CN107084789A - Single pixel detector spectrum reflectivity reconstructing method based on sparse prior - Google Patents
Single pixel detector spectrum reflectivity reconstructing method based on sparse prior Download PDFInfo
- Publication number
- CN107084789A CN107084789A CN201710213402.3A CN201710213402A CN107084789A CN 107084789 A CN107084789 A CN 107084789A CN 201710213402 A CN201710213402 A CN 201710213402A CN 107084789 A CN107084789 A CN 107084789A
- Authority
- CN
- China
- Prior art keywords
- reflectivity
- spectral
- basis function
- pixel detector
- single pixel
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01J—MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
- G01J3/00—Spectrometry; Spectrophotometry; Monochromators; Measuring colours
- G01J3/28—Investigating the spectrum
- G01J3/2823—Imaging spectrometer
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01J—MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
- G01J3/00—Spectrometry; Spectrophotometry; Monochromators; Measuring colours
- G01J3/28—Investigating the spectrum
- G01J3/2823—Imaging spectrometer
- G01J2003/2826—Multispectral imaging, e.g. filter imaging
Landscapes
- Physics & Mathematics (AREA)
- Spectroscopy & Molecular Physics (AREA)
- General Physics & Mathematics (AREA)
- Spectrometry And Color Measurement (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The present invention relates to a kind of single pixel detector spectrum reflectivity reconstructing method based on sparse prior, principal component analysis is carried out to training sample set, first three main composition of the spectral reflectance data of training sample set is obtained, is used as reconstruction basis function vector；The collection of single multispectral test color lumps spectral energy is carried out by single pixel detector, energy value U is obtained；In the solution procedure of training sample set, basis function vector B, basis function vector coefficient a and calculation matrix particular factor are obtainedSingle multispectral test color lumps spectral energy U is gathered by resulting, the reflectivity of test sample is reconstructed.The present invention can make full use of the sparse prior knowledge of spectral reflectivity space sparse features and lighting source relative spectral power distributions based on principal component orthogonal basis, reduce the optics complexity of multispectral data acquisition system, reduce hits, the rebuilding spectrum efficiency of reflectivity is improved, reconstruction precision is improved.
Description
Technical field
The present invention relates to a kind of image processing techniques, more particularly to a kind of single pixel detector spectrum based on sparse prior
Reflectivity reconstructing method.
Background technology
Spectral reflectivity reconstructing method is rebuild, wiener is rebuild, finite dimension by the multispectral image got using pseudoinverse
The methods such as reconstruction obtain the unrelated spectral reflectivity information of, scene unrelated with equipment.The equipment one of multispectral image is obtained at present
As light splitting, above first two are carried out using grating beam splitting, mechanical rotating filtering piece, liquid crystal tunable filter and nanohole array
Mode optical system complexity is higher, poor reliability, and latter two mode is expensive, complex manufacturing technology.
The content of the invention
The problem of requiring high and complicated the present invention be directed to the optical device of existing acquisition multispectral image, it is proposed that Yi Zhongji
In the single pixel detector spectrum reflectivity reconstructing method of sparse prior, the optics that can reduce multispectral data acquisition system is answered
Miscellaneous degree, reduces hits, improves the rebuilding spectrum efficiency of reflectivity, improves reconstruction precision.
The technical scheme is that：A kind of single pixel detector spectrum reflectivity reconstructing method based on sparse prior,
Specifically include following steps：
1) to the processing of training sample set：For the acquisition of the spectral reflectivity of face battle array object, gathered and adjusted using ccd array
The energy value of the spectral reflectivity of object after system, each pixels of CCD are entered as single pixel detector to the relevant position of object
The measurement of row light energy value, and then the multispectral reflectivity of face battle array object is obtained, and using following formula (1) to acquired
Reflectance spectrum carries out principal component analysis, so that obtain J basis function vector of training sample set, i.e., for training sample, its
The energy value u that all spectral reflectivity R of training sample set and corresponding single pixel detector are collected is, it is known that being
The input value of MATLAB softwares, basis function vector B, basis function vector coefficient a and calculation matrix M are tried to achieve by MATLAB softwares
Particular factor
R=Ba=[b after expansion1,b2,...,bJ][a1,a2,...,aJ]T
Wherein calculation matrix
2) to the processing of single multispectral test color lumps：Object is illuminated using the lighting source of known spectra power distribution,
Spectral reflectivity information to object is modulated, and single multispectral test color lumps spectral energy is carried out by single pixel detector
Collection, obtain energy value U；
3) reflectance spectrum of single multispectral test color lumps is rebuild：In step 1) solution procedure of training sample set
In, basis function vector B, basis function vector coefficient a and calculation matrix particular factorAll, it is known that logical
Cross step 2) obtained by single multispectral test color lumps spectral energy U is gathered, the reflectivity process of reconstruct test sample is：
First pass through the basis function vector coefficient (A that formula (3) solves test samples1,A2,...AJ), then reconstructed again by formula (4)
Spectral reflectivity r,
R=BA=[b1,b2,...,bJ][A1,A2,...,AJ]T (4)。
The step 3) in specific restructuring procedure it is as follows：
U=MTR=MTBA carries out the Its Sparse Decomposition based on principal component orthogonal basis for specific calculation matrix M；Then：
bJObtained by PCA, the main composition [b of first three in B1,b2,b3] contribution rate ＞ 95%, choose master
Composition is as sparse base, in view of the orthogonal property of principal component base vector, above formula (5) is expressed as：
Thus construct specific(A can be solved1,A2,…AJ), and then reconstruct spectral reflectance
Rate.
The beneficial effects of the present invention are：Single pixel detector spectrum reflectivity reconstruct side of the invention based on sparse prior
Method, can make full use of spectral reflectivity space sparse features and lighting source relative spectral power distributions to be based on principal component orthogonal
The sparse prior knowledge of base, reduces the optics complexity of multispectral data acquisition system, reduces hits, improves the light of reflectivity
Spectrum rebuilds efficiency, improves reconstruction precision.There is certain reference significance to the improvement of the acquiring technology of multispectral image.
Brief description of the drawings
Fig. 1 rebuilds schematic diagram for reflectance spectrum of the present invention based on iteration method；
Fig. 2 is first three basis function vector figure of the reflectance spectrum method for reconstructing of the specific embodiment of the invention；
Fig. 3 a are the light relatively of the modulation illumination light of the invention based on first principal component of Munsell colors training sample set
Spectral power distributions figure；
Fig. 3 b are the light relatively of the modulation illumination light of the invention based on second principal component of Munsell colors training sample set
Spectral power distributions figure；
Fig. 3 c are the light relatively of the modulation illumination light of the invention based on the 3rd principal component of Munsell colors training sample set
Spectral power distributions figure；
Fig. 4 is the spectral reflectivity figure that Munsell colors training sample of the present invention concentrates a color lump；
Fig. 5 is the single pixel detector spectrum reflectivity reconstruction result figure of the invention based on sparse prior.
Embodiment
Single pixel detector spectrum reflectivity reconstructing method step based on sparse prior is as follows：
1) to the processing of training sample set：Fig. 1 is the spectral reflectance recovery schematic diagram based on single pixel detector, for
The acquisition of the spectral reflectivity of face battle array object, the energy value of the spectral reflectivity of the object after modulation is gathered using ccd array
(each pixels of CCD carry out the measurement of light energy value as single pixel detector to the relevant position of object), and then obtain face battle array
The multispectral reflectivity of object.And principal component analysis is carried out to acquired reflectance spectrum using following formula (1), so that
J basis function vector of training sample set is obtained, first three basis function vector [b1, b2, b3] conduct is taken according to contribution rate size
Basis function vector is rebuild, because the contribution rate of first three basis function vector is to 95%, the contribution rate of basis function vector below
Only 5%.I.e. for training sample, all spectral reflectivity R of its training sample set and the collection of corresponding single pixel detector
To energy value u be that, it is known that be the input value of MATLAB softwares, basis function vector can be tried to achieve from there through MATLAB softwares
B, basis function vector coefficient a (also known as sparse coefficient) and calculation matrix M particular factor
R=Ba=[b after expansion1,b2,...,bJ][a1,a2,...,aJ]T (1)
There is N number of color lump in formula (1), each color lump has the spectral reflectivity that Q is tieed up, then R is that N × Q ties up reflectance spectrum
Vector, B is the basis function vector that N × J is tieed up, and a is the basis function vector coefficient that J × Q is tieed up, and J is basis function vector number, and Q is light
Compose the dimension of reflectivity.
2) to the processing of single multispectral test color lumps：Shone using the lighting source with specific relative spectral power distributions
(i.e. light source power distribution is the light source relative spectral power distributions difference, it is known that different to bright object, uses different spectral power distributions
The obtained value of the light source irradiation final single pixel detector of color lump be also different), the spectral reflectivity information to object is carried out
Modulation, the collection of single multispectral test color lumps spectral energy is carried out by single pixel detector.Obtain U.
The energy value U wherein obtained is the function on calculation matrix M and spectral reflectivity r.Such as formula U3×1=M3×Q×
rQ×1, wherein rQ×1It is the spectral reflectivity of single color lump, dimension is Q, U3×1For the energy value of test color lumps, calculation matrix M3×Q
By relative spectral power distributionsComposition.
J is that the corresponding calculation matrix of basis function vector number is：
3) reflectance spectrum of single multispectral test color lumps is rebuild：In the solution procedure of training sample set, base letter
Number vector B, basis function vector coefficient a and M particular factorAll, it is known that by step 2) obtained by
Single multispectral test color lumps spectral energy U is gathered, the reflectivity process of reconstruct test sample is：First pass through following public affairs
Formula (3) solves the basis function vector coefficient (A of test samples1,A2,...AJ), light is then reconstructed by following formula (4) again
Compose reflectivity r.
R=BA=[b1,b2,...,bJ][A1,A2,...,AJ]T (4)
Wherein, step 3) in specific restructuring procedure it is as follows：
(1) U=MTR=MTBA can carry out the Its Sparse Decomposition based on principal component orthogonal basis for specific calculation matrix M.
(2)
(3) then：
bJObtained by PCA, the main composition [b of first three in B1,b2,b3] contribution rate ＞ 95%, choose principal component and make
For sparse base, in view of the orthogonal property of principal component base vector, above formula (5) can be expressed as：
What wherein basis function vector B was offset.
It is specific by the visible construction of formula (3)(A can be solved1,A2,…AJ), and then reconstruct
Spectral reflectivity.Fig. 5 is to reconstruct spectral reflectivity by pseudoinverse technique and reconstruct spectral reflectivity based on single pixel detector to compare
Figure.It can be seen that compared to pseudoinverse technique, the optical spectrum reconstruction method reconstruction accuracy based on single pixel detector will more accurate one
A bit.
Claims (2)
1. a kind of single pixel detector spectrum reflectivity reconstructing method based on sparse prior, it is characterised in that specifically include as
Lower step：
1) to the processing of training sample set：For the acquisition of the spectral reflectivity of face battle array object, after ccd array collection modulation
Object spectral reflectivity energy value, each pixels of CCD are used as single pixel detector to carry out light to the relevant position of object
The measurement of energy value, and then the multispectral reflectivity of face battle array object is obtained, and using following formula (1) to acquired reflection
Rate spectrum carries out principal component analysis, so as to obtain J basis function vector of training sample set, i.e., for training sample, it is trained
The energy value u that all spectral reflectivity R of sample set and corresponding single pixel detector are collected is, it is known that soft for MATLAB
The input value of part, basis function vector B, basis function vector coefficient a and calculation matrix M particular factor are tried to achieve by MATLAB softwares
<mrow>
<mi>R</mi>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>j</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>J</mi>
</munderover>
<msub>
<mi>a</mi>
<mi>j</mi>
</msub>
<msub>
<mi>b</mi>
<mi>j</mi>
</msub>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
</mrow>
R=Ba=[b after expansion1,b2,...,bJ][a1,a2,...,aJ]T
Wherein calculation matrix
2) to the processing of single multispectral test color lumps：Object is illuminated using the lighting source of known spectra power distribution, to thing
The spectral reflectivity information of body is modulated, and adopting for single multispectral test color lumps spectral energy is carried out by single pixel detector
Collection, obtains energy value U；
3) reflectance spectrum of single multispectral test color lumps is rebuild：In step 1) in the solution procedure of training sample set, base
Functional vector B, basis function vector coefficient a and calculation matrix particular factorAll, it is known that passing through step
2) being gathered to single multispectral test color lumps spectral energy U obtained by, the reflectivity process for reconstructing test sample is：First pass through
Formula (3) solves the basis function vector coefficient (A of test samples1,A2,...AJ), it is then anti-by formula (4) reconstruct spectrum again
Rate r is penetrated,
R=BA=[b1,b2,...,bJ][A1,A2,...,AJ]T (4)。
2. the single pixel detector spectrum reflectivity reconstructing method based on sparse prior according to claim 1, its feature exists
In the step 3) in specific restructuring procedure it is as follows：
U=MTR=MTBA carries out the Its Sparse Decomposition based on principal component orthogonal basis for specific calculation matrix M；
Then：
bJObtained by PCA, the main composition [b of first three in B1,b2,b3] contribution rate ＞ 95%, choose principal component and make
For sparse base, in view of the orthogonal property of principal component base vector, above formula (5) is expressed as：
Thus construct specific(A can be solved1,A2,…AJ), and then reconstruct spectral reflectivity.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710213402.3A CN107084789A (en) | 2017-04-01 | 2017-04-01 | Single pixel detector spectrum reflectivity reconstructing method based on sparse prior |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710213402.3A CN107084789A (en) | 2017-04-01 | 2017-04-01 | Single pixel detector spectrum reflectivity reconstructing method based on sparse prior |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107084789A true CN107084789A (en) | 2017-08-22 |
Family
ID=59614389
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710213402.3A Pending CN107084789A (en) | 2017-04-01 | 2017-04-01 | Single pixel detector spectrum reflectivity reconstructing method based on sparse prior |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107084789A (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107941337A (en) * | 2017-11-16 | 2018-04-20 | 北京理工大学 | Fast illuminated spectrum imaging method, device and optical spectrum imagers |
| CN108507674A (en) * | 2018-03-13 | 2018-09-07 | 北京航空航天大学 | A kind of nominal data processing method of light field light spectrum image-forming spectrometer |
| CN109506780A (en) * | 2018-11-23 | 2019-03-22 | 浙江智彩科技有限公司 | Object spectra reflectivity method for reconstructing based on multispectral LED illumination |
| CN109655233A (en) * | 2018-12-18 | 2019-04-19 | 厦门大学 | A kind of multichannel light spectrum image-forming display screen Systems for optical inspection and its detection method |
| CN110175971A (en) * | 2019-05-27 | 2019-08-27 | 大连海事大学 | Deep learning image reconstruction method for multispectral single-pixel imaging |
| CN110658179A (en) * | 2019-10-09 | 2020-01-07 | 上海理工大学 | Laser Raman gas concentration detection method based on multi-signal superposition and pseudo-inverse method |
| CN110736542A (en) * | 2019-10-28 | 2020-01-31 | 南京林业大学 | spectrum reconstruction method based on RGB values |
| CN110926608A (en) * | 2019-10-14 | 2020-03-27 | 齐鲁工业大学 | Spectrum reconstruction method based on light source screening |
| CN111397733A (en) * | 2020-04-23 | 2020-07-10 | 湖南大学 | Single/multi-frame snapshot type spectral imaging method, system and medium |
| CN112240801A (en) * | 2020-10-13 | 2021-01-19 | 中国科学院长春光学精密机械与物理研究所 | Polarization Imaging System |
| CN112884854A (en) * | 2021-01-13 | 2021-06-01 | 齐鲁工业大学 | Spectrum sparse reconstruction method based on camera response value |
| CN112991299A (en) * | 2021-03-18 | 2021-06-18 | 中国科学院紫金山天文台 | Method for constructing smooth point diffusion function in image processing |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0530818A2 (en) * | 1991-09-04 | 1993-03-10 | X-Rite, Inc. | Multiangular color measuring apparatus |
| CN103528968A (en) * | 2013-11-01 | 2014-01-22 | 上海理工大学 | Reflectance spectrum reconstruction method based on iterative threshold method |
| CN104359556A (en) * | 2014-11-14 | 2015-02-18 | 武汉大学 | Optimal training sample selection method for broad band spectrum imaging system |
| CN104634451A (en) * | 2015-02-11 | 2015-05-20 | 武汉大学 | Spectrum reconstruction method and system based on multichannel imaging system |
-
2017
- 2017-04-01 CN CN201710213402.3A patent/CN107084789A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0530818A2 (en) * | 1991-09-04 | 1993-03-10 | X-Rite, Inc. | Multiangular color measuring apparatus |
| CN103528968A (en) * | 2013-11-01 | 2014-01-22 | 上海理工大学 | Reflectance spectrum reconstruction method based on iterative threshold method |
| CN104359556A (en) * | 2014-11-14 | 2015-02-18 | 武汉大学 | Optimal training sample selection method for broad band spectrum imaging system |
| CN104634451A (en) * | 2015-02-11 | 2015-05-20 | 武汉大学 | Spectrum reconstruction method and system based on multichannel imaging system |
Non-Patent Citations (1)
| Title |
|---|
| 张雷洪等: "《Study on the key technology of spectral reflectivity reconstruction based on sparse prior by a single-pixel detector》", 《PHOTON.RES.》 * |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107941337A (en) * | 2017-11-16 | 2018-04-20 | 北京理工大学 | Fast illuminated spectrum imaging method, device and optical spectrum imagers |
| CN108507674A (en) * | 2018-03-13 | 2018-09-07 | 北京航空航天大学 | A kind of nominal data processing method of light field light spectrum image-forming spectrometer |
| CN109506780B (en) * | 2018-11-23 | 2020-11-06 | 浙江智彩科技有限公司 | Reconstruction Method of Object Spectral Reflectance Based on Multispectral LED Lighting |
| CN109506780A (en) * | 2018-11-23 | 2019-03-22 | 浙江智彩科技有限公司 | Object spectra reflectivity method for reconstructing based on multispectral LED illumination |
| CN109655233A (en) * | 2018-12-18 | 2019-04-19 | 厦门大学 | A kind of multichannel light spectrum image-forming display screen Systems for optical inspection and its detection method |
| CN109655233B (en) * | 2018-12-18 | 2020-02-07 | 厦门大学 | Optical detection system and detection method for multi-channel spectral imaging display screen |
| CN110175971A (en) * | 2019-05-27 | 2019-08-27 | 大连海事大学 | Deep learning image reconstruction method for multispectral single-pixel imaging |
| CN110175971B (en) * | 2019-05-27 | 2022-09-16 | 大连海事大学 | A Deep Learning Image Reconstruction Method for Multispectral Single Pixel Imaging |
| CN110658179A (en) * | 2019-10-09 | 2020-01-07 | 上海理工大学 | Laser Raman gas concentration detection method based on multi-signal superposition and pseudo-inverse method |
| CN110658179B (en) * | 2019-10-09 | 2021-11-19 | 上海理工大学 | Laser Raman gas concentration detection method based on multi-signal superposition and pseudo-inverse method |
| CN110926608A (en) * | 2019-10-14 | 2020-03-27 | 齐鲁工业大学 | Spectrum reconstruction method based on light source screening |
| CN110736542B (en) * | 2019-10-28 | 2021-07-16 | 南京林业大学 | A Spectral Reconstruction Method Based on RGB Values |
| CN110736542A (en) * | 2019-10-28 | 2020-01-31 | 南京林业大学 | spectrum reconstruction method based on RGB values |
| CN111397733B (en) * | 2020-04-23 | 2021-03-02 | 湖南大学 | A single/multi-frame snapshot spectral imaging method, system and medium |
| CN111397733A (en) * | 2020-04-23 | 2020-07-10 | 湖南大学 | Single/multi-frame snapshot type spectral imaging method, system and medium |
| CN112240801A (en) * | 2020-10-13 | 2021-01-19 | 中国科学院长春光学精密机械与物理研究所 | Polarization Imaging System |
| CN112884854A (en) * | 2021-01-13 | 2021-06-01 | 齐鲁工业大学 | Spectrum sparse reconstruction method based on camera response value |
| CN112884854B (en) * | 2021-01-13 | 2022-06-03 | 齐鲁工业大学 | A Spectral Sparse Reconstruction Method Based on Camera Response Values |
| CN112991299A (en) * | 2021-03-18 | 2021-06-18 | 中国科学院紫金山天文台 | Method for constructing smooth point diffusion function in image processing |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107084789A (en) | Single pixel detector spectrum reflectivity reconstructing method based on sparse prior | |
| Agüeros et al. | A new look at an old cluster: the membership, rotation, and magnetic activity of low-mass stars in the 1.3 Gyr old open cluster NGC 752 | |
| CN115326783B (en) | Raman spectrum preprocessing model generation method, system, terminal and storage medium | |
| Villar et al. | The Hα-based star formation rate density of the universe at z= 0.84 | |
| Faisst et al. | Constraints on Quenching of Z≲ 2 Massive Galaxies from the Evolution of the Average Sizes of Star-forming and Quenched Populations in COSMOS | |
| CN108062744B (en) | A deep learning-based super-resolution reconstruction method for mass spectrometry images | |
| Chen et al. | STORM imaging of mitochondrial dynamics using a vicinal-dithiol-proteins-targeted probe | |
| CN103188989B (en) | Method for reconstructing dynamic fluorescence molecular tomography | |
| CN111259954B (en) | D-Resnet-based hyperspectral traditional Chinese medicine tongue fur and tongue quality classification method | |
| CN104897583B (en) | A kind of micro bloodstain the cannot-harm-detection device and method | |
| CN104614321A (en) | Real-time crop growth condition monitoring method based on spectrum images | |
| Weisz et al. | The JWST Resolved Stellar Populations Early Release Science Program. V. DOLPHOT Stellar Photometry for NIRCam and NIRISS | |
| Gong et al. | Spectrum-image dual-modality fusion empowered accurate and efficient classification system for traditional Chinese medicine | |
| Liu et al. | Detection of heterogeneity in multi-spectral transmission image based on spatial pyramid matching model and deep learning | |
| Li et al. | Comparative analysis of pixel-level fusion algorithms and a new high-resolution dataset for SAR and optical image fusion | |
| CN105181606A (en) | Hyperspectral imaging technology-based method for detecting sucrose content distribution of peanut | |
| CN101908211A (en) | A hyperspectral image fusion method based on variational method | |
| CN116310788A (en) | Hyperspectral modeling method of nitrogen content in plant leaves based on map fusion | |
| CN115294434A (en) | A method for implementing a multimodal learning convolutional neural network model based on the acquisition and fusion of molecular spectral features | |
| Donath et al. | Joint Deconvolution of Astronomical Images in the Presence of Poisson Noise | |
| Chen et al. | Bayesian experimental design and parameter estimation for ultrafast spin dynamics | |
| CN103969659B (en) | Celestial body spectrum image imaging method based on compressed sensing | |
| Hussain et al. | Depth separable-CNN for improved spectral super-resolution | |
| CN118603317B (en) | A single-pixel hyperspectral imaging system and method based on dual filters | |
| Nunez et al. | Medium-adaptive Compressive Diffuse Optical Tomography |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170822 |
|
| WD01 | Invention patent application deemed withdrawn after publication |