CN110660112A - Drawing spectrum reconstruction method based on special color card and multispectral imaging - Google Patents
Drawing spectrum reconstruction method based on special color card and multispectral imaging Download PDFInfo
- Publication number
- CN110660112A CN110660112A CN201910936265.5A CN201910936265A CN110660112A CN 110660112 A CN110660112 A CN 110660112A CN 201910936265 A CN201910936265 A CN 201910936265A CN 110660112 A CN110660112 A CN 110660112A
- Authority
- CN
- China
- Prior art keywords
- spectral
- spectral reflectance
- painting
- image
- channel response
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/001—Texturing; Colouring; Generation of texture or colour
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Spectrometry And Color Measurement (AREA)
Abstract
The invention discloses a drawing spectrum reconstruction method based on a special color card and multispectral imaging. For art painting and ancient writing which have artistic value and are inconvenient to contact and measure, the invention introduces a special color card which has the same pigment and substrate as the painting, utilizes a multispectral imaging system to obtain multichannel response values of the target art painting and the special color card, and measures the spectral reflectance of a color block of a training sample of the special color card. Similar training samples are merged by clustering to prevent overfitting and obtain a training sample set containing dedicated color target multi-channel response values and corresponding spectral reflectance. And establishing a conversion matrix of the multi-channel response values and the spectral reflectance in the training sample set, and mapping the multi-channel response value image of the painting into a corresponding spectral reflectance image by using the matrix so as to reconstruct the spectral image of the target painting.
Description
Technical Field
The invention relates to a method for obtaining spectral information and chromaticity information of art painting in an imaging mode, in particular to a method for obtaining reflection spectral information of the art painting by utilizing a multispectral camera with a plurality of visible light wave band channels so as to further obtain the chromaticity information of the art painting.
Background
The chromaticity value of an object obtained by the traditional image capturing method is influenced by illumination light during image capturing, cannot represent the intrinsic characteristics of the surface of the object and is easily influenced by metamerism; in order to be able to obtain the true appearance information of an object, one approach is to obtain physical quantities that are independent of the photographing apparatus and the light source: spectral reflectance of the object. In the digital preservation of art painting, if the spectral reflectance of the digital painting can be accurately obtained, the information can be preserved to the maximum extent, and the color appearance of the painting under any light source can be accurately reproduced.
The existing spectral reconstruction algorithm is mainly based on a method of training samples, namely, a multi-channel response value of a sample shot by a multi-spectral camera and a spectral reflectance corresponding to the sample measured by a photometer are utilized to establish a relation between a multi-channel response value of a target object and the spectral reflectance, so that a reconstructed drawing spectral image is calculated. If the training sample is not consistent with the target object in material, the precision of spectrum reconstruction is greatly reduced, and the target object is directly used as a self-training sample, which may cause irreversible damage to precious art painting and cultural relics, so that a special color card made of the same substrate and pigment as the original painting is selected as the training sample of spectrum reconstruction. However, if the training sample constructed by the special color card is not selected, the local content of the painting is often over-sampled or under-sampled, and the spectral reconstruction precision is reduced.
Disclosure of Invention
In order to realize a simple and easy spectrum reconstruction process, simultaneously improve the consistency of a training sample and a painting material so as to improve the reconstruction precision and prevent oversampling or undersampling, the invention provides a method for selecting a reconstruction training sample by using a special color card to reconstruct a painting spectrum.
The invention has the following inventive concept: for art painting and ancient writing which have artistic value and are inconvenient to contact and measure, a special color card with the same pigment and matrix as the painting is introduced, a multispectral imaging system is utilized to obtain multichannel response values of the target art painting and the special color card, and the spectral reflectance of a color block of a training sample of the special color card is measured. Similar training samples are merged by clustering to prevent overfitting and obtain a training sample set containing dedicated color target multi-channel response values and corresponding spectral reflectance. And establishing a conversion matrix of the multi-channel response values and the spectral reflectance in the training sample set, and mapping the multi-channel response value image of the painting into a corresponding spectral reflectance image by using the matrix so as to reconstruct the spectral image of the target painting.
The purpose of the invention is realized by the following technical scheme:
a drawing spectrum reconstruction method based on special color card and multispectral imaging comprises the following steps:
s1: marking uniform parts of different color blocks on a special color card of the target drawing as sampling points of spectrum reconstruction;
s2: multi-channel image P for shooting target drawing by multispectral camerapKeeping the illumination environment and the position of the multispectral camera unchanged, and shooting the multispectral image P of the special color cardc；
S3: the spectral reflectance r of n color patches in the exclusive color chart was measured using a spectrophotometer according to the marking made in S1c1、rc2、…rcnAnd correspondingly at PcThe multi-channel response value of the middle extracted mark position is recorded as pc1、pc2、…pcnR is tociAnd pciObtaining N pairs of sampling point pairs in a one-to-one correspondence mode, and using the N pairs of sampling point pairs as an alternative sample set N for spectrum recovery;
s4: processing a multi-channel image P of a drawing using a clustering algorithmpThe multi-channel response value of each pixel divides multispectral pixel points of the drawing into m types<N, selecting M pairs of sampling point pairs with the minimum Euclidean distance from the corresponding various centers from the alternative sample set N for spectrum recovery as a training sample set M for spectrum recovery;
s5: using the spectral reflectance r of all sampling point pairs in the training sample set MciWith corresponding multi-channel response values pciEstablishing a corresponding mapping relation between the multi-channel response value and the spectral reflectance, wherein rci∈M，pci∈M；
S6：According to the mapping relation obtained in S5, calculating a multispectral image P of the target drawingpTo the spectral reflectance of each pixel, thereby reconstructing a spectral image R of the drawingp。
Preferably, the special color card is a plurality of color block samples which are drawn by using the same or similar pigment and matrix as the target drawing and are arranged in a color gradient mode.
Preferably, the specific implementation method of step S4 is:
multi-channel image P for drawing target by utilizing K-means clustering algorithmpClustering the multi-channel response value of each pixel, and after the class center iterative optimization is terminated, dividing multispectral pixel points of the painting into m classes to obtain m class centers<n; respectively calculating the sampling point pairs nearest to the center of each class in the alternative sample set N for spectrum recovery to obtain m pairs of sampling point pairs Cn1、Cn2、…CnmCorresponding to the spectral reflectance r of the training samplen1、rn2、…rnmAnd a multi-channel response value pn1、pn2、…pnmA training sample set M for spectral recovery is formed.
Preferably, the specific implementation method of step S5 is:
the multi-channel response value p in the training sample set Mn1、pn2、…pnmForming a multi-channel response value matrix P, corresponding spectral reflectance rn1、rn2、…rnmForming a spectral reflectance matrix R;
calculating a conversion matrix T from the multichannel response values to the spectral reflectance as follows:
T＝RPT(PPT)-1
applying a transformation matrix T to a multispectral image PpEstablishing a corresponding mapping relation between the multi-channel response value and the spectral reflectance for calculating and obtaining a corresponding spectral image Rp：
Rp＝TPp。
The invention has the beneficial effects that: the invention reduces the spectrum reconstruction error caused by the inconsistency of the training sample and the target painting spectrum and simultaneously avoids the damage to the painting caused by direct sampling and measurement on the precious art painting by extracting the multichannel response value and the spectrum reflectance ratio of the special color chart close to the spectrum reflectance ratio of the art painting as the training sample for spectrum reconstruction.
Drawings
FIG. 1 is a color ink painting image (lower image) and a partial special color card color lump image (upper image) according to an embodiment, wherein a sampling area of a training sample is marked by a square box;
FIG. 2 is a spectral reflectance of a portion of a training sample;
fig. 3 is a flow chart of spectral image restoration.
Detailed Description
The invention will be further elucidated and described with reference to the drawings and the detailed description.
The invention discloses a drawing spectrum reconstruction method based on a special color card and multispectral imaging, which comprises the following steps:
s1: and a special color card for marking the target drawing is used for sampling points of spectral reconstruction, wherein the special color card is a plurality of color block samples which are drawn by using pigment and matrix which are the same as or similar to the target drawing and are arranged in a color gradient mode. Due to pen touch and bleeding which may exist in the drawing of the special color card, the color block of the special color card may have a phenomenon of non-uniform color, which affects the extraction of the multichannel response value and the measurement of the spectral reflectance, so that the uniform part of the color block needs to be marked as a training sample.
S2: multi-channel image P for shooting target drawing by multispectral camerapKeeping the illumination environment and the position of the multispectral camera unchanged, and shooting the multispectral image P of the special color cardc. The illumination environment is unchanged to ensure that the mapping relationship established by the multi-channel values and the spectral reflectance of the special color card can be used for the spectral reconstruction based on the multi-channel image.
S3: the spectral reflectance r of n color patches in the exclusive color chart was measured using a spectrophotometer according to the marking made in S1c1、rc2、…rcnAnd correspondingly at PcWell extract markPosition-recorded multi-channel response value, recorded as pc1、pc2、…pcnR is tociAnd pciAnd obtaining N pairs of sampling point pairs in a one-to-one correspondence mode, wherein the one-to-one correspondence sampling point set serves as an alternative sampling set N for spectrum recovery.
S4: processing a multi-channel image P of a drawing using a clustering algorithmpThe multi-channel response value of each pixel divides multispectral pixel points of the drawing into m types<And N, selecting M pairs of sampling point pairs with the minimum Euclidean distance from the corresponding various centers from the alternative sample set N for spectrum recovery as a training sample set M for spectrum recovery. In the present invention, this step can be specifically realized by the following method:
multi-channel image P for drawing target by utilizing K-means clustering algorithmpClustering the multi-channel response value of each pixel, and after the class center iterative optimization is terminated, dividing multispectral pixel points of the painting into m classes to obtain m class centers<n; respectively calculating the sampling point pairs nearest to the center of each class in the alternative sample set N for spectrum recovery to obtain m pairs of sampling point pairs Cn1、Cn2、…CnmCorresponding to the spectral reflectance r of the training samplen1、rn2、…rnmAnd a multi-channel response value pn1、pn2、…pnmA training sample set M for spectral recovery is formed.
S5: using the spectral reflectance r of all sampling point pairs in the training sample set MciWith corresponding multi-channel response values pciEstablishing a corresponding mapping relation between the multi-channel response value and the spectral reflectance based on a pseudo-inverse method, wherein rci∈M，pciE.g. M. In the present invention, this step can be specifically realized by the following method:
the multi-channel response value p in the training sample set Mn1、pn2、…pnmForming a multi-channel response value matrix P, corresponding spectral reflectance rn1、rn2、…rnmForming a spectral reflectance matrix R;
calculating a conversion matrix T from the multichannel response values to the spectral reflectance as follows:
T＝RPT(PPT)-1
applying a transformation matrix T to a multispectral image PpEstablishing a corresponding mapping relation between the multi-channel response value and the spectral reflectance for calculating and obtaining a corresponding spectral image Rp：
Rp＝TPp。
S6: according to the mapping relation obtained in S5, calculating a multispectral image P of the target drawingpTo the spectral reflectance of each pixel, thereby reconstructing a spectral image R of the drawingp。
The above method is applied in the following embodiments in order to better demonstrate the technical effect thereof.
Examples
In this embodiment, a method for selecting a reconstruction training sample for reconstructing a drawing spectrum using a color filter wheel type multispectral camera and a handheld spectrophotometer is described, taking an example of using a color filter wheel type multispectral camera and a handheld spectrophotometer in cooperation with a special color chart of a color wash painting to measure and reconstruct a color wash painting.
At present, most of image spectral reflectance recovery algorithms based on multispectral images use a universal color chart as a training sample for spectral recovery, the training sample has a fixed and known spectral reflectance and does not need to be additionally measured and calibrated, but because the spectral reflectance of a real object is often different from that of the universal color chart to a certain extent, the spectral reflectance of the image recovered by the universal color chart generates errors, and the image spectral reflectance is distorted. The invention provides a spectral image recovery method based on a special color chart for painting, which selects a proper special color chart training sample by a K-means clustering method to obtain a more accurate spectral image recovery result.
FIG. 1 shows the appearance of a color ink-wash painting and a portion of a custom color chart in an embodiment, and the location of the sampling point for measuring the spectral reflectance of a training sample.
FIG. 2 is a graph of spectral reflectance of a portion of the proprietary color chart of the example.
Fig. 3 is a drawing spectrum reconstruction flow based on a special color chart training sample.
1. The process for selecting the special color card training sample comprises the following steps:
in order to select a proper number of training samples and prevent the spectral recovery precision from being reduced due to the oversampling of the training samples on the basis of ensuring the coverage of a color space as large as possible, the training samples need to be clustered, and the most representative sample color block is selected from a training sample subset to be used as a training sample finally input into a spectral reconstruction algorithm.
To have a reflectance of r for n spectrac1、rc2、…rcnThe color-related clustering of the color blocks of the training samples is carried out, and the chromaticity values of the color blocks are acquired firstly. For this purpose, the training samples are converted into the CIELAB space. The object tristimulus value based on the CIE1931 chromaticity system can be calculated by the following equation:
x, Y, Z are the tristimulus values of the object, respectively; p (lambda) is the spectral power distribution of the light source, in the present invention the luminance is 100cd/m2Standard D65 light source; ρ (λ) is the reflectivity of the surface of the object, and in the present invention is rcn(ii) a x (λ), y (λ), z (λ) are CIE1931 standard chromaticity observer spectrum tristimulus values, also known as color matching functions CMF of human eyes, i.e. colormatching functions; k is a normalized value in the CIE1931 chromaticity system; lambda [ alpha ]min、λmaxThe upper and lower limits of the visible light band are usually λmin＝400nm，λmax＝700nm。
The XYZ tristimulus values of the training sample are not linear with their color appearance, and need to be converted to a uniform color space in order to better characterize the effect of human eyes viewing the training sample. The present invention uses CIEL a b space and the correlation transformation can be calculated by:
wherein
In the formula Xn、Yn、ZnFor reference to white point tristimulus values, X is under standard D65 illuminantn＝95.047，Yn＝100，Zn＝108.883。
By the above conversion, the reflectance r is obtainedc1、rc2、…rcnL of training samples*a*b*Value, denoted as C1、C2、…Cn。
Clustering of training samples at L*a*b*The method is performed in space, and the embodiment clusters n sample points by using a K-means clustering method. And selecting a proper clustering number m, wherein the size of m is at least 5 to 10 times of the number of the types of the painting pigments used by the special color card, and the number of m can be adjusted according to the number n of the training samples. And the K-means clustering optimizes the selection of class centers through continuous iteration, and initially selects m random points as the centers of m classes. One iterative process is as follows: calculate all sample points at L*a*b*The Euclidean distance of the center of each class in the space, and each point is allocated to the class with the minimum Euclidean distance from the center of the class; and recalculating the class center of each class, namely the point with the minimum sum of Euclidean distances from each point in the class. Repeating for multiple times until the class center does not change any more, and obtaining m training samples C with minimum Euclidean distance from each class centern1、Cn2、…CnmCorresponding to the spectral reflectance r of the training samplen1、rn2、…rnmAnd a multi-channel response value pn1、pn2、…pnmA training sample set M for spectral recovery is formed.
2. The spectrum reconstruction process comprises the following steps:
the special color chip is marked for the sampling point of the spectrum reconstruction. In order to avoid that the color unevenness of the color block of the special color card affects the extraction of the multichannel response value of the shooting and the measurement of the spectral reflectance, the color block even part needs to be marked for the subsequent shooting sampling and the spectral reflectance measurement.
Capturing a multi-channel image P of a painting under a high color rendering LED light source using a multi-spectral camerapKeeping the illumination environment and the position of the multispectral camera unchanged, replacing the painting with the special color card, and shooting the multispectral image P of the special color cardcAnd the mapping relation established by the multichannel value and the spectral reflectance of the special color card can be used for spectral reconstruction based on the multichannel image.
After the training sample set M is obtained through the aforementioned training sample selection procedure, a conversion matrix T for converting to spectral reflectance using multi-pass may be calculated using the training sample set M. There are various methods for obtaining the transformation matrix, and the present embodiment adopts a pseudo-inverse method. Training multichannel response value p in sample set Mn1、pn2、…pnmForming a multi-channel response value matrix P, forming a spectral reflectance matrix R by the corresponding spectral reflectance, and converting the matrix T to minimize the square error of the spectral reflectance reconstructed by P and the actual spectral reflectance R, namely:
min(||R-TP||2)
wherein | · | | represents the frobenius norm of the matrix, and the transformation matrix T from the obtained multichannel response value to the spectral reflectance is:
T＝RPT(PPT)-1
the transformation matrix T obtained by calculation can be applied to the multispectral image PpCalculating to obtain the corresponding spectral image Rp：
Rp＝TPp
The spectral image RpMore in-depth spectroscopy and coloristic applications are possible.
The above embodiment is merely a preferred embodiment of the present invention, but not intended to limit the present invention. Various changes and modifications may be made by one skilled in the relevant art without departing from the spirit and scope of the invention. Therefore, the technical scheme obtained by adopting the modes of equivalent replacement or equivalent transformation and the like falls into the protection scope of the invention.
Claims (5)
1. A drawing spectrum reconstruction method based on special color card and multispectral imaging is characterized by comprising the following steps:
s1: marking uniform parts of different color blocks on a special color card of the target drawing as sampling points of spectrum reconstruction;
s2: multi-channel image P for shooting target drawing by multispectral camerapKeeping the illumination environment and the position of the multispectral camera unchanged, and shooting the multispectral image P of the special color cardc；
S3: the spectral reflectance r of n color patches in the exclusive color chart was measured using a spectrophotometer according to the marking made in S1c1、rc2、…rcnAnd correspondingly at PcThe multi-channel response value of the middle extracted mark position is recorded as pc1、pc2、…pcnR is tociAnd pciObtaining N pairs of sampling point pairs in a one-to-one correspondence mode, and using the N pairs of sampling point pairs as an alternative sample set N for spectrum recovery;
s4: processing a multi-channel image P of a drawing using a clustering algorithmpThe multi-channel response value of each pixel divides multispectral pixel points of the drawing into m types<N, selecting M pairs of sampling point pairs with the minimum Euclidean distance from the corresponding various centers from the alternative sample set N for spectrum recovery as a training sample set M for spectrum recovery;
s5: using the spectral reflectance r of all sampling point pairs in the training sample set MciWith corresponding multi-channel response values pciEstablishing a corresponding mapping relation between the multi-channel response value and the spectral reflectance, wherein rci∈M，pci∈M；
S6: according to the mapping relation obtained in S5, calculating a multispectral image P of the target drawingpTo the spectral reflectance of each pixel, thereby reconstructing a spectral image R of the drawingp。
2. The method for spectral reconstruction of a painting based on specialized color chips and multi-spectral imaging according to claim 1, wherein the specialized color chips are color gradient arranged color patch samples drawn using the same or similar pigments and matrices as the target painting.
3. The method for reconstructing a drawing spectrum based on dedicated color chips and multispectral imaging as claimed in claim 1, wherein the step S4 is implemented by:
multi-channel image P for drawing target by utilizing K-means clustering algorithmpClustering the multi-channel response value of each pixel, and after the class center iterative optimization is terminated, dividing multispectral pixel points of the painting into m classes to obtain m class centers<n; respectively calculating the sampling point pairs nearest to the center of each class in the alternative sample set N for spectrum recovery to obtain m pairs of sampling point pairs Cn1、Cn2、…CnmCorresponding to the spectral reflectance r of the training samplen1、rn2、…rnmAnd a multi-channel response value pn1、pn2、…pnmA training sample set M for spectral recovery is formed.
4. The method for reconstructing a drawing spectrum based on dedicated color chips and multispectral imaging as claimed in claim 1, wherein the step S5 is implemented by:
the multi-channel response value p in the training sample set Mn1、pn2、…pnmForming a multi-channel response value matrix P, corresponding spectral reflectance rn1、rn2、…rnmForming a spectral reflectance matrix R;
calculating a conversion matrix T from the multichannel response values to the spectral reflectance as follows:
T＝RPT(PPT)-1
applying a transformation matrix T to a multispectral image PpEstablishing a corresponding mapping relation between the multi-channel response value and the spectral reflectance for calculating and obtaining a corresponding spectral image Rp：
Rp＝TPp。
5. A method for spectral reconstruction of a painting based on specialized color chips and multi-spectral imaging as claimed in claim 1, wherein said target painting is an artistic painting or historical relic that is not convenient for touch measurement.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910936265.5A CN110660112B (en) | 2019-09-29 | 2019-09-29 | Drawing spectrum reconstruction method based on special color card and multispectral imaging |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910936265.5A CN110660112B (en) | 2019-09-29 | 2019-09-29 | Drawing spectrum reconstruction method based on special color card and multispectral imaging |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110660112A true CN110660112A (en) | 2020-01-07 |
| CN110660112B CN110660112B (en) | 2021-09-24 |
Family
ID=69038563
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910936265.5A Active CN110660112B (en) | 2019-09-29 | 2019-09-29 | Drawing spectrum reconstruction method based on special color card and multispectral imaging |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110660112B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111750993A (en) * | 2020-06-04 | 2020-10-09 | 武汉纺织大学 | Open measurement environment spectrum measurement method based on imaging condition correction |
| CN113538285A (en) * | 2021-07-23 | 2021-10-22 | 奥比中光科技集团股份有限公司 | Method and device for determining spectral image, terminal and storage medium |
| CN113840135A (en) * | 2021-09-03 | 2021-12-24 | 大连中科创达软件有限公司 | Color cast detection method, device, equipment and storage medium |
| CN113935894A (en) * | 2021-09-09 | 2022-01-14 | 完美世界(北京)软件科技发展有限公司 | Ink and wash style scene rendering method and equipment and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8285018B2 (en) * | 2010-03-02 | 2012-10-09 | National Chung Cheng University | Method for reconstructing color images |
| US20130293702A1 (en) * | 2010-11-01 | 2013-11-07 | The Hong Kong Research Institute Of Textiles And Apparel Limited | Multispectral imaging color measurement system and method for processing imaging signals thereof |
| CN106153192A (en) * | 2016-07-22 | 2016-11-23 | 浙江大学 | A kind of method utilizing multispectral camera virtual responsive value to obtain spectral reflectance |
| CN106153193A (en) * | 2016-07-22 | 2016-11-23 | 浙江大学 | A kind of method utilizing the double light source response value of multispectral camera to obtain spectral reflectance |
| CN106841055A (en) * | 2017-03-22 | 2017-06-13 | 浙江大学 | A kind of training sample selection method for reconstructing art drawing spectrum picture |
| CN108896176A (en) * | 2018-05-14 | 2018-11-27 | 浙江大学 | A kind of Space Consistency bearing calibration of multi-optical spectrum imaging system |
| CN109839189A (en) * | 2018-12-27 | 2019-06-04 | 浙江农林大学暨阳学院 | Utilize the method for multispectral camera self-adapting reconstruction spectral reflectance |
-
2019
- 2019-09-29 CN CN201910936265.5A patent/CN110660112B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8285018B2 (en) * | 2010-03-02 | 2012-10-09 | National Chung Cheng University | Method for reconstructing color images |
| US20130293702A1 (en) * | 2010-11-01 | 2013-11-07 | The Hong Kong Research Institute Of Textiles And Apparel Limited | Multispectral imaging color measurement system and method for processing imaging signals thereof |
| CN106153192A (en) * | 2016-07-22 | 2016-11-23 | 浙江大学 | A kind of method utilizing multispectral camera virtual responsive value to obtain spectral reflectance |
| CN106153193A (en) * | 2016-07-22 | 2016-11-23 | 浙江大学 | A kind of method utilizing the double light source response value of multispectral camera to obtain spectral reflectance |
| CN106841055A (en) * | 2017-03-22 | 2017-06-13 | 浙江大学 | A kind of training sample selection method for reconstructing art drawing spectrum picture |
| CN108896176A (en) * | 2018-05-14 | 2018-11-27 | 浙江大学 | A kind of Space Consistency bearing calibration of multi-optical spectrum imaging system |
| CN109839189A (en) * | 2018-12-27 | 2019-06-04 | 浙江农林大学暨阳学院 | Utilize the method for multispectral camera self-adapting reconstruction spectral reflectance |
Non-Patent Citations (3)
| Title |
|---|
| PENG XU ET AL.: ""Self-training-based spectral image reconstruction for art paintings with multispectral imaging"", 《APPLIED OPTICS》 * |
| 徐鹏: ""基于多光谱成像技术的光谱图像重构与显示研究"", 《中国博士学位论文全文数据库信息科技辑》 * |
| 林美琪: ""面向国画的光谱重建技术的研究"", 《中国优秀硕士学位论文全文数据库基础科学辑》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111750993A (en) * | 2020-06-04 | 2020-10-09 | 武汉纺织大学 | Open measurement environment spectrum measurement method based on imaging condition correction |
| CN111750993B (en) * | 2020-06-04 | 2022-10-14 | 武汉纺织大学 | Open measurement environment spectrum measurement method based on imaging condition correction |
| CN113538285A (en) * | 2021-07-23 | 2021-10-22 | 奥比中光科技集团股份有限公司 | Method and device for determining spectral image, terminal and storage medium |
| CN113840135A (en) * | 2021-09-03 | 2021-12-24 | 大连中科创达软件有限公司 | Color cast detection method, device, equipment and storage medium |
| CN113840135B (en) * | 2021-09-03 | 2023-10-20 | 大连中科创达软件有限公司 | Color cast detection method, device, equipment and storage medium |
| CN113935894A (en) * | 2021-09-09 | 2022-01-14 | 完美世界(北京)软件科技发展有限公司 | Ink and wash style scene rendering method and equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110660112B (en) | 2021-09-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110660112B (en) | Drawing spectrum reconstruction method based on special color card and multispectral imaging | |
| CN108020519B (en) | Virtual multi-light-source spectrum reconstruction method based on color constancy | |
| KR100748844B1 (en) | Method for colour matching by means of an electronic imaging device | |
| CN111047539A (en) | Fabric image color calibration algorithm based on spectral reflectivity reconstruction | |
| CN109963133A (en) | Color calibration system and method | |
| CN108520488B (en) | Method for reconstructing spectrum and copying spectrum and electronic equipment | |
| CN108896499A (en) | In conjunction with principal component analysis and the polynomial spectral reflectance recovery method of regularization | |
| CN110926609B (en) | Spectrum reconstruction method based on sample feature matching | |
| WO2022247840A1 (en) | Light source spectrum and multispectral reflectivity image acquisition methods and apparatuses, and electronic device | |
| Liang et al. | Research on filter selection method for broadband spectral imaging system based on ancient murals | |
| JP2020012668A (en) | Evaluation device, measurement device, evaluation method and evaluation program | |
| CN112484856A (en) | Method for obtaining high-precision chromaticity and spectrum image | |
| CN112488997B (en) | Method for detecting and evaluating color reproduction of ancient painting printed matter based on characteristic interpolation | |
| Shrestha et al. | Evaluation and comparison of multispectral imaging systems | |
| CN108267426B (en) | Drawing pigment identification system and method based on multispectral imaging | |
| Finlayson et al. | Root-polynomial colour correction | |
| Guo et al. | Multispectral photometric stereo for spatially-varying spectral reflectances: A well posed problem? | |
| CN110926608A (en) | Spectrum reconstruction method based on light source screening | |
| Lin et al. | Investigating the upper-bound performance of sparse-coding-based spectral reconstruction from RGB images | |
| El-Rifai et al. | Enhanced spectral reflectance reconstruction using pseudo-inverse estimation method | |
| CN106896068B (en) | Commodity color acquisition and reproduction method and system for solving online shopping color distortion | |
| CN113259658B (en) | Method for acquiring favorite memory color of object and favorite memory color standard color card | |
| CN115964850A (en) | Operation illumination light source design method based on color commercial imaging equipment | |
| Shrestha et al. | Assessment of two fast multispectral systems for imaging of a cultural heritage artifact-a Russian icon | |
| Párraga et al. | Accurate mapping of natural scenes radiance to cone activation space: a new image dataset |
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 |