CN106251368B - The fusion method of SAR image and multispectral image based on BEMD - Google Patents
The fusion method of SAR image and multispectral image based on BEMD Download PDFInfo
- Publication number
- CN106251368B CN106251368B CN201610407820.1A CN201610407820A CN106251368B CN 106251368 B CN106251368 B CN 106251368B CN 201610407820 A CN201610407820 A CN 201610407820A CN 106251368 B CN106251368 B CN 106251368B
- Authority
- CN
- China
- Prior art keywords
- component
- image
- imf
- residual components
- sar image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000007500 overflow downdraw method Methods 0.000 title claims abstract description 23
- 230000009466 transformation Effects 0.000 claims abstract description 51
- 230000004927 fusion Effects 0.000 claims abstract description 27
- 238000000354 decomposition reaction Methods 0.000 claims description 17
- 238000001914 filtration Methods 0.000 claims description 15
- 230000006870 function Effects 0.000 claims description 10
- 238000001228 spectrum Methods 0.000 claims description 6
- 238000012952 Resampling Methods 0.000 claims description 5
- 238000001514 detection method Methods 0.000 claims description 4
- 238000012545 processing Methods 0.000 abstract description 8
- 238000004458 analytical method Methods 0.000 abstract description 6
- VMXUWOKSQNHOCA-UKTHLTGXSA-N ranitidine Chemical compound [O-][N+](=O)\C=C(/NC)NCCSCC1=CC=C(CN(C)C)O1 VMXUWOKSQNHOCA-UKTHLTGXSA-N 0.000 abstract description 6
- 238000000034 method Methods 0.000 description 25
- 230000008569 process Effects 0.000 description 10
- 230000003287 optical effect Effects 0.000 description 7
- 238000003384 imaging method Methods 0.000 description 5
- 238000005070 sampling Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 241000282693 Cercopithecidae Species 0.000 description 1
- 241001269238 Data Species 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000007499 fusion processing Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000002844 melting Methods 0.000 description 1
- 230000008018 melting Effects 0.000 description 1
- 238000002156 mixing Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000035515 penetration Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000000513 principal component analysis Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011946 reduction process Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 238000000844 transformation Methods 0.000 description 1
- 238000013519 translation 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
- G06T5/70—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
- G06T2207/10016—Video; Image sequence
-
- 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
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
A kind of fusion method of SAR image and multispectral image of the present invention, comprising: IHS is carried out to multispectral image and converts to obtain I, H, S component；BEMD is carried out respectively to SAR image and I component to convert to obtain the IMF component and residual components of SAR image and I component；The IMF component and residual components for adjusting SAR image and I component are consistent the IMF component number of the two, and corresponding residual components are added in extra IMF component；The IMF component of SAR image adjusted and I component and IMF component and residual components are merged with residual components respectively；BEMD inverse transformation is carried out according to fusion results, obtains fused I component；IHS inverse transformation is carried out to fused I component, H component and S component, obtains fused image.The fusion method of SAR image and multispectral image based on BEMD of the invention saves preferable texture information, and reliable guarantee can be provided for the specific subdivision of similar atural object, is improved to non-linear, non-stationary signal analysis and processing capacity.
Description
Technical field
The present invention relates to melting for image processing field more particularly to a kind of SAR image and multispectral image based on BEMD
Conjunction method.
Background technique
It is multi-platform, stage construction, multidate, multisensor, more with the fast development of sensor technology, remote sensing technology etc.
The remote sensing image of spectrum and multiresolution constitutes the multi-source remote sensing information of multizone.Different sensors has different imagings
Mechanism and different wave-length coverages is worked in, therefore image obtained often reflects the feature of atural object different aspect.Pass through list
Information provided by one sensor may be not comprehensive, inconsistent, even inaccurate.In order to which more effectively utilization is increasingly multiple
Miscellaneous Multi-source Remote Sensing Image Information, associated image fusion technology come into being.
Synthetic aperture radar (Synthetic Aperture Radar, SAR) is a kind of microwave remote sensor, is passed through actively
Transmitting electromagnetic wave and the electromagnetic wave information realization remotely sensed image for receiving return.For optical sensor, synthetic aperture thunder
Up to round-the-clock, round-the-clock, high-resolution imaging capability, and its operating distance is remote, and areas imaging is wide, penetration capacity
By force, solve the problems, such as that optical remote sensing sensor is influenced by time and weather.
Currently, common fusion method has: IHS transformation, PCA transformation, Brovey transformation and wavelet transformation, wherein IHS
Transformation, PCA transformation and Brovey transformation normally result in more serious spectrum distortion, and wavelet transformation then solves spectrum torsion
Bent problem.However, the above method is mainly used for the fusion of multispectral image and panchromatic image, proposed in this paper to optics shadow
As a large amount of texture informations can be lost using the above method when being merged with High-resolution SAR Images, carried out to subtle atural object
Classification has great difficulty when extracting.
In the following, a more detailed description is done to the prior art in above-mentioned Remote Sensing Image Fusion field, so as to more preferable
Ground understands the present invention.
Traditional Remote Sensing Image Fusion generally uses optics panchromatic image to be merged with multispectral image, traditional fusion
Method include: IHS transformation, principal component transform (PCA), ratio fusion method, Principal Component Analysis (Karhunen-Loeve transformation), Weighted Fusion method,
Wavelet Transform etc..
By taking IHS is converted as an example: the different-waveband remotely-sensed data that will be obtained by modes such as optics, thermal infrared and radars (microwave)
One RGB color that object color attribute is described of synthesis is transformed to brightness (Intensity), coloration H
(Hue), saturation degree S (Saturation) describes the color space IHS of image.
The blending algorithm of IHS transformation is as follows:
It is positive to become formula:
H=arctg (v1/v2)
In formula: I indicates brightness, and H indicates coloration, and S indicates saturation degree, and v1, v2 are to calculate the intermediate of I, H introducing and become
Amount.
It is counter to change formula:
By taking wavelet transformation as an example: the process of Wavelet Transform Fusion is as follows:
Wavelet transformation is defined as:
Convert kernel function are as follows:
Wherein,For a wavelet, also known as morther wavelet or wavelet basis, it is by t=0 be centered on
Band pass function, and time domain average value
Wavelet transformation is, low-frequency information is handled with space large scale, with the small scale processing high-frequency information in space.
Currently, the multiscale analysis method used in the fusion method of SAR image and optical image specifically include that two dimension from
Dissipate wavelet transformation, shift-invariant spaces and indicate anisotropic warp wavelet etc..Those skilled in the art answer
Work as understanding, wavelet transformation is inherently a kind of linear transformation, and to non-linear, the processing capacity of non-stationary signal is limited；Moreover,
Wavelet transformation is non-adaptive, selection of the syncretizing effect depending on wavelet basis.
Therefore, it is necessary to a kind of method for merging High-resolution SAR Images with optical image, this method can overcome
Existing method is handling non-linear, more out of strength on non-stationary signal defect, and at the same time being able to maintain in multi-source image
Spectral information, marginal information and texture information.
The meaning of the abbreviation of Partial key term in this specification is shown herein, comprising: the one-dimensional empirical modal of EMD
It decomposes (Empirical Mode Decomposition, EMD)；Two-dimensional empirical mode decomposition
(BidimensionalEmpirical Mode Decomposition,BEMD)；Intrinsic mode functions (Intrinsic Mode
Function,IMF)；Synthetic aperture radar ((SyntheticApertureRadar, SAR).
Summary of the invention
The object of the present invention is to provide synthetic aperture radar SAR images and multispectral shadow that one kind can overcome drawbacks described above
The fusion method of picture.
In a first aspect, the present invention provides the fusion method of a kind of synthetic aperture radar SAR image and multispectral image,
It is characterized by: carrying out IHS transformation to multispectral image, the I component, H component and S component of the multispectral image are obtained；It is right
The SAR image and the I component carry out two-dimensional empirical mode decomposition BEMD transformation respectively, obtain the intrinsic of the SAR image
The IMF component and residual components of modular function IMF component and residual components and the I component；To IMF points of the SAR image
The IMF component and residual components of amount and residual components and the I component are adjusted, and make the IMF component of the SAR image
It is consistent with the IMF component number of the I component, and corresponding residual components is added in extra IMF component；Respectively by institute
The residual components and I for stating the IMF component of SAR image adjusted and the IMF component of I component and SAR image adjusted divide
The residual components of amount are merged；BEMD inverse transformation is carried out according to fusion results, obtains fused I component；To fused I
Component, the H component and the S component carry out IHS inverse transformation, obtain fused image.
Preferably, the step of BEMD is converted includes: to initialize to the image of pending BEMD transformation；Identification is just
The extreme point in image after beginningization；In the case where the extreme point meets predetermined threshold, by being carried out to the extreme point
Fitting, obtains the average envelope face of image；The local trend of image is extracted according to the average envelope face；In the local trend
In the case where meeting scheduled IMF condition, the local trend is determined as IMF component；It is calculated according to the IMF component remaining
Component；In the case where the residual components meet predetermined monotony condition, the residual components are determined as final remnants points
Amount.
Preferably, it in the case where the local trend is unsatisfactory for IMF decision condition, is replaced initially with the local trend
Image after change continually looks for extreme point therein；In the case where the residual components are unsatisfactory for monotony condition, with described residual
Remaining component replaces the image after initialization, continually looks for extreme point therein.
Preferably, the IMF decision condition is according in the extreme point in the image for wherein identifying extreme point and between zero point
Relationship and the average envelope face value determine；The monotony condition by two in succession determine come IMF component meter
It obtains.
Preferably, before the step of carrying out IHS transformation to multispectral image, further includes: gone to the SAR image
It makes an uproar；Pass through the pixel resolution of SAR image and the multispectral image after the unified denoising of resampling；After resampling
SAR image and multispectral image carry out Image registration.
Preferably, before the fusion the step of, further includes: IMF component and I point to the SAR image adjusted
The IMF component of amount carries out Laplce's high-pass filtering, and the remnants of the residual components and I component to SAR image adjusted
Component carries out Laplce's low-pass filtering；Result based on the filtering calculates separately IMF component in the SAR image and residual
The weight of each pixel in remaining component；Wherein, the fusion is carried out based on the weight.
The fusion method of SAR image and multispectral image based on BEMD of the invention saves preferable texture information,
Reliable guarantee can be provided for the specific subdivision of similar atural object, improved to non-linear, non-stationary signal analysis and processing
Ability.
Detailed description of the invention
Fig. 1 is the flow chart according to the fusion method of the SAR and multispectral image based on BEMD of the embodiment of the present invention；
Fig. 2 is the flow chart of the BEMD transformation in Fig. 1；
Fig. 3 is the fusion results figure of the multi-source Remote Sensing Images of the Guilin Area based on wavelet transformation；
Fig. 4 is the fusion results figure of the multi-source Remote Sensing Images of the Guilin Area based on fusion method of the invention.
Specific embodiment
Below by drawings and examples, technical scheme of the present invention will be described in further detail.
Non-linear, more out of strength on non-stationary signal lack is being handled for existing Remote Sensing Image Fusion transform method
It falls into, the present invention proposes a kind of new Remote Sensing Image Fusion transform method.BEMD transformation is suitable for non-linear, non-stationary signal point
Analysis has compared with high s/n ratio；In addition, the time first reading feature of BEMD basis signal itself realizes signal decomposition, final fusion effect
Fruit does not depend on the selection of any basic function, has complete adaptivity.Using BEMD transform method, the present invention realizes SAR
Image is merged with high score optical image.
Firstly, image fusing method of the present invention is broadly described as follows several steps.
1) the object image of multi-source RS Images Fusion is pre-processed.
For example, the multispectral image that the SAR image obtained with TerraSAR-X satellite and GF-1 camera obtain can be chosen
Carry out visual fusion.Preferably, be to guarantee syncretizing effect, respectively first to two kinds of image datas carry out geographical calibrations, geometric correction,
The pretreatment such as Image registration, filtering and noise reduction.Since two kinds of images have different spatial resolutions, before carrying out Image registration
It need to unify the pixel resolution of image；In addition, need to disappear in preprocessing process since there are incidence angles when obtaining for SAR image
Except ghost image.
2) for the calculating circulation during visual fusion the problems such as reasonably chooses the big of image pixel and imaging window
It is small.Image pixel size can choose 8 multiple, be preferably best with square；And image should not be too large, such as can select
Taking imaging window size is 4000*4000.
3) BEMD decomposition is carried out to multispectral image and SAR image.
Multispectral image contains there are three wave band R, G, B, and the color space IHS three-component has relative independentability, Neng Gougeng
Good embodiment color information.Therefore, IHS transformation is first carried out to multispectral image, it is rear to carry out BEMD decomposition, i.e., to three of them wave band
It is decomposed respectively.In an experiment, decomposition can be defined and obtain four layers of component, respectively three IMF components and a remnants divide
Amount.First IMF component is most high-frequency information, and second and three-component frequency information successively reduce, and residual components are predominantly bright
Spend information.
Only one wave band of SAR image, i.e. gray scale wave band, then also carrying out BEMD decomposition to the gray scale wave band of SAR image.
4) by after decomposition SAR image greyscale wave band with decompose after multispectral image I component merge, then with H,
The merging of S component is reconverted into R, G, B, finally obtains fusion evaluation.
In the following, the detailed description side according to the present invention of fusion based on BEMD high-resolution SAR image and optical image
Method.
Fig. 1 is the process according to the fusion method of the SAR image and multispectral image based on BEMD of the embodiment of the present invention
Figure.
Step 1: Yunnan snub-nosed monkey, process mainly includes image resampling, the denoising of SAR image, Image registration.
It is possible, firstly, to pass through the pixel resolution of the unified two width images of down-sampling.
Secondly as SAR image is vulnerable to influence of noise, it is empty with image so as to cause the non-linear of image, non-stationary signal
Between resolution ratio improve and further obvious, intrinsic noise is also more prominent therewith.Therefore, it is necessary to preceding first to it in the sampling of SAR image
Denoising is carried out, such as noise reduction process can be carried out to SAR image using Goldstein filtering is improved.
Finally, acquiring corresponding image points to progress for the SAR image after denoising, sampling and the multispectral image after sampling
Image registration.
Step 2: SAR image is merged with multispectral image, specific steps are as follows:
1, to multispectral image carry out IHS transformation (i.e. color space conversion in Fig. 1), obtain multispectral image I, H,
S component.
H=arctg (v1/v2)
Note: v1, v2 are intermediate variable.
2, BEMD transformation is carried out respectively to pretreated SAR image and transformed multispectral image I component.
By BEMD transformation respectively, the IMF component and residual components R of SAR image are obtained, and obtain multispectral image
IMF component and residual components R.After BEMD is decomposed, SAR image is represented byMultispectral shadow
As I component is represented byWherein, I, R are respectively IMF component and residual components；M, N is respectively
The number of the decomposable IMF of the I component of SAR and multispectral image.
3, the image decomposition amount of SAR image and multispectral image is handled, makes the two Decomposition order having the same.
Since BEMD has complete adaptivity, the I component of SAR image and multispectral image after BEMD decomposition by obtaining
IMF number may be different, so as to cause inconsistent in subsequent fusion process.Therefore, it is necessary to the decomposition amounts to image
It is handled, keeps the Decomposition order of image to be fused consistent.Specifically, it setting can will be obtained after decomposing in BEMD transformation
IMF number, so that the finally obtained IMF number of two kinds of images is identical.For example, set n=min { M, N }, by taking n=M as an example, mostly light
IMFs component after n-th of IMF of the I component of spectrum image will be added into residual components, and making its IMF number is n.This
When, the I component of multispectral image can indicate are as follows:
Wherein,
4, Laplce's filtering is carried out by the component that BEMD is converted to SAR image and multispectral image, with detection
Marginal information in image.For example, the IMF component to SAR image and multispectral image carries out Laplce's high-pass filtering, to two
The residual components of person carry out Laplce's low-pass filtering.Marginal information in image mainly corresponds to high-frequency information, that is, passes through BEMD
The IMF component respectively obtained, correspondingly, low-frequency information therein are residual components.
5, the pixel window determined based on Laplace operator, it is whether smooth in detection window region.Pass through detection, judgement
Whether the information of identical frequency is obtained after previous step filtering.Such as it is judged as smooth, then show that high-pass filtering obtains is all high
Frequency information, what low-pass filtering obtained is all low-frequency information.In other words, smoothly show that the information obtained after filtering does not have in frequency
Jump sex differernce, i.e., no edge.
6, in the case where judging that pixel window inner region is smooth, indicate that frequency information is same or similar, then deducibility picture
It is same atural object in plain window, then calculating its average weight；Otherwise it needs to calculate separately weight.For example, can basis
Calculate the weight of each pixel in each component of SAR image based on the fusion rule of provincial characteristics, and by the IMF component of SAR image
It is denoted as α respectively with the weight of residual componentsn、β。
7, by IMF componentAnd residual componentsThe fusion of different scale is carried out respectively, and fusion is public
Formula is as follows:
Rfuse=βRSF+(1-β)RMI
6, BEMD inverse transformation is carried out, using fusion results as new I component Ifuse, wherein inverse transformation formula is as follows:
7, by new I component IfuseColor space inverse transformation is carried out with H, S component of multispectral image, i.e., by IfuseHMSM
Rgb space is changed in inversion, obtains fused image.
Fig. 2 is the flow chart of the BEMD transformation in Fig. 1.
In step 201, the image of pending BEMD transformation is initialized.
The extreme point of image after step 202, identification initialization specifically finds its all local maximum
Point and minimum point.
In step 203, judge whether the number of extreme point meets predetermined condition.If it is satisfied, then process proceeds to 204；
Otherwise process proceeds to step 207.
In step 204, the maximum point and minimum point of effective interpolation algorithm difference fitted signal are selected, is obtained up and down
Enveloping surface, and average envelope face is calculated.
In step 205, the local trend of image is extracted according to average envelope face；
In step 206, judge whether the local trend meets IMF condition.If satisfied, then process proceeds to step 207；It is no
Then, the image of the initialization in step 202 is replaced with the local trend, process returns to step 202.
IMF decision condition are as follows:
1) for entire data set, | extreme point-zero point |≤1；
2) the average envelope face amount determined for any point on data set, local maximum and local minimum is zero.
In step 207, which is set as IMF component.
In step 208, residual components are calculated according to the IMF component.
Whether the residual components being calculated in step 209, judgment step 207 meet monotony condition.As met, then flow
Journey terminates, which is determined as to the residual components of final output；Otherwise it is replaced with the residual components first in step 202
Beginningization image, process return to step 202.
Residual components whether Dan Tiao criterion are as follows:
1) any IMF is not included；Or
2) residual components be less than SD, 0.2 < SD < 0.3,
In general, the range of SD setting is 0.2-0.3, i.e., when SD meets 0.2 < SD < 0.3, screening process terminates.
Fig. 3 is the fusion results figure of the multi-source Remote Sensing Images of the Guilin Area based on wavelet transformation；Fig. 4 is based on the present invention
Fusion method Guilin Area multi-source Remote Sensing Images fusion results figure.It can obviously be seen by the comparison of Fig. 3 and Fig. 4
Out, fusion method according to the present invention saves preferable texture information, and the specific subdivision of similar atural object can be provided can
By guaranteeing, improve to non-linear, non-stationary signal analysis and processing capacity.
The present invention is based on BEMD transformation to merge to multi-source Remote Sensing Images, can as the basis of classification of remote-sensing images
Guarantee while handling linear perhaps non-linear steady or non-stationary signal, guarantees the light that multi-source Remote Sensing Images are included
Spectrum information, marginal information and texture information are not lost.Traditional SAR mainly uses two-dimensional discrete small with optical image fusion method
The linear transformations such as wave conversion, translation invariant wavelet, these methods largely abandon non-linear and non-stationary letter
Number processing.Integration program provided by the invention is more suitable for non-linear, non-stationary signal analysis, noise with higher
Than.In addition, BEMD transformation is that it is different from wavelet transformation when to signal processing dependent on wavelet basis better than wavelet transformation
Selection, it is to rely on signal temporal scale feature to realize signal decomposition, has complete adaptivity.
Professional should further appreciate that, described in conjunction with the examples disclosed in the embodiments of the present disclosure
Unit and algorithm steps, can be realized with electronic hardware, computer software, or a combination of the two, hard in order to clearly demonstrate
The interchangeability of part and software generally describes each exemplary composition and step according to function in the above description.
These functions are implemented in hardware or software actually, the specific application and design constraint depending on technical solution.
Professional technician can use different methods to achieve the described function each specific application, but this realization
It should not be considered as beyond the scope of the present invention.
The step of method described in conjunction with the examples disclosed in this document or algorithm, can be executed with hardware, processor
The combination of software module or the two is implemented.Software module can be placed in random access memory (RAM), memory, read-only memory
(ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or technical field
In any other form of storage medium well known to interior.
Above-described specific embodiment has carried out further the purpose of the present invention, technical scheme and beneficial effects
It is described in detail.It should be understood that being not intended to limit the present invention the foregoing is merely a specific embodiment of the invention
Protection scope.All within the spirits and principles of the present invention, any modification, equivalent substitution, improvement and etc. done should all include
Within protection scope of the present invention.
Claims (5)
1. a kind of fusion method of synthetic aperture radar SAR image and multispectral image based on BEMD, it is characterised in that:
IHS transformation is carried out to multispectral image, obtains the I component, H component and S component of the multispectral image；
Two-dimensional empirical mode decomposition BEMD transformation is carried out to the SAR image and the I component respectively, obtains the SAR image
Intrinsic mode functions IMF component and residual components and the I component IMF component and residual components；
The IMF component and residual components of IMF component and residual components and the I component to the SAR image are adjusted
It is whole, it is consistent the IMF component of the SAR image and the IMF component number of the I component, and extra IMF component is added
Enter corresponding residual components；
The IMF component of IMF component and I component to the SAR image adjusted carries out Laplce's high-pass filtering, and right
The residual components of SAR image adjusted and the residual components of I component carry out Laplce's low-pass filtering；
Whether smoothly to be determined in detection window region based on the result of the filtering；
If smooth, its average weight is calculated；
If unsmooth, the weight of each pixel in the IMF component and residual components in the SAR image is calculated separately；
Wherein, the fusion is carried out based on the weight；
Respectively by the IMF component and SAR image adjusted of the IMF component of the SAR image adjusted and I component
Residual components are merged with the residual components of I component；
BEMD inverse transformation is carried out according to fusion results, obtains fused I component；
IHS inverse transformation is carried out to fused I component, the H component and the S component, obtains fused image；
While handling linear perhaps non-linear steady or non-stationary signal, guarantee the light that multi-source Remote Sensing Images are included
Spectrum information, marginal information and texture information are not lost.
2. the fusion method of the synthetic aperture radar SAR image and multispectral image according to claim 1 based on BEMD, special
The step of sign is, the BEMD is converted include:
The image of pending BEMD transformation is initialized；
The extreme point in image after identification initialization；
In the case where the extreme point meets predetermined threshold, by being fitted to the extreme point, being averaged for image is obtained
Enveloping surface；
The local trend of image is extracted according to the average envelope face；
In the case where the local trend meets scheduled IMF condition, the local trend is determined as IMF component；
Residual components are calculated according to the IMF component；
In the case where the residual components meet predetermined monotony condition, the residual components are determined as final remnants points
Amount.
3. the fusion method of the synthetic aperture radar SAR image and multispectral image according to claim 2 based on BEMD, special
Sign is:
In the case where the local trend is unsatisfactory for IMF decision condition, the image after initializing is replaced with the local trend,
Continually look for extreme point therein；
In the case where the residual components are unsatisfactory for monotony condition, the image after initializing is replaced with the residual components, after
It is continuous to find extreme point therein.
4. the fusion method of the synthetic aperture radar SAR image and multispectral image according to claim 2 based on BEMD, special
Sign is:
The IMF decision condition is according in the extreme point in the image for wherein identifying extreme point and relationship and institute between zero point
The value for stating average envelope face determines；
The monotony condition by two in succession determine come IMF component be calculated.
5. the fusion method of the synthetic aperture radar SAR image and multispectral image according to claim 1 based on BEMD, special
Sign is, before the step of carrying out IHS transformation to multispectral image, further includes:
The SAR image is denoised；
Pass through the pixel resolution of SAR image and the multispectral image after the unified denoising of resampling；
To after resampling SAR image and multispectral image carry out Image registration.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610407820.1A CN106251368B (en) | 2016-06-12 | 2016-06-12 | The fusion method of SAR image and multispectral image based on BEMD |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610407820.1A CN106251368B (en) | 2016-06-12 | 2016-06-12 | The fusion method of SAR image and multispectral image based on BEMD |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106251368A CN106251368A (en) | 2016-12-21 |
| CN106251368B true CN106251368B (en) | 2019-05-14 |
Family
ID=57612923
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610407820.1A Active CN106251368B (en) | 2016-06-12 | 2016-06-12 | The fusion method of SAR image and multispectral image based on BEMD |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106251368B (en) |
Families Citing this family (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107480619B (en) * | 2017-08-03 | 2019-11-15 | 中国地质大学（武汉） | Noise-reduction method and system based on EEMD and the Ground Penetrating Radar B-scan image for arranging entropy |
| CN107644408A (en) * | 2017-09-27 | 2018-01-30 | 安徽大学 | A kind of polarization radar image denoising method based on Anisotropic diffusion |
| CN108549902B (en) * | 2018-03-14 | 2022-08-16 | 中国科学院遥感与数字地球研究所 | Improved SAR image and multi-spectrum optical image fusion method |
| CN109118463B (en) * | 2018-07-27 | 2021-10-19 | 中国科学院国家空间科学中心 | SAR image and optical image fusion method based on HSL and image entropy |
| CN109345555B (en) * | 2018-10-15 | 2020-08-25 | 中科卫星应用德清研究院 | Method for identifying rice based on multi-temporal multi-source remote sensing data |
| CN109886904B (en) * | 2019-01-25 | 2021-08-10 | 北京市遥感信息研究所 | SAR image and low-resolution multispectral image fusion method and system |
| CN111598115B (en) * | 2019-02-21 | 2022-06-28 | 兰州交通大学 | SAR image fusion method based on cross cortical neural network model |
| CN109924990A (en) * | 2019-03-27 | 2019-06-25 | 兰州大学 | A kind of EEG signals depression identifying system based on EMD algorithm |
| CN111090129B (en) * | 2019-12-31 | 2022-06-28 | 核工业北京地质研究院 | Fast searching method for ore control structure of hard rock type uranium ore based on multi-source data fusion |
| CN112258853A (en) * | 2020-10-19 | 2021-01-22 | 洛阳云感科技有限公司 | Chain visibility monitoring and early warning system in highway fog zone becomes more meticulous |
| CN113436069B (en) * | 2021-06-16 | 2022-03-01 | 中国电子科技集团公司第五十四研究所 | Remote sensing image fusion method based on maximum signal-to-noise ratio projection |
| CN117252785B (en) * | 2023-11-16 | 2024-03-12 | 安徽省测绘档案资料馆(安徽省基础测绘信息中心) | Cloud removing method based on combination of multisource SAR and optical image |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2013173436A1 (en) * | 2012-05-16 | 2013-11-21 | Synta Pharmaceuticals Corp. | Pre-selection of subjects for therapeutic treatment with an hsp90 inhibitor based on hypoxic status |
| CN104021536A (en) * | 2014-06-16 | 2014-09-03 | 西北工业大学 | Self-adaptation SAR image and multispectral image fusion method |
| CN105825491A (en) * | 2016-03-17 | 2016-08-03 | 江苏科技大学 | Image fusion method based on hybrid model |
-
2016
- 2016-06-12 CN CN201610407820.1A patent/CN106251368B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2013173436A1 (en) * | 2012-05-16 | 2013-11-21 | Synta Pharmaceuticals Corp. | Pre-selection of subjects for therapeutic treatment with an hsp90 inhibitor based on hypoxic status |
| CN104021536A (en) * | 2014-06-16 | 2014-09-03 | 西北工业大学 | Self-adaptation SAR image and multispectral image fusion method |
| CN105825491A (en) * | 2016-03-17 | 2016-08-03 | 江苏科技大学 | Image fusion method based on hybrid model |
Non-Patent Citations (5)
| Title |
|---|
| "二维Hilbert-Huang变换及其在图像处理中的应用";乔丽红;《中国博士学位论文全文数据库 信息科技辑》;20101015;论文第33页 * |
| "使用二维EMD 的红外和彩色可见光图像融合";张秀琼等;《计算机工程与应用》;20101231;第46卷(第16期);论文第182-184页 * |
| "基于Contourlet 与HIS 变换的SAR与多光谱影像融合方法";黄登山等;《宇航学报》;20110131;第32卷(第1期);论文第188-192页 * |
| "基于EMD的SAR与全色影像融合算法";陈云等;《计算机应用与软件》;20160531;第33卷(第5期);论文第177-180页 * |
| "多源遥感影像数据融合理论与技术";杨丽坤等;《北京测绘》;20141231;论文第101-104页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106251368A (en) | 2016-12-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106251368B (en) | The fusion method of SAR image and multispectral image based on BEMD | |
| Agrawal et al. | A novel joint histogram equalization based image contrast enhancement | |
| Hu et al. | Single image defogging based on illumination decomposition for visual maritime surveillance | |
| US8457437B2 (en) | System and method for enhancing registered images using edge overlays | |
| Shahdoosti et al. | Pansharpening of clustered MS and Pan images considering mixed pixels | |
| Khan et al. | Localization of radiance transformation for image dehazing in wavelet domain | |
| EP2157544A1 (en) | System for adaptive removal of speckle noise in digital images and generation of a colour composite product based on automated analysis of speckle characteristics | |
| Ganasala et al. | Contrast enhanced multi sensor image fusion based on guided image filter and NSST | |
| CN112307901A (en) | Landslide detection-oriented SAR and optical image fusion method and system | |
| Cai et al. | Infrared and visible image fusion based on BEMSD and improved fuzzy set | |
| Bi et al. | Haze removal for a single remote sensing image using low-rank and sparse prior | |
| Khosravi et al. | An introduction to ENVI tools for synthetic aperture radar (SAR) image despeckling and quantitative comparison of denoising filters | |
| CN110348442A (en) | A kind of shipborne radar image sea oil film recognition methods based on support vector machines | |
| CN107451608B (en) | SAR image non-reference quality evaluation method based on multi-view amplitude statistical characteristics | |
| Asghar et al. | Automatic enhancement of digital images using cubic Bézier curve and Fourier transformation | |
| Boldt et al. | SAR image segmentation using morphological attribute profiles | |
| Trongtirakul et al. | Fractional Contrast Stretching for Image Enhancement of Aerial and Satellite Images. | |
| Khosravi et al. | A tutorial and performance analysis on ENVI tools for SAR image despeckling | |
| Reddy | Digital image processing: Principles and applications | |
| Parihar | Histogram modification and DCT based contrast enhancement | |
| Li et al. | Simultaneous smoothing and sharpening using iwgif | |
| Priyanka et al. | Contrast enhancement of night images | |
| Pandey et al. | Single Image Defogging Algorithm-Based on Modified Haze Line Prior | |
| Pappas et al. | Curvelet fusion of panchromatic and SAR satellite imagery using fractional lower order moments | |
| Starovoitov | Comparative Analysis of Some Dynamic Range Reduction Methods for SAR Image Visualization |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |