method, specifically
Step is as follows:
1) original SAR image data sequence is pre-processed, using square acquisition haplopia SAR intensity maps of plural modulus
As sequence;
2) registration is carried out to pretreated haplopia SAR intensity image sequences, obtains the haplopia SAR intensity images after registration
Sequence;
For the validity of qualitative and quantitative analysis the method for the present invention, data use X-band Germany TerraSAR-X radars
Satellite single-polarized data, image incident angle are 37 °, and spatial resolution is 3m × 3m (distance to × orientation).In order to enhance
The estimated accuracy of hypothesis testing carries out registration to original image first, using maximum intensity cross correlation algorithm, chooses match point
256 × 256, match window size is 64 × 64.
It should be noted that:The method of the present invention is applicable not only to the TerraSAR-X data chosen in the experiment, to it
He is spaceborne and on-board data is equally applicable, simply needs to choose different according to the topographical features in different research areas with punctual
With quantity and window size.
3) haplopia SAR intensity images after registration are obeyed under the hypothesis of index statistical distribution, obtain likelihood ratio test
Region of rejection;
For obeying the time samples { x of exponential distribution1,x2,…,xnAnd { y1,y2,…,yn, it obtains under null hypothesis seemingly
So than the reduced form examined, set null hypothesis has same distribution as two time samples:
Wherein Λ is the equivalent-simplification statistic of likelihood ratio test,WithCenter reference pixel and neighbour to be measured are represented respectively
The average value of two time samples of domain pixel;F (2n, 2n) represents that obey degree of freedom is distributed for the F of (2n, 2n), and n represents time sequence
Row SAR image quantity.
4) Fig. 2 is the same distribution pixel sample set schematic diagram that the present invention is extracted using the likelihood ratio test with region of rejection
(a) to be superimposed upon 15 × 15 rectangular slide windows in SAR image;(b) the same distribution pixel sample set chosen in window.If
The rectangular slide window that a fixed size is m × m, m are pixel number, m=15, the time samples { y of each space pixel i1 i,
y2 i,…,yn iSee dot in Fig. 2 (a), center reference pixel { x1,x2,…,xnSee the triangulation points of Fig. 2 (a) centers, by
The statistics similarity of a relatively the two, i=1,2 ..., m × m, under the conditions of given level of signifiance α=0.05, when likelihood ratio is examined
The estimated result of the equivalent-simplification statistic Λ tested falls in region of rejection, overthrows null hypothesis, i.e. two spaces pixel is dissimilar, should
Space pixel is heterogeneous pixel with center reference pixel;Otherwise null hypothesis is received, two pixels are with distribution pixel, see Fig. 2 (b)
In point, discrimination formula is as follows:
Or
WhereinWithRepresent that F (2n, 2n) is distributed upper respectivelyWithDivide position
Point.
It should be noted that:The size of sliding window is determined by the resolution ratio of selected SAR image, which can
With other spaceborne and carried SAR data point reuses according to selection.
5) step 4) is repeated, travels through each space pixel in whole image size, until obtaining each space pixel
Statistics is the same as distribution sample;
6) using Lee wave filters, by the pixel in original Directional Windows mouth in wave filter instead of above-mentioned each space pixel
The statistics of acquisition is filtered SAR image with sample is distributed, and exports result.Fig. 3 is raw noise image and various filtering sides
Method result schematic diagram;(a) it is raw noise image, (b) is rule window filter result, and (c) is existing tradition Lee filtering knots
Fruit, (d) are the result that the present invention filters.
For the effect that statistics of the invention in quantitative analysis Fig. 3 (d) is filtered with distribution samples selection in SAR image, adopt
Keep the precision of filter result in index SNR evaluation inventions with intensity of speckle noise index SSI and resolution ratio, and in Fig. 3 (a)
Raw noise image, the rule window filter result in Fig. 3 (b), traditional Lee filter results in Fig. 3 (c) are compared,
Table 1 is the precision comparative result of different filtering methods.
It should be noted that:Rule window filters and what tradition Lee filtering used is 7 × 7 sizes.Refer in evaluation
In number, intensity of speckle noise index SSI is smaller, illustrates that noise suppression effect is better;Resolution ratio keeps SNR numerical value bigger, explanation
Edge keeps effect better.
The precision of the different filtering methods of table 1 compares
The statistics of the present invention compared with rule window and tradition Lee filtering methods is can be seen that with distribution from table 1 and Fig. 3
Samples selection has better effect:In the filter result obtained using institute's extracting method of the present invention, the border of atural object it is apparent and
Continuously, the similar region of scattering mechanism has obtained preferable differentiation, and final result is more nearly with real surface.
The above is only the preferred embodiment of the present invention, it is noted that for the ordinary skill people of the art
For member, without departing from the technical principles of the invention, several improvement and deformation can also be made, these are improved and deformation
Also it should be regarded as protection scope of the present invention.
Claims (6)
1. a kind of statistics based on sequential SAR image is the same as distribution space pixel selecting method, which is characterized in that including following step
Suddenly:
1) original SAR image data sequence is pre-processed, obtains haplopia SAR intensity image sequences;
2) registration is carried out to pretreated haplopia SAR intensity image sequences, obtains the haplopia SAR intensity image sequences after registration
Row;
3) haplopia SAR intensity images after registration are obeyed under the hypothesis of index statistical distribution, obtain the refusal of likelihood ratio test
Domain;
4) rectangular slide window of the size as m × m is set, m is pixel number, and m is odd number, each empty in comparison window one by one
Between pixel time samples and center reference pixel time samples statistics similarity, set null hypothesis as two time samples
With same distribution, if null hypothesis is overthrown in likelihood ratio test under the conditions of given level of signifiance α, which joins with center
It is heterogeneous pixel to examine pixel;Otherwise likelihood ratio test receives null hypothesis, i.e., the space pixel and center reference pixel are same are distributed
Pixel;
5) step 4) is repeated, travels through each space pixel in whole image size, the statistics until obtaining each space pixel
With distribution sample;
6) it is using wave filter, the pixel in original Directional Windows mouth in wave filter is same instead of the statistics of above-mentioned each space pixel
Sample is distributed, original SAR image is filtered, exports result.
With distribution space pixel selecting method, 2. feature exists the statistics according to claim 1 based on sequential SAR image
In, in the step 1) using plural modulus square method obtain haplopia SAR intensity image sequences.
With distribution space pixel selecting method, 3. feature exists the statistics according to claim 1 based on sequential SAR image
In registration uses maximum intensity cross correlation algorithm in step 2).
With distribution space pixel selecting method, 4. feature exists the statistics according to claim 1 based on sequential SAR image
In obtaining the region of rejection of likelihood ratio test in the step 3) specifically, time samples { x for obeying exponential distribution1,x2,…,
xnAnd { y1,y2,…,yn, the reduced form of the likelihood ratio test under null hypothesis is obtained, sets null hypothesis as two time samples
With same distribution:
<mrow>
<mi>&Lambda;</mi>
<mo>=</mo>
<mfrac>
<mover>
<mi>x</mi>
<mo>&OverBar;</mo>
</mover>
<mover>
<mi>y</mi>
<mo>&OverBar;</mo>
</mover>
</mfrac>
<mo>~</mo>
<mi>F</mi>
<mrow>
<mo>(</mo>
<mn>2</mn>
<mi>n</mi>
<mo>,</mo>
<mn>2</mn>
<mi>n</mi>
<mo>)</mo>
</mrow>
</mrow>
Wherein Λ is the equivalent-simplification statistic of likelihood ratio test,WithThe average value of two time samples is represented respectively;F(2n,
2n) represent that obey degree of freedom is distributed for the F of (2n, 2n), n represents time series SAR image quantity.
With distribution space pixel selecting method, 5. feature exists the statistics according to claim 1 based on sequential SAR image
In, in step 4) in comparison window the time samples and the time samples of center reference pixel of each space pixel statistics phase
Like degree specifically, in the window for being m × m in size, the time samples { y of more each space pixel i one by one1 i,y2 i,…,yn i}
With center reference pixel { x1,x2,…,xnStatistics similarity, i=1,2 ..., m × m, under the conditions of given level of signifiance α,
When the estimated result of the equivalent-simplification statistic Λ of likelihood ratio test falls in region of rejection, null hypothesis, i.e. two spaces picture are overthrown
Plain dissimilar, which is heterogeneous pixel with center reference pixel;Otherwise null hypothesis is received, two pixels are same distribution images
Element, discrimination formula are as follows:
Or
WhereinWithRepresent that F (2n, 2n) is distributed upper respectivelyWithQuantile.
With distribution space pixel selecting method, 6. feature exists the statistics according to claim 1 based on sequential SAR image
In in step 6) median filter using Lee wave filters.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810141076.4A CN108062767B (en) | 2018-02-11 | 2018-02-11 | Statistical same-distribution spatial pixel selection method based on time sequence SAR image |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810141076.4A CN108062767B (en) | 2018-02-11 | 2018-02-11 | Statistical same-distribution spatial pixel selection method based on time sequence SAR image |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108062767A true CN108062767A (en) | 2018-05-22 |
CN108062767B CN108062767B (en) | 2022-02-11 |
Family
ID=62134370
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810141076.4A Active CN108062767B (en) | 2018-02-11 | 2018-02-11 | Statistical same-distribution spatial pixel selection method based on time sequence SAR image |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108062767B (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110412574A (en) * | 2019-09-05 | 2019-11-05 | 河海大学 | A kind of distributed object InSAR timing sequence process method and apparatus of temporal and spatial coherence enhancing |
CN113610783A (en) * | 2021-07-22 | 2021-11-05 | 中山大学 | Time sequence SAR intensity image variation coefficient-based change detection method and device |
CN115267781A (en) * | 2022-09-28 | 2022-11-01 | 中山大学 | InSAR coherence estimation method based on multi-view SAR data set |
CN115291214A (en) * | 2022-09-28 | 2022-11-04 | 中山大学 | Time sequence multi-polarization SAR homogeneous sample selection method |
CN116226789A (en) * | 2023-05-08 | 2023-06-06 | 锋睿领创(珠海)科技有限公司 | Data co-distribution judging method, device, equipment and medium based on artificial intelligence |
CN117848351A (en) * | 2024-03-05 | 2024-04-09 | 河海大学 | Underwater autonomous navigation system and method based on credibility and statistical similarity measurement |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101894215A (en) * | 2010-07-06 | 2010-11-24 | 厦门大学 | Likelihood ratio test error detection method |
ES2467066B1 (en) * | 2012-04-13 | 2015-03-31 | Universitat Politècnica De Catalunya | Procedure for the Validation of the Hypothesis of the Random-Volume-over-Ground Model and its Application in the Estimation of Data in the Study and Characterization of Vegetable Coverings through Multichannel Coherent Systems |
CN104515980A (en) * | 2014-12-08 | 2015-04-15 | 广西大学 | Method and device for ground moving target indication based on InSAR (interferometric synthetic aperture radar) formation |
CN106960443A (en) * | 2017-03-21 | 2017-07-18 | 民政部国家减灾中心 | The method and device of non-supervisory change detection based on complete polarization sequential SAR image |
CN107507209A (en) * | 2017-08-07 | 2017-12-22 | 西安理工大学 | The sketch map extracting method of Polarimetric SAR Image |
-
2018
- 2018-02-11 CN CN201810141076.4A patent/CN108062767B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101894215A (en) * | 2010-07-06 | 2010-11-24 | 厦门大学 | Likelihood ratio test error detection method |
ES2467066B1 (en) * | 2012-04-13 | 2015-03-31 | Universitat Politècnica De Catalunya | Procedure for the Validation of the Hypothesis of the Random-Volume-over-Ground Model and its Application in the Estimation of Data in the Study and Characterization of Vegetable Coverings through Multichannel Coherent Systems |
CN104515980A (en) * | 2014-12-08 | 2015-04-15 | 广西大学 | Method and device for ground moving target indication based on InSAR (interferometric synthetic aperture radar) formation |
CN106960443A (en) * | 2017-03-21 | 2017-07-18 | 民政部国家减灾中心 | The method and device of non-supervisory change detection based on complete polarization sequential SAR image |
CN107507209A (en) * | 2017-08-07 | 2017-12-22 | 西安理工大学 | The sketch map extracting method of Polarimetric SAR Image |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110412574A (en) * | 2019-09-05 | 2019-11-05 | 河海大学 | A kind of distributed object InSAR timing sequence process method and apparatus of temporal and spatial coherence enhancing |
CN110412574B (en) * | 2019-09-05 | 2021-05-25 | 河海大学 | Distributed target InSAR time sequence processing method and device with enhanced space-time coherence |
CN113610783A (en) * | 2021-07-22 | 2021-11-05 | 中山大学 | Time sequence SAR intensity image variation coefficient-based change detection method and device |
CN115267781A (en) * | 2022-09-28 | 2022-11-01 | 中山大学 | InSAR coherence estimation method based on multi-view SAR data set |
CN115291214A (en) * | 2022-09-28 | 2022-11-04 | 中山大学 | Time sequence multi-polarization SAR homogeneous sample selection method |
CN115291214B (en) * | 2022-09-28 | 2022-12-16 | 中山大学 | Time sequence multi-polarization SAR homogeneous sample selection method |
CN116226789A (en) * | 2023-05-08 | 2023-06-06 | 锋睿领创(珠海)科技有限公司 | Data co-distribution judging method, device, equipment and medium based on artificial intelligence |
CN116226789B (en) * | 2023-05-08 | 2023-08-18 | 锋睿领创(珠海)科技有限公司 | Data co-distribution judging method, device, equipment and medium based on artificial intelligence |
CN117848351A (en) * | 2024-03-05 | 2024-04-09 | 河海大学 | Underwater autonomous navigation system and method based on credibility and statistical similarity measurement |
CN117848351B (en) * | 2024-03-05 | 2024-05-28 | 河海大学 | Underwater autonomous navigation system and method based on credibility and statistical similarity measurement |
Also Published As
Publication number | Publication date |
---|---|
CN108062767B (en) | 2022-02-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108062767A (en) | Statistics based on sequential SAR image is the same as distribution space pixel selecting method | |
Chierchia et al. | Multitemporal SAR image despeckling based on block-matching and collaborative filtering | |
CN102609701B (en) | Remote sensing detection method based on optimal scale for high-resolution SAR (synthetic aperture radar) | |
CN105205484B (en) | Synthetic aperture radar target detection method based on warp wavelet and Wiener filtering | |
US7116265B2 (en) | Recognition algorithm for the unknown target rejection based on shape statistics obtained from orthogonal distance function | |
CN105759268B (en) | The adaptive quick determination methods of SAR image CFAR based on multithreading | |
CN101727662A (en) | SAR image nonlocal mean value speckle filtering method | |
CN108171193B (en) | Polarized SAR (synthetic aperture radar) ship target detection method based on super-pixel local information measurement | |
CN105160648B (en) | Radar target and Shadow segmentation method based on small echo and constant false alarm rate | |
JP2008292449A (en) | Automatic target identifying system for detecting and classifying object in water | |
Iqbal et al. | SAR image despeckling by selective 3D filtering of multiple compressive reconstructed images | |
CN106128121A (en) | Vehicle queue length fast algorithm of detecting based on Local Features Analysis | |
CN106295498A (en) | Remote sensing image target area detection apparatus and method | |
CN107680061A (en) | Dual-polarization SAR image speckle filtering method based on similarity test | |
Chu et al. | Technical framework for shallow-water bathymetry with high reliability and no missing data based on time-series sentinel-2 images | |
CN101794437A (en) | Method for detecting abnormal target in hyperspectral remotely sensed image | |
CN108805057A (en) | A kind of SAR image oil depot area detection method based on joint significance analysis | |
Ding et al. | Coastline detection in SAR images using multiscale normalized cut segmentation | |
Fayad et al. | Hy-TeC: a hybrid vision transformer model for high-resolution and large-scale mapping of canopy height | |
CN110390338A (en) | A kind of SAR high-precision matching process based on non-linear guiding filtering and ratio gradient | |
CN112989940A (en) | Raft culture area extraction method based on high-resolution three-satellite SAR image | |
CN105204010A (en) | Ground object target detection method of low signal-to-clutter ratio synthetic aperture radar image | |
CN115390047B (en) | Satellite-borne photon counting laser radar data denoising and filtering method and device | |
Ayhan et al. | Practical considerations in unsupervised change detection using SAR images | |
Stevens et al. | Quantitative data quality metrics for 3D laser radar systems |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |