CN110148103A - EO-1 hyperion and Multispectral Image Fusion Methods, computer readable storage medium, electronic equipment based on combined optimization - Google Patents

EO-1 hyperion and Multispectral Image Fusion Methods, computer readable storage medium, electronic equipment based on combined optimization Download PDF

Info

Publication number
CN110148103A
CN110148103A CN201910355496.7A CN201910355496A CN110148103A CN 110148103 A CN110148103 A CN 110148103A CN 201910355496 A CN201910355496 A CN 201910355496A CN 110148103 A CN110148103 A CN 110148103A
Authority
CN
China
Prior art keywords
image
matrix
spectrum
spatial resolution
abundance
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
Application number
CN201910355496.7A
Other languages
Chinese (zh)
Other versions
CN110148103B (en
Inventor
郑向涛
陈文静
卢孝强
刘康
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
XiAn Institute of Optics and Precision Mechanics of CAS
Original Assignee
XiAn Institute of Optics and Precision Mechanics of CAS
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by XiAn Institute of Optics and Precision Mechanics of CAS filed Critical XiAn Institute of Optics and Precision Mechanics of CAS
Priority to CN201910355496.7A priority Critical patent/CN110148103B/en
Publication of CN110148103A publication Critical patent/CN110148103A/en
Application granted granted Critical
Publication of CN110148103B publication Critical patent/CN110148103B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00Image enhancement or restoration
    • G06T5/50Image enhancement or restoration using two or more images, e.g. averaging or subtraction
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/10Image acquisition modality
    • G06T2207/10032Satellite or aerial image; Remote sensing
    • G06T2207/10036Multispectral image; Hyperspectral image
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/20Special algorithmic details
    • G06T2207/20212Image combination
    • G06T2207/20221Image fusion; Image merging

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Image Processing (AREA)

Abstract

The present invention provides a kind of EO-1 hyperion based on combined optimization and Multispectral Image Fusion Methods, computer readable storage medium, electronic equipment, it solves existing EO-1 hyperion and Multispectral Image Fusion Methods relies on space degenerate matrix, the problem for causing high spectrum image spatial resolution lower.Method includes the following steps: step 1, input true value;Step 2 pre-processes true value, obtains observable low spatial resolution high spectrum image and observable high spatial resolution multi-spectral image;Step 3, according to line spectrum aliasing model, using Non-negative Matrix Factorization, to image IhWith image ImIt is mixed to carry out spectrum solution, end member matrix and abundance matrix needed for combined optimization obtain image IhEnd member matrix E, abundance matrix AhWith image ImAbundance matrix A;Step 4 estimates high spatial resolution high spectrum image, is denoted as image Z.

Description

It is EO-1 hyperion and Multispectral Image Fusion Methods based on combined optimization, computer-readable Storage medium, electronic equipment
Technical field
The present invention relates to technical field of image processing, and in particular to a kind of EO-1 hyperion and multispectral figure based on combined optimization As fusion method, computer readable storage medium, electronic equipment, can be applied to environmental monitoring, target detection, target classification and The fields such as military surveillance.
Background technique
Hyperspectral imager samples electromagnetic spectrum in many continuous and very narrow spectral band, acquisition High spectrum image has high spectral resolution.In order to obtain more wave bands, sensor has a light splitting before receiving luminous energy Light is divided into many parts by process, i.e. spectroscope, then each wave band only has small portion after light splitting under light projectile energy certain condition Energy is divided to reach sensor.Sensor obtains certain luminous energy and could respond, in order to ensure enough signal-to-noise ratio, it is necessary to increase chip picture Plain size.Pixel Dimensions refer to the area of each pixel, and when Pixel Dimensions increase, pixel quantity reduces in unit area, this meeting The image spatial resolution obtained is caused to reduce.Based on the above reasons, had by the high spectrum image that Hyperspectral imager obtains There is low spatial resolution.Due to various hardware limitations, existing technical staff proposes to improve high spectrum image using software approach Spatial resolution, i.e. EO-1 hyperion and Multispectral Image Fusion Methods, by with the multispectral image with high spatial resolution Fusion is to improve high spectrum image spatial resolution.
In recent years, researchers propose many EO-1 hyperions and Multispectral Image Fusion Methods.The purpose of these methods It is, by observable low spatial resolution-high spectral resolution high spectrum image and the low spectrum of observable high spatial resolution- Resolution multi-spectral image generates unobservable high spatial resolution-high spectral resolution high spectrum image.Such method is big More dependences and space degenerate matrix, space degenerate matrix contains high spectral resolution high spectrum image and low spatial resolution is more Space degeneration relationship between spectrum picture.Z.H.Nezhad et al. is in document " Z.H.Nezhad, A.Karami, R.Heylen and P. Scheunders,“Fusion of Hyperspectral and Multispectral Images Using Spectral Unmixing and Sparse Coding,”IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,vol.9,no.6,pp.2377-2389,2016.” It is middle to propose a kind of EO-1 hyperion and Multispectral Image Fusion Methods that sparse coding is mixed based on spectrum solution, the fusion process of this method An ill posed inverse problem, first with based on the regularization term that sparse coding constructs be converted into it is suitable determine inverse problem, then use Some high spatial resolution multi-spectral images or full-colour image and then one suitable dictionary of building, are based on the word in uncorrelated scene Allusion quotation and sparse coding is estimated by the initial high spatial resolution high spectrum image that line spectrum solution mixes model estimation, reused dilute It dredges coding and is used as regularization term, fixed inverse problem is fitted to calculate abundance by solution, is finally obtained from the abundance and end member obtained Required high spatial resolution high spectrum image.However, in practical applications, it is very tired accurately to estimate space degenerate matrix Difficulty, such method need estimation space degenerate matrix, and the evaluated error of generation can be propagated in fusion process, influence fusion method Performance.
In conclusion existing EO-1 hyperion and Multispectral Image Fusion Methods excessively rely on space degenerate matrix, lead to height Spectrum picture spatial resolution is lower, it is made to there is centainly restricted.
Summary of the invention
Space degenerate matrix is relied in order to solve existing EO-1 hyperion and Multispectral Image Fusion Methods, leads to high spectrum image The lower problem of spatial resolution, the invention proposes a kind of based on the EO-1 hyperion of combined optimization and Multispectral Image Fusion side Method, computer readable storage medium, electronic equipment are mainly used for improving the low problem of high spectrum image spatial resolution, to mention The spatial resolution of high high spectrum image.
The technical solution of the invention is as follows:
A kind of EO-1 hyperion and Multispectral Image Fusion Methods based on combined optimization, comprising the following steps:
Step 1, input true value, the true value refer to true high spatial resolution high spectrum image;
Step 2 pre-processes true value, obtains observable low spatial resolution high spectrum image and observable They are denoted as image I by high spatial resolution multi-spectral image respectivelyhWith image Im
Step 3, according to line spectrum aliasing model, using Non-negative Matrix Factorization, to image IhWith image ImCarry out spectrum solution It is mixed, end member matrix and abundance matrix needed for combined optimization;After reaching maximum number of iterations, image I is obtainedhEnd member matrix E, abundance matrix AhWith image ImAbundance matrix A;
Step 4, solution are mixed to rebuild, according to image IhEnd member matrix E and image ImAbundance matrix A, calculate high spatial point Resolution high spectrum image, i.e., final fusion results are denoted as image Z.
Further, the step 2 specifically:
Step 2.1) carries out space degeneration processing to true value, i.e., first carries out fuzzy operation to true value, then to gained knot Fruit carries out down-sampling operation, obtains observable low spatial resolution high spectrum image, is denoted as image Ih
Step 2.2) carries out Spectrum curve degradation processing to true value, that is, allows true value and spectral response matrix multiple, and obtaining can The high spatial resolution multi-spectral image of observation, is denoted as image Im
Further, the step 3 specifically:
Step 3.1) input picture Ih, image ImWith spectral response matrix Gm, image IhEnd member matrix and abundance matrix It is denoted as E and A respectivelyh, image ImAbundance matrix be denoted as A, according to line spectrum aliasing model Ih≈EAhAnd Im≈GmEA is right Image IhSpectrum solution, which is carried out, using Non-negative Matrix Factorization mixes available E and Ah, to image ImLight is carried out using Non-negative Matrix Factorization Spectrum solution mixes available A;
Step 3.2) initializes end member matrix E, abundance matrix AhWith abundance matrix A, the number of iterations k=0;
Step 3.3) updates end member matrix E, abundance matrix A simultaneouslyhWith abundance matrix A:
Wherein, α(k)It is the step-length of kth time iteration, L is loss function;
Wherein, parameter lambda=8, ‖ ‖FIndicate F norm,WithL is respectively indicated to E, AhPartial derivative is sought with A;
Wherein, ()TRepresenting matrix transposition;
Step 3.4) judges whether to reach maximum number of iterations, the end member matrix E that exports if reaching, abundance matrix AhWith with abundance matrix A, otherwise continue step 3.3).
Further, the step 4 specifically:
Step 4.1) is according to image IhEnd member matrix E and image ImAbundance matrix A, calculate high spatial resolution height Spectrum picture, i.e., final fusion results are denoted as image Z:
Z=EA.
Further, the step 4 further includes step 4.2):
Step 4.2) movement images Z and true value, Calculation Estimation index.
Meanwhile the present invention also provides a kind of computer readable storage mediums, are stored thereon with computer program, the calculating The step of above method is realized when machine program is executed by processor.
In addition, the present invention also provides a kind of electronic equipment, the electronic equipment includes:
Processor;
Computer readable storage medium, is stored thereon with computer program, and the computer program is transported by the processor The step of above method is executed when row.
Compared with the conventional method, the invention has the following advantages:
1. image interfusion method of the present invention does not need the priori knowledge of Spectrum curve degradation matrix, using the side based on combined optimization Method merges low spatial resolution high spectrum image and high spatial resolution multi-spectral image, according to line spectrum aliasing mould Type carries out spectrum to low spatial resolution high spectrum image and high spatial resolution multi-spectral image using Non-negative Matrix Factorization Solution is mixed, and end member matrix and abundance matrix needed for combined optimization finally carry out them to solve mixed reconstruction, and available quality is higher High spatial resolution high spectrum image, this method do not need the participation of Spectrum curve degradation matrix, overcomes conventional method to spectrum The dependence of degenerate matrix improves the spatial resolution of high spectrum image, has the advantages that syncretizing effect is good, complexity is low.
2. image interfusion method application range of the present invention is wider, can be applied to environmental monitoring, target detection, target classification with And the fields such as military surveillance.
Detailed description of the invention
Fig. 1 is that the present invention is based on the flow charts of the EO-1 hyperion of combined optimization and Multispectral Image Fusion Methods.
Specific embodiment
The contents of the present invention are described in further detail below in conjunction with the drawings and specific embodiments:
The method of the present invention is according to line spectrum aliasing model, using Non-negative Matrix Factorization, to low spatial resolution EO-1 hyperion Image and high spatial resolution multi-spectral image carry out spectrum solution and mix, end member matrix and abundance matrix needed for combined optimization, most They are carried out afterwards to solve mixed reconstruction, the available higher high spatial resolution high spectrum image of quality.The present invention and existing side Method is compared, and the participation of Spectrum curve degradation matrix is not needed, and overcomes dependence of the conventional method to Spectrum curve degradation matrix, has fusion effect The advantages of fruit is good, complexity is low and has wide range of applications.
As shown in Figure 1, the EO-1 hyperion and Multispectral Image Fusion Methods provided by the present invention based on combined optimization, including Following steps:
Step 1, input true value, true value refer to true high spatial resolution high spectrum image;
Step 2 pre-processes true value, obtains observable low spatial resolution high spectrum image and observable They are denoted as image I by high spatial resolution multi-spectral image respectivelyhWith image Im
Step 2.1) carries out space degeneration processing to true value, i.e., first carries out fuzzy operation to true value, then to gained knot Fruit carries out down-sampling operation, obtains observable low spatial resolution high spectrum image, is denoted as image Ih
Step 2.2) carries out Spectrum curve degradation processing to true value, that is, allows true value and spectral response matrix multiple, and obtaining can The high spatial resolution multi-spectral image of observation, is denoted as image Im
Step 3, according to line spectrum aliasing model, using Non-negative Matrix Factorization, to image IhWith image ImCarry out spectrum solution Mixed, end member matrix and abundance matrix needed for combined optimization obtain image I after reaching maximum number of iterationshEnd member matrix E, abundance matrix AhWith image ImAbundance matrix A;
Step 3.1) input picture Ih, image ImWith spectral response matrix Gm, image IhEnd member matrix and abundance matrix It is denoted as E and A respectivelyh, image ImAbundance matrix be denoted as A, according to line spectrum aliasing model Ih≈EAhAnd Im≈GmEA is right Image IhSpectrum solution, which is carried out, using Non-negative Matrix Factorization mixes available E and Ah, to image ImLight is carried out using Non-negative Matrix Factorization Spectrum solution mixes available A;
Step 3.2) initializes end member matrix E, abundance matrix AhWith abundance matrix A, the number of iterations k=0;
Step 3.3) updates end member matrix E, abundance matrix A simultaneouslyhWith abundance matrix A:
Wherein, α(k)It is the step-length of kth time iteration, its calculation method is in document " C.-J.Lin, " Projected gradient methods for nonnegative matrix factorization,”Neural Computation, Vol. 2779 19, no.10, pp.2756-, in 2007. " Algorithm 4, L is loss function,
Wherein, parameter lambda=8, ‖ ‖FIndicate F norm,WithL is respectively indicated to E, AhPartial derivative is sought with A,
Wherein, ()TRepresenting matrix transposition;
Step 3.4) judges whether to reach maximum number of iterations, the end member matrix E that exports if reaching, abundance matrix AhWith with abundance matrix A, otherwise continue step 3.3;
The mixed reconstruction of step 4, solution and Calculation Estimation index
Step 4.1) is according to image IhEnd member matrix E and image ImAbundance matrix A, estimate high spatial resolution height Spectrum picture, i.e., final fusion results are denoted as image Z:
Z=EA.
Step 4.2) movement images Z and true value, Calculation Estimation index.By comparing the similar of fusion results and true value Size is spent, to evaluate the performance of fusion method, i.e. the similarity of the fusion results and true value matter that shows fusion results more greatly Amount is better.
Effect of the invention is further explained below by way of specific emulation experiment.
1, simulated conditions
The method of the present invention be central processing unit be Intel (R) Core (TM) i7 5930k, memory 64GB, Ubuntu grasp Make in system, the emulation carried out with MATLAB software;
2, emulation content
For the experimental data used for Pavia University database, which includes 1 true high spectrum image, Picture material is Pavia University, and the spatial resolution of image is 1.3 meters.The size of image is 200 × 200 × 103, Wherein 200 × 200 be image space size, 103 be spectral band number.It will be original in Pavia University database High spectrum image as true value, low spatial resolution height of the image obtained again by fuzzy and down-sampling 4 as test Spectrum picture.
On Pavia University database, complete inventive algorithm (a kind of EO-1 hyperion based on combined optimization and Multispectral Image Fusion Methods) experiment.In order to prove the validity of algorithm, comprehensively considers the popular, newness of algorithm, select 4 kinds of control methods have been taken to be compared: Bicubic, FUSE, PALM and CO-CNMF.Wherein, Bicubic is classical benchmark side Method, by Zeyde et al. in document " R.Zeyde, M.Elad, and M.Protter, " On single image scale-up Usingsparse-representations, " the middle proposition of Curves and Surfaces, 2012, pages 711-730. "; FUSE is in document " Q.Wei, N.Dobigeon, and J.-Y.Tourneret, " Fast fusion of multi- bandimages based on solving a sylvester equation,”IEEE Transactions onImage It is proposed in Processing, vol.24, no.11, pp.4109-4121,2015. ";PALM document " C.Lanaras, E.Baltsavias,and K.Schindler,“Hyperspectralsuperresolution by coupled spectral unmixing,”in IEEE International Conference on Computer Vision,2015, Pp.3586-3594. it is proposed in ";CO-CNMF is in document " C.-H.Lin, F.Ma, C.-Y. Chi, and C.-H.Hsieh, " A convex optimization-based coupled nonnegative matrix factorization algorithm for hyperspectral and multispectral data fusion,”IEEE Transactions on It is proposed in Geoscienceand Remote Sensing, vol.56, no.3, pp.1652-1667,2018. ".
Fusion side is measured using PSNR, UIQI, RMSE, ERGAS and SAM of true value and super-resolution high spectrum image The performance of method.On Pavia University database, it is compared with 4 kinds of control methods, the results are shown in Table 1.
As seen from Table 1, fusion results of the invention are more preferable than existing fusion method.This is because existing method needs to estimate Space degenerate matrix, the error generated in estimation procedure can be transmitted in fusion process, this will affect the performance of fusion method, this Invention does not need the participation of space degenerate matrix, and alleviating the error generated during estimation space degenerate matrix can merge The problem of being transmitted in journey.Therefore than other methods, more robust more effectively further demonstrates of the invention advanced this method Property.
The comparing result of the different fusion methods of table 1
Bicubic FUSE PALM CO-CNMF The present invention
PSNR 25.885 32.032 33.708 32.105 36.329
UIQI 0.768 0.949 0.963 0.950 0.975
RMSE 13.261 6.906 5.367 6.629 4.260
ERGAS 8.086 4.256 3.473 3.955 3.049
SAM 6.271 4.840 4.038 4.280 3.650
The embodiment of the present invention also provides a kind of computer readable storage medium, and for storing program, program is performed reality Now the step of EO-1 hyperion and Multispectral Image Fusion Methods based on combined optimization.In some possible embodiments, this hair Bright various aspects are also implemented as a kind of form of program product comprising program code, when described program product is at end When running in end equipment, said program code is for executing the terminal device described in this specification above method part The step of various illustrative embodiments according to the present invention.
For realizing the program product of the above method, portable compact disc read only memory (CD-ROM) can be used And including program code, and can be run on terminal device, such as PC.However, program product of the invention is unlimited In this, in this document, readable storage medium storing program for executing can be any tangible medium for including or store program, which can be referred to Enable execution system, device or device use or in connection.
Program product can be using any combination of one or more readable mediums.Readable medium can be readable signal Jie Matter or readable storage medium storing program for executing.Readable storage medium storing program for executing for example can be but be not limited to electricity, magnetic, optical, electromagnetic, infrared ray or partly lead System, device or the device of body, or any above combination.More specific example (the non exhaustive column of readable storage medium storing program for executing Table) it include: the electrical connection with one or more conducting wires, portable disc, hard disk, random access memory (RAM), read-only storage Device (ROM), erasable programmable read only memory (EPROM or flash memory), optical fiber, portable compact disc read only memory (CD- ROM), light storage device, magnetic memory device or above-mentioned any appropriate combination.

Claims (7)

1. a kind of EO-1 hyperion and Multispectral Image Fusion Methods based on combined optimization, which comprises the following steps:
Step 1, input true value, the true value refer to true high spatial resolution high spectrum image;
Step 2 pre-processes true value, obtains observable low spatial resolution high spectrum image and observable high-altitude Between resolution multi-spectral image, they are denoted as to image I respectivelyhWith image Im
Step 3, according to line spectrum aliasing model, using Non-negative Matrix Factorization, to image IhWith image ImIt is mixed to carry out spectrum solution, End member matrix and abundance matrix needed for combined optimization;After reaching maximum number of iterations, image I is obtainedhEnd member matrix E, Abundance matrix AhWith image ImAbundance matrix A;
Step 4, solution are mixed to rebuild, according to image IhEnd member matrix E and image ImAbundance matrix A, calculate high spatial resolution High spectrum image, i.e., final fusion results are denoted as image Z.
2. the EO-1 hyperion and Multispectral Image Fusion Methods according to claim 1 based on combined optimization, which is characterized in that The step 2 specifically:
Step 2.1) to true value carry out space degeneration processing, i.e., first to true value carry out fuzzy operation, then to acquired results into The operation of row down-sampling, obtains observable low spatial resolution high spectrum image, is denoted as image Ih
Step 2.2) carries out Spectrum curve degradation processing to true value, that is, allows true value and spectral response matrix multiple, obtains Observable High spatial resolution multi-spectral image, be denoted as image Im
3. the EO-1 hyperion and Multispectral Image Fusion Methods according to claim 1 based on combined optimization, which is characterized in that The step 3 specifically:
Step 3.1) input picture Ih, image ImWith spectral response matrix Gm, image IhEnd member matrix and abundance matrix difference It is denoted as E and Ah, image ImAbundance matrix be denoted as A, according to line spectrum aliasing model Ih≈EAhAnd Im≈GmEA, to image IhSpectrum solution, which is carried out, using Non-negative Matrix Factorization mixes available E and Ah, to image ImSpectrum solution is carried out using Non-negative Matrix Factorization Mix available A;
Step 3.2) initializes end member matrix E, abundance matrix AhWith abundance matrix A, the number of iterations k=0;
Step 3.3) updates end member matrix E, abundance matrix A simultaneouslyhWith abundance matrix A:
Wherein, α(k)It is the step-length of kth time iteration, L is loss function;
Wherein, parameter lambda=8, ‖ ‖FIndicate F norm,WithL is respectively indicated to E, AhPartial derivative is sought with A;
Wherein, ()TRepresenting matrix transposition;
Step 3.4) judges whether to reach maximum number of iterations, end member matrix E, the abundance matrix A exported if reachinghWith With abundance matrix A, otherwise continue step 3.3).
4. the EO-1 hyperion and Multispectral Image Fusion Methods according to claim 1 or 2 or 3 based on combined optimization, special Sign is, the step 4 specifically:
Step 4.1) is according to image IhEnd member matrix E and image ImAbundance matrix A, calculate high spatial resolution EO-1 hyperion Image, i.e., final fusion results are denoted as image Z:
Z=EA.
5. the EO-1 hyperion and Multispectral Image Fusion Methods according to claim 4 based on combined optimization, which is characterized in that The step 4 further includes step 4.2):
Step 4.2) movement images Z and true value, Calculation Estimation index.
6. a kind of computer readable storage medium, is stored thereon with computer program, it is characterised in that: the computer program quilt The step of claim 1-4 any the method is realized when processor executes.
7. a kind of electronic equipment, which is characterized in that the electronic equipment includes:
Processor;
Computer readable storage medium is stored thereon with computer program, when the computer program is run by the processor Perform claim requires the step of 1 to 4 any the method.
CN201910355496.7A 2019-04-29 2019-04-29 Hyperspectral and multispectral image fusion method based on joint optimization, computer-readable storage medium and electronic device Active CN110148103B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910355496.7A CN110148103B (en) 2019-04-29 2019-04-29 Hyperspectral and multispectral image fusion method based on joint optimization, computer-readable storage medium and electronic device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910355496.7A CN110148103B (en) 2019-04-29 2019-04-29 Hyperspectral and multispectral image fusion method based on joint optimization, computer-readable storage medium and electronic device

Publications (2)

Publication Number Publication Date
CN110148103A true CN110148103A (en) 2019-08-20
CN110148103B CN110148103B (en) 2020-12-25

Family

ID=67594585

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910355496.7A Active CN110148103B (en) 2019-04-29 2019-04-29 Hyperspectral and multispectral image fusion method based on joint optimization, computer-readable storage medium and electronic device

Country Status (1)

Country Link
CN (1) CN110148103B (en)

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111369487A (en) * 2020-05-26 2020-07-03 湖南大学 Hyperspectral and multispectral image fusion method, system and medium
CN113205453A (en) * 2021-04-06 2021-08-03 武汉大学 Hyperspectral fusion method based on space-spectrum total variation regularization
CN113516646A (en) * 2021-07-26 2021-10-19 云南电网有限责任公司电力科学研究院 Method and device for enhancing high spectral resolution of multi-sensor image data
CN113822207A (en) * 2021-09-27 2021-12-21 海南长光卫星信息技术有限公司 Hyperspectral remote sensing image identification method and device, electronic equipment and storage medium
CN113888413A (en) * 2021-12-03 2022-01-04 湖南大学 Different-time different-source multispectral image blind spectrum super-resolution method and system
CN114092834A (en) * 2022-01-24 2022-02-25 南京理工大学 Unsupervised hyperspectral image blind fusion method and system based on space-spectrum combined residual correction network
CN116245779A (en) * 2023-05-11 2023-06-09 四川工程职业技术学院 Image fusion method and device, storage medium and electronic equipment
CN116757925A (en) * 2023-05-16 2023-09-15 宁波大学 Method and device for generating satellite remote sensing image with high space-time spectrum resolution

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101697008A (en) * 2009-10-20 2010-04-21 北京航空航天大学 Hyperspectral unmixing method for estimating regularized parameter automatically
CN103325096A (en) * 2013-06-25 2013-09-25 中国科学院遥感与数字地球研究所 Method for reconstructing wide hyperspectral image based on fusion of multispectral/hyperspectral images
CN108460749A (en) * 2018-03-20 2018-08-28 西安电子科技大学 A kind of rapid fusion method of EO-1 hyperion and multispectral image
CN109636769A (en) * 2018-12-18 2019-04-16 武汉大学 EO-1 hyperion and Multispectral Image Fusion Methods based on the intensive residual error network of two-way

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101697008A (en) * 2009-10-20 2010-04-21 北京航空航天大学 Hyperspectral unmixing method for estimating regularized parameter automatically
CN103325096A (en) * 2013-06-25 2013-09-25 中国科学院遥感与数字地球研究所 Method for reconstructing wide hyperspectral image based on fusion of multispectral/hyperspectral images
CN108460749A (en) * 2018-03-20 2018-08-28 西安电子科技大学 A kind of rapid fusion method of EO-1 hyperion and multispectral image
CN109636769A (en) * 2018-12-18 2019-04-16 武汉大学 EO-1 hyperion and Multispectral Image Fusion Methods based on the intensive residual error network of two-way

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
MOUSSA SOFIANE KAROUT ETC.: ""JOINT NONNEGATIVE MATRIX FACTORIZATION FOR HYPERSPECTRAL AND MULTISPECTRAL REMOTE SENSING DATA FUSION"", 《2013 5TH WORKSHIP ON HYPERSPECTRAL IMAGE AND SIGNAL PROCESSING: EVOLUTION IN REMOTE SENSING》 *

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111369487A (en) * 2020-05-26 2020-07-03 湖南大学 Hyperspectral and multispectral image fusion method, system and medium
CN111369487B (en) * 2020-05-26 2020-09-18 湖南大学 Hyperspectral and multispectral image fusion method, system and medium
CN113205453A (en) * 2021-04-06 2021-08-03 武汉大学 Hyperspectral fusion method based on space-spectrum total variation regularization
CN113516646A (en) * 2021-07-26 2021-10-19 云南电网有限责任公司电力科学研究院 Method and device for enhancing high spectral resolution of multi-sensor image data
CN113822207A (en) * 2021-09-27 2021-12-21 海南长光卫星信息技术有限公司 Hyperspectral remote sensing image identification method and device, electronic equipment and storage medium
CN113822207B (en) * 2021-09-27 2024-02-02 海南长光卫星信息技术有限公司 Hyperspectral remote sensing image identification method and device, electronic equipment and storage medium
CN113888413A (en) * 2021-12-03 2022-01-04 湖南大学 Different-time different-source multispectral image blind spectrum super-resolution method and system
CN113888413B (en) * 2021-12-03 2022-02-22 湖南大学 Different-time different-source multispectral image blind spectrum super-resolution method and system
CN114092834A (en) * 2022-01-24 2022-02-25 南京理工大学 Unsupervised hyperspectral image blind fusion method and system based on space-spectrum combined residual correction network
CN116245779A (en) * 2023-05-11 2023-06-09 四川工程职业技术学院 Image fusion method and device, storage medium and electronic equipment
CN116245779B (en) * 2023-05-11 2023-08-22 四川工程职业技术学院 Image fusion method and device, storage medium and electronic equipment
CN116757925A (en) * 2023-05-16 2023-09-15 宁波大学 Method and device for generating satellite remote sensing image with high space-time spectrum resolution

Also Published As

Publication number Publication date
CN110148103B (en) 2020-12-25

Similar Documents

Publication Publication Date Title
CN110148103A (en) EO-1 hyperion and Multispectral Image Fusion Methods, computer readable storage medium, electronic equipment based on combined optimization
Cai et al. Mst++: Multi-stage spectral-wise transformer for efficient spectral reconstruction
CN110135366B (en) Shielded pedestrian re-identification method based on multi-scale generation countermeasure network
Cai et al. Coarse-to-fine sparse transformer for hyperspectral image reconstruction
Nie et al. Deeply learned filter response functions for hyperspectral reconstruction
CN110363215B (en) Method for converting SAR image into optical image based on generating type countermeasure network
Tychsen-Smith et al. Denet: Scalable real-time object detection with directed sparse sampling
WO2018166438A1 (en) Image processing method and device and electronic device
CN109993072B (en) Low-resolution pedestrian re-identification system and method based on super-resolution image generation
Rao et al. A residual convolutional neural network for pan-shaprening
CN107525588B (en) Rapid reconstruction method of dual-camera spectral imaging system based on GPU
CN111127374B (en) Pan-sharing method based on multi-scale dense network
CN110501072B (en) Reconstruction method of snapshot type spectral imaging system based on tensor low-rank constraint
CN105761234A (en) Structure sparse representation-based remote sensing image fusion method
CN104318243B (en) High-spectral data dimension reduction method based on rarefaction representation and empty spectrum Laplce's figure
Lyu et al. Neural radiance transfer fields for relightable novel-view synthesis with global illumination
Rockwell et al. The 8-point algorithm as an inductive bias for relative pose prediction by vits
CN109146787A (en) A kind of real-time reconstruction method of the double camera spectrum imaging system based on interpolation
CN112819737A (en) Remote sensing image fusion method of multi-scale attention depth convolution network based on 3D convolution
CN111680579B (en) Remote sensing image classification method for self-adaptive weight multi-view measurement learning
Zhou et al. Spatial orthogonal attention generative adversarial network for MRI reconstruction
CN114298997A (en) Method and device for detecting forged picture and storage medium
CN107644393B (en) GPU-based abundance estimation algorithm parallel implementation method
Cao et al. Universal high spatial resolution hyperspectral imaging using hybrid-resolution image fusion
Polasek et al. Vision UFormer: Long-range monocular absolute depth estimation

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