CN110533600A - A kind of same/heterogeneous remote sensing image high-fidelity broad sense sky-spectrum fusion method - Google Patents
A kind of same/heterogeneous remote sensing image high-fidelity broad sense sky-spectrum fusion method Download PDFInfo
- Publication number
- CN110533600A CN110533600A CN201910619429.1A CN201910619429A CN110533600A CN 110533600 A CN110533600 A CN 110533600A CN 201910619429 A CN201910619429 A CN 201910619429A CN 110533600 A CN110533600 A CN 110533600A
- Authority
- CN
- China
- Prior art keywords
- remote sensing
- image
- sensing image
- denoted
- fidelity
- 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
- 238000001228 spectrum Methods 0.000 title claims abstract description 55
- 238000007500 overflow downdraw method Methods 0.000 title claims abstract description 24
- 230000004927 fusion Effects 0.000 claims abstract description 69
- 230000003595 spectral effect Effects 0.000 claims abstract description 43
- 238000012952 Resampling Methods 0.000 claims abstract description 32
- 238000011156 evaluation Methods 0.000 claims abstract description 28
- 238000013461 design Methods 0.000 claims abstract description 22
- 230000002708 enhancing effect Effects 0.000 claims abstract description 6
- 238000000034 method Methods 0.000 claims description 59
- 238000012545 processing Methods 0.000 claims description 32
- 238000002347 injection Methods 0.000 claims description 9
- 239000007924 injection Substances 0.000 claims description 9
- 238000005516 engineering process Methods 0.000 claims description 8
- 230000005855 radiation Effects 0.000 claims description 8
- 238000001914 filtration Methods 0.000 claims description 5
- 239000000243 solution Substances 0.000 claims description 5
- 241000287196 Asthenes Species 0.000 claims description 3
- 238000012937 correction Methods 0.000 claims description 3
- 230000008901 benefit Effects 0.000 abstract description 4
- 239000004615 ingredient Substances 0.000 abstract description 3
- 238000000701 chemical imaging Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 230000000295 complement effect Effects 0.000 description 3
- 230000000007 visual effect Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000007499 fusion processing Methods 0.000 description 2
- 238000002156 mixing Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 239000000155 melt Substances 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/89—Radar or analogous systems specially adapted for specific applications for mapping or imaging
- G01S13/90—Radar or analogous systems specially adapted for specific applications for mapping or imaging using synthetic aperture techniques, e.g. synthetic aperture radar [SAR] techniques
- G01S13/9021—SAR image post-processing techniques
- G01S13/9027—Pattern recognition for feature extraction
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/86—Combinations of radar systems with non-radar systems, e.g. sonar, direction finder
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/89—Radar or analogous systems specially adapted for specific applications for mapping or imaging
- G01S13/90—Radar or analogous systems specially adapted for specific applications for mapping or imaging using synthetic aperture techniques, e.g. synthetic aperture radar [SAR] techniques
-
- 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
- G06T5/00—Image enhancement or restoration
- G06T5/73—Deblurring; Sharpening
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N2021/1793—Remote sensing
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30181—Earth observation
Landscapes
- Engineering & Computer Science (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Electromagnetism (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Artificial Intelligence (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a kind of same/heterogeneous remote sensing image high-fidelity broad sense sky-spectrum fusion methods, it can meet panchromatic/multispectral simultaneously, panchromatic/EO-1 hyperion homogeneity image high-fidelity fusion, and high spatial resolution SAR image and the multispectral heterogeneous image high-fidelity of low spatial resolution merge, generate while having the remote sensing image of high spatial resolution and high spectral resolution, it considers demand of the practical engineering application to quickly handling, basic framework is merged based on simple and quick ingredient replacement class, extract the high-frequency information of high spatial resolution remote sense image, and the wave band feature design high-frequency information based on high spectral resolution remote sensing image injects weight, the high spectral resolution remote sensing image that high-frequency information injects after resampling is obtained into spatial altitude and sharpens fusion evaluation；Advantage is that it has more robustness to the spatial registration heterogeneous image, and can meet homogeneity/heterogeneous image sky-spectrum high-fidelity fusion demand simultaneously in view of space enhancing, the influence of spectrally compensating and noise simultaneously.
Description
Technical field
The present invention relates to a kind of Remote Sensing Image Processing Technologies, wide more particularly, to a kind of same/heterogeneous remote sensing image high-fidelity
Adopted sky-spectrum fusion method.
Background technique
Remote sensing image is the important carrier for obtaining earth's surface information, wherein while there is high spatial resolution and EO-1 hyperion point
The remote sensing image of resolution plays a significant role in fields such as information interpretation, terrain classification identifications.However, by the limit of satellite sensor
The influence of system and other factors, the remote sensing image of acquisition mutually restrict in high spatial resolution and two aspect of high spectral resolution and
It can not get both.Such as IKONOS, QuickBird, high score No.1, high score two domestic and international remote sensing satellites provide panchromatic image simultaneously
And multispectral image, wherein panchromatic image has high spatial resolution, but only one wave band；Multispectral image has opposite
Preferable spectral resolution, but its spatial resolution is often lower.Sky-spectrum integration technology can integrate between multi-source Remote Sensing Images
High spatial resolution and high spectral resolution complementary advantage generate while having the distant of high spatial resolution and high spectral resolution
Feel image.
Currently, having a large amount of skies-spectrum fusion method, such as panchromatic/multispectral image merges, panchromatic/Hyperspectral imaging merges,
Multispectral/Hyperspectral imaging fusion, exemplary process have ingredient to replace class fusion method, multiresolution analysis fusion method, base
In variation Optimum Fusion, fusion method based on deep learning etc..These methods are often referred to sky-spectrum under chivalrous understanding and melt
Conjunction method, i.e. sky between homogeneity remote sensing image-spectrum fusion.Though secondly, SAR/ multispectral image Pixel-level fusion belong to it is heterogeneous
Remote Sensing Image Fusion obtains high spatial point if it is intended to the spatially and spectrally complementary information using SAR image and multispectral image
Resolution multispectral image also belongs to broad sense sky-spectrum fusion scope.Currently, SAR/ multispectral image fusion method is adopted mostly
With or use for reference panchromatic/Multi-spectral image fusion frame and method, but compared to homogeneity remote sensing image sky-spectrum fusion, due to SAR image with
Multispectral image otherness is larger, therefore existing method is more challenged in terms of the enhancing of fusion evaluation space and spectrum fidelity.Always
For body, on the one hand, although part homogeneity remote sensing image sky-spectrum fusion method, such as ingredient replacement class fusion method and more resolutions
Rate analysis fusion method has been integrated in professional remote sensing software platform, but most existing methods are difficult in terms of fusion accuracy and efficiency
It gets both；On the other hand, most existing panchromatic/multispectral homogeneity Remote Sensing Image Fusions and SAR/ multispectral image are heterogeneous distant
Although sense visual fusion using similar or consistent frame, is difficult to carry out sky-spectrum fusion of homogeneity remote sensing image and heterogeneous distant
Feel sky-spectrum fusion collaboration processing of image.Therefore, how to develop same/heterogeneous collaboration fusion method of multi-source Remote Sensing Images, simultaneously
Meet the fusion of the homogeneities remote sensing image high-fidelity such as panchromatic/multispectral, panchromatic/EO-1 hyperion, the multispectral heterogeneous remote sensing image height of SAR/
Fidelity merges demand, this is undoubtedly of great significance and practical application value.
Summary of the invention
Technical problem to be solved by the invention is to provide a kind of same/heterogeneous remote sensing image high-fidelity broad sense sky-spectrum fusions
Method generates by the sky of fusion multi-source homogeneity or heterogeneous remote sensing image, spectrum complementary information while having high spatial resolution
With the remote sensing image of high spectral resolution, panchromatic/multispectral, panchromatic/EO-1 hyperion, multispectral/EO-1 hyperion not only can satisfy
Homogeneity remote sensing image high-fidelity sky-spectrum merges demand, and can satisfy the multispectral heterogeneous remote sensing image high-fidelity sky-spectrum of SAR/
Fusion demand.
The technical scheme of the invention to solve the technical problem is: a kind of same/heterogeneous remote sensing image high-fidelity is wide
Adopted sky-spectrum fusion method, it is characterised in that the following steps are included:
Step 1: choosing the multi-source Remote Sensing Images for being directed to Same Scene, respectively original high spatial resolution remote sense image
With original high spectral resolution remote sensing image, correspondence is denoted as IgkAnd Igg；Then to IgkAnd IggIt is pre-processed, will be pre-processed
High spatial resolution remote sense image and high spectral resolution remote sensing the image correspondence obtained afterwards is denoted as I* gkAnd I* gg；
Step 2: judging I* gkIn whether include noise, use Denoising Algorithm to I if comprising noise* gkQuickly denoised
Processing, the high spatial resolution remote sense image obtained after denoising is denoted asDirectly by I if not including noise* gkAgain
It is denoted as
Step 3: extractingHigh-frequency information, detailed process are as follows:
Step 3_1: in IgkSpectral region and IggSpectral region known in situation, according to IgkSpectral region and
IggSpectral region, select by IgkSpectral region covering IggWave band i.e. select by IgkSpectral region covering's
Wave band；Then basisSize, to I* ggCarry out Design Based on Spatial Resampling, the high spectral resolution that will be obtained after Design Based on Spatial Resampling
Rate remote sensing image is denoted asThen according to I* gg, and by selectionWave band carry out linear combination obtainBrightness
Component；
Or in IgkSpectral region and IggSpectral region it is unknown in the case where, according toSize, to I* ggInto
The high spectral resolution remote sensing image obtained after Design Based on Spatial Resampling is denoted as by row Design Based on Spatial ResamplingThen according to I* gg, and lead to
It is right to crossAll wave bands carry out linear combination obtainLuminance component；
It is above-mentioned, it is rightAll wave bands carry out linear combination used by combination coefficient acquisition process are as follows: it is rightIt carries out
The high spatial resolution remote sense image obtained after space desampling is denoted as by space desamplingIn IgkSpectral region and Igg
Spectral region known in situation, according to IgkSpectral region and IggSpectral region, select by IgkSpectral region covering
IggWave band i.e. select by IgkSpectral region coveringWave band, using least square method pairWith selection
I* ggWave band handled, solution obtain combination coefficient；Or in IgkSpectral region and IggThe unknown situation of spectral region
Under, using least square method pairAnd I* ggAll wave bands handled, solution obtain combination coefficient；
Step 3_2: willLuminance component as reference, it is rightMatch by moment is carried out to weakenLuminance component with
Between radiation difference, the high spatial resolution remote sense image obtained after match by moment is denoted as
Step 3_3: willWithLuminance component subtract each other, using obtained error image asBasic high fdrequency component；
Then pass through laplacian spectral radius filter pairBasic high fdrequency component carry out space enhancing, obtainAdjustable height frequency division
Amount；Again willBasic high fdrequency component andTunable high-frequency component weighted array obtainHigh-frequency information；Wherein,'s
The value range of the weight of tunable high-frequency component is [0,5]；
Step 4: utilizing I* ggEach wave band gradient, calculate I* ggEach wave band required forHigh-frequency information
Inject weight；
Step 5: by I* ggEach wave band required forHigh-frequency information injection weight withHigh-frequency information phase
Multiply；Then obtained result is injected intoIn, there is the preliminary sky of high spatial resolution and high spectral resolution simultaneously
Between highly sharpen fusion evaluation, be denoted as Irh；
Step 6: utilizing I* ggTo IrhSpectrally compensating and correction are carried out, the optimal spatial height for obtaining spectrum high-fidelity sharpens
Fusion evaluation.
In the step 1, I* gkAnd I* ggAcquisition process are as follows:
Step 1_1: according to IgkAnd IggBetween space resolution rate, to IggDesign Based on Spatial Resampling is carried out, space is adopted again
The high spectral resolution remote sensing image obtained after sample, is denoted asSize and IgkSize it is consistent；Then
It willAs reference, to IgkGeometrical registration is carried out, using the high spatial resolution remote sense image obtained after geometrical registration as first
High spatial resolution remote sense image is denoted as I'gk；And by IggAs the first high spectral resolution remote sensing image, it is denoted as I'gg；
Step 1_2: willAs reference, to I'gkRadiation registration is carried out to weaken I'gkWith I'ggBetween radiation difference,
Using the high spatial resolution remote sense image obtained after radiation registration as the second high spatial resolution remote sense image, it is denoted as I* gk；And
By I'ggAs the second high spectral resolution remote sensing image, it is denoted as I* gg。
In the step 1_1, geometrical registration utilizes professional remote sensing image processing software ENVI/ERDAS.
In the step 1_2, using match by moment technology to I'gkRadiation registration is carried out,Wherein, Pgk,stdIndicate I'gkIn all pixels point pixel value standard
Difference, Pgk,meanIndicate I'gkIn all pixels point pixel value mean value,It indicatesLuminance component in all pictures
The standard deviation of the pixel value of vegetarian refreshments,It indicatesLuminance component in all pixels point pixel value mean value.
DescribedLuminance component acquisition process are as follows: it is rightAll wave bands take mean value to obtainBrightness point
Amount.
In the step 2, using Denoising Algorithm to I* gkCarry out the process of quick denoising are as follows: for homogeneity remote sensing
Visual fusion, using Wiener filtering technology as Denoising Algorithm to I* gkCarry out quick denoising；Heterogeneous remote sensing image is melted
It closes, using Lee filtering technique as Denoising Algorithm to I* gkCarry out quick denoising.
The detailed process of the step 4 are as follows:
Step 4_1: to I* ggAll wave bands take mean value, obtain mean value component；
Step 4_2: I is calculated* ggEach wave band average gradient, and calculate the average gradient of mean value component；
Step 4_3: by I* ggThe average gradient of each wave band and the average gradient of mean value component be divided by, obtain I* ggPair
It answers required for wave bandHigh-frequency information injection weight；For I* ggQ-th of wave band, by I* ggQ-th of wave band it is flat
The result that the average gradient of equal gradient and mean value component is divided by is as I* ggQ-th of wave band required forHigh frequency letter
The injection weight of breath；Wherein, q is positive integer, and the initial value of q is 1, and 1≤q≤Q, Q indicate I* ggWave band number.
The detailed process of the step 6 are as follows:
Step 6_1: to IrhFuzzy Processing is carried out, the fusion evaluation obtained after Fuzzy Processing is denoted as I'rh；Wherein, it obscures
Ambiguity function used by handling selects image sensor modulation /demodulation function MTF；
Step 6_2: to I'rhDown-sampled processing is carried out, so that the size of the fusion evaluation obtained after down-sampled processing
With I* ggSize it is consistent, the fusion evaluation obtained after down-sampled processing is denoted as I "rh；
Step 6_3: by I "rhWith I* ggSubtract each other to obtain residual component；
Step 6_4: carrying out resampling processing to residual component, so that the size of the residual component obtained after resampling processing
Size and IrhSize it is consistent；
Step 6_5: the residual component obtained after resampling is handled is added to IrhIn, it obtains the sharpening of optimal spatial height and melts
Group photo picture.
Compared with the prior art, the advantages of the present invention are as follows:
1) the method for the present invention fully takes into account the influence of noise and heterogeneous Remote Sensing Image Fusion in remote sensing image and often produces
The phenomenon that third contact of a total solar or lunar eclipse blending algorithm, introduce fusion remote sensing image spectrally compensating strategy, not only can satisfy homogeneity it is panchromatic/multispectral remote sensing
Visual fusion, panchromatic/target in hyperspectral remotely sensed image fusion, multispectral/target in hyperspectral remotely sensed image fusion, and it is more to can satisfy SAR/
The fusion demand of the heterogeneous remote sensing image such as spectrum.
2) the method for the present invention considers space enhancing and spectrally compensating simultaneously, can promote high spectral resolution remote sensing image
Spatial resolution, the blending image for obtaining having spatial altitude to sharpen meets fusion evaluation vision interpretation demand, while last
Obtained optimal spatial height sharpens fusion evaluation spectrum fidelity with higher, can meet quantification application need well
It asks.
3) the method for the present invention is simple and efficient, and can satisfy the rapid fusion process demand of practical engineering application.
Detailed description of the invention
Fig. 1 is the overall procedure block diagram of the method for the present invention；
Fig. 2 a is a panchromatic remote sensing image；
Fig. 2 b is that a width and the multi-spectrum remote sensing image of Fig. 2 a Same Scene obtain after the step 3_1 of the method for the present invention processing
Multi-spectrum remote sensing image after the Design Based on Spatial Resampling arrived；
Fig. 2 c is high using optimal spatial of the method for the present invention to Fig. 2 a and Fig. 2 b the spectrum high-fidelity handled
Degree sharpens fusion evaluation；
Fig. 3 a is a panchromatic remote sensing image；
Fig. 3 b is that a width and the target in hyperspectral remotely sensed image of Fig. 3 a Same Scene obtain after the step 3_1 of the method for the present invention processing
Target in hyperspectral remotely sensed image after the Design Based on Spatial Resampling arrived；
Fig. 3 c is high using optimal spatial of the method for the present invention to Fig. 3 a and Fig. 3 b the spectrum high-fidelity handled
Degree sharpens fusion evaluation；
Fig. 4 a is a width SAR remote sensing image；
Fig. 4 b is that a width and the multi-spectrum remote sensing image of Fig. 4 a Same Scene obtain after the step 3_1 of the method for the present invention processing
Multi-spectrum remote sensing image after the Design Based on Spatial Resampling arrived；
Fig. 4 c is high using optimal spatial of the method for the present invention to Fig. 4 a and Fig. 4 b the spectrum high-fidelity handled
Degree sharpens fusion evaluation.
Specific embodiment
The present invention will be described in further detail below with reference to the embodiments of the drawings.
Fully take into account homogeneity image sky-spectrum high-fidelity fusion, heterogeneous image sky-spectrum high-fidelity fusion demand, Yi Jishi
Border engineer application is to remote sensing image rapid fusion process demand, and the invention proposes a kind of same/heterogeneous remote sensing image high-fidelity is wide
Adopted sky-spectrum fusion method not only can produce the best fusion evaluation of spatial altitude sharpening, while the spectrum of best fusion evaluation
Fidelity is high, and can meet homogeneity Remote Sensing Image Fusion and heterogeneous Remote Sensing Image Fusion simultaneously.
A kind of same/heterogeneous remote sensing image high-fidelity broad sense sky-spectrum fusion method proposed by the present invention, overall procedure block diagram
As shown in Figure 1, itself the following steps are included:
Step 1: choosing the multi-source Remote Sensing Images for being directed to Same Scene, respectively original high spatial resolution remote sense image
With original high spectral resolution remote sensing image, correspondence is denoted as IgkAnd Igg；Then to IgkAnd IggIt is pre-processed, will be pre-processed
High spatial resolution remote sense image and high spectral resolution remote sensing the image correspondence obtained afterwards is denoted as I* gkAnd I* gg。
In this particular embodiment, in step 1, I* gkAnd I* ggAcquisition process are as follows:
Step 1_1: according to IgkAnd IggBetween space resolution rate, to IggDesign Based on Spatial Resampling is carried out, space is adopted again
The high spectral resolution remote sensing image obtained after sample, is denoted as Size and IgkSize it is consistent；Then willAs reference, to IgkGeometrical registration is carried out, the high spatial resolution remote sense image obtained after geometrical registration is high as first
Spatial resolution remote sensing image, is denoted as I'gk；And by IggAs the first high spectral resolution remote sensing image, it is denoted as I'gg。
Step 1_2: willAs reference, to I'gkRadiation registration is carried out to weaken I'gkWith I'ggBetween radiation difference,
Using the high spatial resolution remote sense image obtained after radiation registration as the second high spatial resolution remote sense image, it is denoted as I* gk；And
By I'ggAs the second high spectral resolution remote sensing image, it is denoted as I* gg。
In this particular embodiment, in step 1_1, geometrical registration utilizes professional remote sensing image processing software ENVI/
ERDAS。
In this particular embodiment, in step 1_2, it is contemplated that high spatial resolution remote sense image and high spectral resolution are distant
Feeling the image capturing time, there may be differences, and especially heterogeneous Remote Sensing Image Fusion is relatively conventional, therefore use match by moment technology
To I'gkRadiation registration is carried out,Wherein, Pgk,stdIndicate I'gkIn all pictures
The standard deviation of the pixel value of vegetarian refreshments, Pgk,meanIndicate I'gkIn all pixels point pixel value mean value,It indicates's
The standard deviation of the pixel value of all pixels point in luminance component,It indicatesLuminance component in all pixels points
The mean value of pixel value.
In this particular embodiment,Luminance component acquisition process are as follows: it is rightAll wave bands take mean value to obtainLuminance component.
Step 2: judging I* gkIn whether include noise, use Denoising Algorithm to I if comprising noise* gkQuickly denoised
Processing, the high spatial resolution remote sense image obtained after denoising is denoted asDirectly by I if not including noise* gkWeight
Newly it is denoted as
In this particular embodiment, in step 2, using Denoising Algorithm to I*gkCarry out the process of quick denoising are as follows: right
In homogeneity Remote Sensing Image Fusion, such as IgkFor panchromatic image, and IggWhen for multispectral image or Hyperspectral imaging, filtered using wiener
Wave technology is as Denoising Algorithm to I* gkCarry out quick denoising；For heterogeneous Remote Sensing Image Fusion, such as IgkFor SAR image, and
IggWhen for multispectral image or Hyperspectral imaging, using Lee filtering technique as Denoising Algorithm to I* gkIt carries out at quick denoising
Reason.
Step 3: extractingHigh-frequency information, detailed process are as follows:
Step 3_1: in IgkSpectral region and IggSpectral region known in situation, according to IgkSpectral region and
IggSpectral region, select by IgkSpectral region covering IggWave band i.e. select by IgkSpectral region covering's
Wave band；Then basisSize, to I* ggCarry out Design Based on Spatial Resampling, the high spectral resolution that will be obtained after Design Based on Spatial Resampling
Rate remote sensing image is denoted asThen according to I* gg, and by selectionWave band carry out linear combination obtainBrightness
Component.
Or in IgkSpectral region and IggSpectral region it is unknown in the case where, according toSize, to I* ggInto
The high spectral resolution remote sensing image obtained after Design Based on Spatial Resampling is denoted as by row Design Based on Spatial ResamplingThen according to I* gg, and lead to
It is right to crossAll wave bands carry out linear combination obtainLuminance component.
It is above-mentioned, it is rightAll wave bands carry out linear combination used by combination coefficient acquisition process are as follows: for response shadow
It is right as quick process demandSpace desampling is carried out, the high spatial resolution remote sense image obtained after space desampling is remembered
ForIn IgkSpectral region and IggSpectral region known in situation, according to IgkSpectral region and IggSpectrum model
It encloses, selects by IgkSpectral region covering IggWave band i.e. select by IgkSpectral region coveringWave band, use
Least square method pairWith the I of selection* ggWave band handled, solution obtain combination coefficient；Or in IgkSpectral region
WithSpectral region it is unknown in the case where, using least square method pairAnd I* ggAll wave bands handled, solve
To combination coefficient.
Step 3_2: willLuminance component as reference, it is rightMatch by moment is carried out to weakenLuminance component with
Between radiation difference, the high spatial resolution remote sense image obtained after match by moment is denoted as
Step 3_3: willWithLuminance component subtract each other, using obtained error image asThe high frequency division in basis
Amount；Then pass through laplacian spectral radius filter pairBasic high fdrequency component carry out space enhancing, obtainTunable high-frequency
Component；Again willBasic high fdrequency component andTunable high-frequency component weighted array obtainHigh-frequency information；Wherein,
Tunable high-frequency component weight value range be [0,5], in actual treatmentTunable high-frequency component weight according to
Demand is manually set.
Step 4: utilizing I* ggEach wave band gradient, calculate I* ggEach wave band required forHigh-frequency information
Inject weight.
In this particular embodiment, the detailed process of step 4 are as follows:
Step 4_1: to I* ggAll wave bands take mean value, obtain mean value component.
Step 4_2: I is calculated* ggEach wave band average gradient, and calculate the average gradient of mean value component.
Step 4_3: by I* ggThe average gradient of each wave band and the average gradient of mean value component be divided by, obtain I* ggPair
It answers required for wave bandHigh-frequency information injection weight；For I* ggQ-th of wave band, by I* ggQ-th of wave band it is flat
The result that the average gradient of equal gradient and mean value component is divided by is as I* ggQ-th of wave band required forHigh frequency letter
The injection weight of breath；Wherein, q is positive integer, and the initial value of q is 1, and 1≤q≤Q, Q indicate I* ggWave band number.
Step 5: by I* ggEach wave band required forHigh-frequency information injection weight withHigh-frequency information phase
Multiply；Then obtained result is injected intoIn, there is the preliminary sky of high spatial resolution and high spectral resolution simultaneously
Between highly sharpen fusion evaluation, be denoted as Irh。
Step 6: utilizing I* ggTo IrhSpectrally compensating and correction are carried out, the optimal spatial height for obtaining spectrum high-fidelity sharpens
Fusion evaluation.
In this particular embodiment, the detailed process of step 6 are as follows:
Step 6_1: to IrhFuzzy Processing is carried out, the fusion evaluation obtained after Fuzzy Processing is denoted as I'rh；Wherein, it obscures
Ambiguity function used by handling selects image sensor modulation /demodulation function MTF.
Step 6_2: to I'rhDown-sampled processing is carried out, so that the size of the fusion evaluation obtained after down-sampled processing
With I* ggSize it is consistent, the fusion evaluation obtained after down-sampled processing is denoted as I "rh。
Step 6_3: by I "rhWith I* ggSubtract each other to obtain residual component.
Step 6_4: carrying out resampling processing to residual component, so that the size of the residual component obtained after resampling processing
Size and IrhSize it is consistent.
Step 6_5: the residual component obtained after resampling is handled is added to IrhIn further to promote fusion evaluation
Spectrum fidelity obtains optimal spatial height and sharpens fusion evaluation.
For the feasibility and validity for verifying the method for the present invention, the method for the present invention is tested.
Fig. 2 a gives a panchromatic remote sensing image, and Fig. 2 b gives the multispectral remote sensing shadow of a width Yu Fig. 2 a Same Scene
Multi-spectrum remote sensing image after the Design Based on Spatial Resampling that picture obtains after the step 3_1 of the method for the present invention processing, Fig. 2 c give benefit
Fusion evaluation is sharpened with optimal spatial height of the method for the present invention to Fig. 2 a and Fig. 2 b the spectrum high-fidelity handled.Figure
3a gives a panchromatic remote sensing image, and Fig. 3 b gives the target in hyperspectral remotely sensed image of a width and Fig. 3 a Same Scene through the present invention
Target in hyperspectral remotely sensed image after the Design Based on Spatial Resampling obtained after the step 3_1 processing of method, Fig. 3 c, which gives, utilizes present invention side
Method sharpens fusion evaluation to the optimal spatial height of Fig. 3 a and Fig. 3 b the spectrum high-fidelity handled.Fig. 4 a gives one
Width SAR remote sensing image, Fig. 4 b give the step of multi-spectrum remote sensing image of a width and Fig. 4 a Same Scene is through the method for the present invention
Multi-spectrum remote sensing image after 3_1 processing after obtained Design Based on Spatial Resampling, Fig. 4 c give using the method for the present invention to Fig. 4 a and
The optimal spatial height for the spectrum high-fidelity that Fig. 4 b is handled sharpens fusion evaluation.Fig. 2 c, Fig. 3 c and Fig. 4 c are observed, it can
To find out either panchromatic/multi-spectrum remote sensing image fusion, panchromatic/target in hyperspectral remotely sensed image fusion, SAR/ multispectral remote sensing shadow
As fusion, fusion results all have the spatial structural form more sharpened, while the spectral color of fusion results and original height
Spectral resolution remote sensing image consistency with higher, spectrum retention property are preferable.
Claims (8)
1. a kind of same/heterogeneous remote sensing image high-fidelity broad sense sky-spectrum fusion method, it is characterised in that the following steps are included:
Step 1: choosing the multi-source Remote Sensing Images for being directed to Same Scene, respectively original high spatial resolution remote sense image and original
The high spectral resolution remote sensing image of beginning, correspondence are denoted as IgkAnd Igg；Then to IgkAnd IggIt is pre-processed, will be obtained after pretreatment
To high spatial resolution remote sense image and high spectral resolution remote sensing image correspondence be denoted as I* gkAnd I* gg；
Step 2: judging I* gkIn whether include noise, use Denoising Algorithm to I if comprising noise* gkIt carries out at quick denoising
Reason, the high spatial resolution remote sense image obtained after denoising is denoted asDirectly by I if not including noise* gkAgain
It is denoted as
Step 3: extractingHigh-frequency information, detailed process are as follows:
Step 3_1: in IgkSpectral region and IggSpectral region known in situation, according to IgkSpectral region and Igg's
Spectral region is selected by IgkSpectral region covering IggWave band i.e. select by IgkSpectral region coveringWave
Section；Then basisSize, to I* ggCarry out Design Based on Spatial Resampling, the high spectral resolution that will be obtained after Design Based on Spatial Resampling
Remote sensing image is denoted asThen according to I* gg, and by selectionWave band carry out linear combination obtainBrightness
Component；
Or in IgkSpectral region and IggSpectral region it is unknown in the case where, according toSize, to I* ggCarry out space
The high spectral resolution remote sensing image obtained after Design Based on Spatial Resampling is denoted as by resamplingThen according to I* gg, and by pairAll wave bands carry out linear combination obtainLuminance component；
It is above-mentioned, it is rightAll wave bands carry out linear combination used by combination coefficient acquisition process are as follows: it is rightCarry out space
It is down-sampled, the high spatial resolution remote sense image obtained after space desampling is denoted asIn IgkSpectral region and IggLight
In situation known to spectral limit, according to IgkSpectral region and IggSpectral region, select by IgkSpectral region covering Igg
Wave band i.e. select by IgkSpectral region coveringWave band, using least square method pairWith the I of selection* gg's
Wave band is handled, and solution obtains combination coefficient；Or in IgkSpectral region and IggSpectral region it is unknown in the case where, adopt
With least square method pairAnd I* ggAll wave bands handled, solution obtain combination coefficient；
Step 3_2: willLuminance component as reference, it is rightMatch by moment is carried out to weakenLuminance component withBetween
Radiation difference, the high spatial resolution remote sense image obtained after match by moment is denoted as
Step 3_3: willWithLuminance component subtract each other, using obtained error image asBasic high fdrequency component；Then
Pass through laplacian spectral radius filter pairBasic high fdrequency component carry out space enhancing, obtainTunable high-frequency component；Again
It willBasic high fdrequency component andTunable high-frequency component weighted array obtainHigh-frequency information；Wherein,It is adjustable
The value range of the weight of high fdrequency component is [0,5]；
Step 4: utilizing I* ggEach wave band gradient, calculate I* ggEach wave band required forHigh-frequency information injection
Weight；
Step 5: by I* ggEach wave band required forHigh-frequency information injection weight withHigh-frequency information be multiplied；So
Obtained result is injected into afterwardsIn, had the preliminary space of high spatial resolution and high spectral resolution high simultaneously
Degree sharpens fusion evaluation, is denoted as Irh；
Step 6: utilizing I* ggTo IrhSpectrally compensating and correction are carried out, the optimal spatial height for obtaining spectrum high-fidelity sharpens fusion
Image.
2. a kind of same/heterogeneous remote sensing image high-fidelity broad sense sky-spectrum fusion method according to claim 1, feature exist
In the step 1, I* gkAnd I* ggAcquisition process are as follows:
Step 1_1: according to IgkAnd IggBetween space resolution rate, to IggDesign Based on Spatial Resampling is carried out, after Design Based on Spatial Resampling
Obtained high spectral resolution remote sensing image, is denoted asSize and IgkSize it is consistent；Then will
As reference, to IgkGeometrical registration is carried out, using the high spatial resolution remote sense image obtained after geometrical registration as the first high-altitude
Between resolution remote sense image, be denoted as I'gk；And by IggAs the first high spectral resolution remote sensing image, it is denoted as I'gg；
Step 1_2: willAs reference, to I'gkRadiation registration is carried out to weaken I'gkWith I'ggBetween radiation difference, by spoke
The high spatial resolution remote sense image obtained after registration is penetrated as the second high spatial resolution remote sense image, is denoted as I* gk；And it will
I'ggAs the second high spectral resolution remote sensing image, it is denoted as I* gg。
3. a kind of same/heterogeneous remote sensing image high-fidelity broad sense sky-spectrum fusion method according to claim 2, feature exist
In the step 1_1, geometrical registration utilizes professional remote sensing image processing software ENVI/ERDAS.
4. a kind of same/heterogeneous remote sensing image high-fidelity broad sense sky-spectrum fusion method according to claim 2 or 3, feature
It is in the step 1_2, using match by moment technology to I'gkRadiation registration is carried out,Wherein, Pgk,stdIndicate I'gkIn all pixels point pixel value standard
Difference, Pgk,meanIndicate I'gkIn all pixels point pixel value mean value,It indicatesLuminance component in all pictures
The standard deviation of the pixel value of vegetarian refreshments,It indicatesLuminance component in all pixels point pixel value mean value.
5. a kind of same/heterogeneous remote sensing image high-fidelity broad sense sky-spectrum fusion method according to claim 4, feature exist
In describedLuminance component acquisition process are as follows: it is rightAll wave bands take mean value to obtainLuminance component.
6. a kind of same/heterogeneous remote sensing image high-fidelity broad sense sky-spectrum fusion method according to claim 1, feature exist
In the step 2, using Denoising Algorithm to I* gkCarry out the process of quick denoising are as follows: homogeneity remote sensing image is melted
It closes, using Wiener filtering technology as Denoising Algorithm to I* gkCarry out quick denoising；For heterogeneous Remote Sensing Image Fusion, adopt
Use Lee filtering technique as Denoising Algorithm to I* gkCarry out quick denoising.
7. a kind of same/heterogeneous remote sensing image high-fidelity broad sense sky-spectrum fusion method according to claim 1, feature exist
In the detailed process of the step 4 are as follows:
Step 4_1: to I* ggAll wave bands take mean value, obtain mean value component；
Step 4_2: I is calculated* ggEach wave band average gradient, and calculate the average gradient of mean value component；
Step 4_3: by I* ggThe average gradient of each wave band and the average gradient of mean value component be divided by, obtain I* ggCorrespondence wave
Required for sectionHigh-frequency information injection weight；For I* ggQ-th of wave band, by I* ggQ-th of wave band average ladder
The result that the average gradient of degree and mean value component is divided by is as I* ggQ-th of wave band required forHigh-frequency information
Inject weight；Wherein, q is positive integer, and the initial value of q is 1, and 1≤q≤Q, Q indicate I* ggWave band number.
8. a kind of same/heterogeneous remote sensing image high-fidelity broad sense sky-spectrum fusion method according to claim 1, feature exist
In the detailed process of the step 6 are as follows:
Step 6_1: to IrhFuzzy Processing is carried out, the fusion evaluation obtained after Fuzzy Processing is denoted as I'rh；Wherein, Fuzzy Processing
Used ambiguity function selects image sensor modulation /demodulation function MTF；
Step 6_2: to I'rhDown-sampled processing is carried out, so that the size and I of the fusion evaluation obtained after down-sampled processing* gg
Size it is consistent, the fusion evaluation obtained after down-sampled processing is denoted as I "rh；
Step 6_3: by I "rhWith I* ggSubtract each other to obtain residual component；
Step 6_4: carrying out resampling processing to residual component, so that the size of the residual component obtained after resampling processing
With IrhSize it is consistent；
Step 6_5: the residual component obtained after resampling is handled is added to IrhIn, it obtains optimal spatial height and sharpens fusion shadow
Picture.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910619429.1A CN110533600B (en) | 2019-07-10 | 2019-07-10 | Same/heterogeneous remote sensing image high-fidelity generalized space-spectrum fusion method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910619429.1A CN110533600B (en) | 2019-07-10 | 2019-07-10 | Same/heterogeneous remote sensing image high-fidelity generalized space-spectrum fusion method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110533600A true CN110533600A (en) | 2019-12-03 |
| CN110533600B CN110533600B (en) | 2022-07-19 |
Family
ID=68659619
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910619429.1A Active CN110533600B (en) | 2019-07-10 | 2019-07-10 | Same/heterogeneous remote sensing image high-fidelity generalized space-spectrum fusion method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110533600B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111091113A (en) * | 2019-12-30 | 2020-05-01 | 贵阳欧比特宇航科技有限公司 | Hyperspectral image data fusion method |
| CN111145132A (en) * | 2019-12-05 | 2020-05-12 | 东南大学 | Spectrum-keeping type rapid remote sensing image fusion method |
| CN113436123A (en) * | 2021-06-22 | 2021-09-24 | 宁波大学 | High-resolution SAR and low-resolution multispectral image fusion method based on cloud removal-resolution improvement cooperation |
| CN117253125A (en) * | 2023-10-07 | 2023-12-19 | 珠江水利委员会珠江水利科学研究院 | Space-spectrum mutual injection image fusion method, system and readable storage medium |
| CN117274763A (en) * | 2023-11-21 | 2023-12-22 | 珠江水利委员会珠江水利科学研究院 | Remote sensing image space-spectrum fusion method, system, equipment and medium based on balance point analysis |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102446351A (en) * | 2010-10-15 | 2012-05-09 | 江南大学 | Multispectral and high-resolution full-color image fusion method study |
| CN103236047A (en) * | 2013-03-28 | 2013-08-07 | 北京航空航天大学 | Method for fusing full-color and multi-spectral images on basis of fitting for substituted components |
| CN103942769A (en) * | 2013-12-10 | 2014-07-23 | 珠江水利委员会珠江水利科学研究院 | Satellite remote sensing image fusion method |
| CN112166692B (en) * | 2012-06-29 | 2014-07-30 | 二十一世纪空间技术应用股份有限公司 | Remote sensing image cross fusion method |
| CN104851091A (en) * | 2015-04-28 | 2015-08-19 | 中山大学 | Remote sensing image fusion method based on convolution enhancement and HCS transform |
| CN106651800A (en) * | 2016-12-23 | 2017-05-10 | 中国科学院遥感与数字地球研究所 | PAN modulation and multiple linear regression-based MS and PAN image fusion method |
| CN108765359A (en) * | 2018-05-31 | 2018-11-06 | 安徽大学 | Fusion method of hyperspectral remote sensing image and full-color image based on JSK model and NSCT technology |
| CN109166089A (en) * | 2018-07-24 | 2019-01-08 | 重庆三峡学院 | The method that a kind of pair of multispectral image and full-colour image are merged |
| CN109509160A (en) * | 2018-11-28 | 2019-03-22 | 长沙理工大学 | Hierarchical remote sensing image fusion method utilizing layer-by-layer iteration super-resolution |
| CN109886870A (en) * | 2018-12-29 | 2019-06-14 | 西北大学 | Remote sensing image fusion method based on binary channels neural network |
-
2019
- 2019-07-10 CN CN201910619429.1A patent/CN110533600B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102446351A (en) * | 2010-10-15 | 2012-05-09 | 江南大学 | Multispectral and high-resolution full-color image fusion method study |
| CN112166692B (en) * | 2012-06-29 | 2014-07-30 | 二十一世纪空间技术应用股份有限公司 | Remote sensing image cross fusion method |
| CN103236047A (en) * | 2013-03-28 | 2013-08-07 | 北京航空航天大学 | Method for fusing full-color and multi-spectral images on basis of fitting for substituted components |
| CN103942769A (en) * | 2013-12-10 | 2014-07-23 | 珠江水利委员会珠江水利科学研究院 | Satellite remote sensing image fusion method |
| CN104851091A (en) * | 2015-04-28 | 2015-08-19 | 中山大学 | Remote sensing image fusion method based on convolution enhancement and HCS transform |
| CN106651800A (en) * | 2016-12-23 | 2017-05-10 | 中国科学院遥感与数字地球研究所 | PAN modulation and multiple linear regression-based MS and PAN image fusion method |
| CN108765359A (en) * | 2018-05-31 | 2018-11-06 | 安徽大学 | Fusion method of hyperspectral remote sensing image and full-color image based on JSK model and NSCT technology |
| CN109166089A (en) * | 2018-07-24 | 2019-01-08 | 重庆三峡学院 | The method that a kind of pair of multispectral image and full-colour image are merged |
| CN109509160A (en) * | 2018-11-28 | 2019-03-22 | 长沙理工大学 | Hierarchical remote sensing image fusion method utilizing layer-by-layer iteration super-resolution |
| CN109886870A (en) * | 2018-12-29 | 2019-06-14 | 西北大学 | Remote sensing image fusion method based on binary channels neural network |
Non-Patent Citations (3)
| Title |
|---|
| SAJJAD EGHBALIAN等: "Multi spectral image fusion by deep convolutional neural network", 《INTERNATIONAL JOURNAL OF REMOTE SENSING》 * |
| 孟祥超: "多源时—空—谱光学遥感影像的变分融合方法", 《中国优秀博硕士学位论文全文数据库(博士)基础科学辑》 * |
| 朱吉: "顾及相机抖动与运动目标的多重曝光影像融合", 《中国优秀博硕士学位论文全文数据库(硕士)信息科技辑》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111145132A (en) * | 2019-12-05 | 2020-05-12 | 东南大学 | Spectrum-keeping type rapid remote sensing image fusion method |
| CN111145132B (en) * | 2019-12-05 | 2022-11-18 | 东南大学 | Spectrum-preserving type rapid remote sensing image fusion method |
| CN111091113A (en) * | 2019-12-30 | 2020-05-01 | 贵阳欧比特宇航科技有限公司 | Hyperspectral image data fusion method |
| CN113436123A (en) * | 2021-06-22 | 2021-09-24 | 宁波大学 | High-resolution SAR and low-resolution multispectral image fusion method based on cloud removal-resolution improvement cooperation |
| CN117253125A (en) * | 2023-10-07 | 2023-12-19 | 珠江水利委员会珠江水利科学研究院 | Space-spectrum mutual injection image fusion method, system and readable storage medium |
| CN117253125B (en) * | 2023-10-07 | 2024-03-22 | 珠江水利委员会珠江水利科学研究院 | Space-spectrum mutual injection image fusion method, system and readable storage medium |
| CN117274763A (en) * | 2023-11-21 | 2023-12-22 | 珠江水利委员会珠江水利科学研究院 | Remote sensing image space-spectrum fusion method, system, equipment and medium based on balance point analysis |
| CN117274763B (en) * | 2023-11-21 | 2024-04-05 | 珠江水利委员会珠江水利科学研究院 | Remote sensing image space-spectrum fusion method, system, equipment and medium based on balance point analysis |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110533600B (en) | 2022-07-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110533600A (en) | A kind of same/heterogeneous remote sensing image high-fidelity broad sense sky-spectrum fusion method | |
| CN111080724B (en) | Fusion method of infrared light and visible light | |
| Choi et al. | A new adaptive component-substitution-based satellite image fusion by using partial replacement | |
| Klonus et al. | Image fusion using the Ehlers spectral characteristics preservation algorithm | |
| CN107563964B (en) | Rapid splicing method for large-area-array sub-meter-level night scene remote sensing images | |
| Zhang | Understanding image fusion | |
| CN104683767B (en) | Penetrating Fog image generating method and device | |
| US8761506B1 (en) | Pan sharpening digital imagery | |
| Bungert et al. | Blind image fusion for hyperspectral imaging with the directional total variation | |
| Teke | Satellite image processing workflow for RASAT and Göktürk-2 | |
| Song et al. | An adaptive pansharpening method by using weighted least squares filter | |
| CN112184604B (en) | Color image enhancement method based on image fusion | |
| CN111340895B (en) | Image color uniformizing method based on pyramid multi-scale fusion | |
| CN103049898A (en) | Method for fusing multispectral and full-color images with light cloud | |
| Seo et al. | UPSNet: Unsupervised pan-sharpening network with registration learning between panchromatic and multi-spectral images | |
| CN112307901A (en) | Landslide detection-oriented SAR and optical image fusion method and system | |
| WO2015057425A1 (en) | Converting an image from a dual-band sensor to a visible color image | |
| CN108961325A (en) | Method for registering between more/high-spectrum remote sensing wave band | |
| Rao et al. | Spatiotemporal data fusion using temporal high-pass modulation and edge primitives | |
| Qian et al. | Enhancing spatial resolution of hyperspectral imagery using sensor's intrinsic keystone distortion | |
| CN112669232A (en) | Depth image enhancement processing method and device | |
| Zhong et al. | Attention_FPNet: Two-branch remote sensing image pansharpening network based on attention feature fusion | |
| CN107507151B (en) | Multispectral remote sensing image real color restoration method and system | |
| Gehrke et al. | Radiometric normalization of large airborne image data sets acquired by different sensor types | |
| KR20210096925A (en) | Flexible Color Correction Method for Massive Aerial Orthoimages |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |