CN113792710A

CN113792710A - Spectrum reconstruction method and device and electronic equipment

Info

Publication number: CN113792710A
Application number: CN202111344528.7A
Authority: CN
Inventors: 李遂贤
Original assignee: Binzhou University
Current assignee: Binzhou University
Priority date: 2021-11-15
Filing date: 2021-11-15
Publication date: 2021-12-14

Abstract

The application discloses a spectrum reconstruction method, a spectrum reconstruction device and electronic equipment, wherein the method comprises the following steps: acquiring a human skin color database, wherein the data set comprises a training set and a verification set; acquiring RGB response values of samples in a training set and reflectivity corresponding to the RGB response values; performing polynomial expansion of different orders on the RGB response value to obtain polynomial terms corresponding to the polynomial expansion of different orders, and obtaining different numbers of eigenvectors of the reflectivity by adopting a principal component analysis method for the reflectivity; combining polynomial terms corresponding to polynomial extensions of different orders and different numbers of each eigenvector, and determining spectral reconstruction performance indexes corresponding to each order under different numbers, wherein the indexes comprise chromatic aberration and spectral error; and determining the target order and the target number of the eigenvector of the reflectivity principal component based on the index, and performing spectrum reconstruction based on the target order and the target number of the eigenvector. By adopting the embodiment of the application, the spectrum reconstruction performance is improved.

Description

Spectrum reconstruction method and device and electronic equipment

Technical Field

The present application relates to the field of computer technologies, and in particular, to a spectrum reconstruction method and apparatus, and an electronic device.

Background

With the rapid development of computer image processing technology, color images and multispectral images are increasingly applied in the fields of representation, transmission and reproduction of color information. Spectral reflectivities within certain ranges ensure that color information can be accurately represented under a variety of lighting and viewing conditions. The spectral reflectivity is the ratio of the luminous flux of the object to the luminous flux of the incident light, the object reflects the residual colored light after selectively absorbing the light source, the object is not influenced by external factors, and the color generation characteristics of the object under different light sources can be predicted according to the spectral reflectivity.

Disclosure of Invention

In order to solve the above problem, embodiments of the present application provide a spectral reconstruction method, an apparatus, and an electronic device, which improve spectral reconstruction performance.

In a first aspect, an embodiment of the present application provides a spectral reconstruction method, including:

acquiring a human skin color database, wherein the human skin color database comprises a training set and a verification set;

acquiring an original RGB response value of a sample in the training set and a reflectivity corresponding to the original RGB response value;

performing polynomial expansion of different orders on the RGB response value to obtain polynomial terms corresponding to the polynomial expansion of different orders, and obtaining different numbers of eigenvectors of main components of the reflectivity by adopting a main component analysis method for the reflectivity;

combining polynomial terms corresponding to the polynomial extensions with different orders and different numbers of the feature vectors, and determining indexes of spectrum reconstruction performance corresponding to each order under different numbers, wherein the indexes comprise chromatic aberration and spectrum error;

and determining a target order and the target number of the feature vectors of the reflectivity principal components based on the indexes, and performing spectrum reconstruction based on the target order and the target number of the feature vectors.

In a second aspect, embodiments of the present application provide a spectral reconstruction apparatus, the apparatus including:

the system comprises a data set acquisition module, a data processing module and a data processing module, wherein the data set acquisition module is used for acquiring a human skin color database, and the human skin color database comprises a training set and a verification set;

a response value obtaining module, configured to obtain an original RGB response value of the sample in the training set and a reflectivity corresponding to the original RGB response value;

the data processing module is used for performing polynomial expansion of different orders on the RGB response value to obtain polynomial terms corresponding to the polynomial expansion of different orders, and obtaining different numbers of eigenvectors of the reflectivity principal component by adopting a principal component analysis method for the reflectivity;

the index calculation module is used for combining polynomial terms corresponding to the polynomial extensions with different orders and different numbers of the feature vectors to determine indexes of spectrum reconstruction performance corresponding to each order under different numbers, wherein the indexes comprise chromatic aberration and spectrum errors;

and the spectrum reconstruction module is used for determining a target order and the target number of the feature vectors of the reflectivity principal components based on the indexes and performing spectrum reconstruction based on the target order and the target number of the feature vectors.

In a third aspect, an embodiment of the present application provides an electronic device, which may include: a processor and a memory; wherein the memory stores a computer program adapted to be loaded by the processor and to perform the method steps of the first aspect described above.

The beneficial effects brought by the technical scheme provided by some embodiments of the application at least comprise:

in the embodiment of the application, by obtaining a human skin color database containing a training set and a verification set, obtaining an original RGB response value of a sample in the training set and a reflectivity corresponding to the original RGB response value, performing polynomial expansion of different orders on the original RGB response value to obtain polynomial terms corresponding to the polynomial expansion of different orders, and obtaining different numbers of characteristic vectors of the main components of the reflectivity by adopting a principal component analysis method for the reflectivity corresponding to the original RGB response value, combining the polynomial terms corresponding to the polynomial expansion of different orders and the different numbers of the characteristic vectors, determining indexes of spectrum reconstruction performance respectively corresponding to each order under the different numbers of the characteristic vectors, wherein the indexes comprise chromatic aberration and spectrum error, determining the target order and the target number of the characteristic vectors of the main component of the reflectivity based on the indexes, and performing spectrum reconstruction based on the target order and the target number of the characteristic vectors. The method comprises the steps of obtaining polynomial terms through polynomial expansion of an original RGB response value and obtaining different numbers of characteristic vectors through principal component analysis of reflectivity corresponding to the original RGB response value, combining the polynomial expanded terms and the different numbers of the characteristic vectors, respectively calculating chromatic aberration and spectral errors of spectral reconstruction, determining the target polynomial order and the target number of the characteristic vectors according to the chromatic aberration and the spectral errors, and performing spectral reconstruction, so that the spectral reconstruction performance is improved.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings without creative efforts.

Fig. 1 is a schematic diagram of an architecture of a spectral reconstruction system according to an embodiment of the present disclosure;

fig. 2 is a schematic flowchart of a spectral reconstruction method according to an embodiment of the present disclosure;

fig. 3 is a schematic flowchart of a spectral reconstruction method according to an embodiment of the present disclosure;

fig. 4 is a schematic flowchart of a spectral reconstruction method according to an embodiment of the present disclosure;

fig. 5 is a schematic flowchart of a spectral reconstruction method according to an embodiment of the present disclosure;

fig. 6 is a schematic structural diagram of a spectrum reconstruction apparatus according to an embodiment of the present disclosure;

fig. 7 is a schematic structural diagram of a spectrum reconstruction apparatus according to an embodiment of the present disclosure;

fig. 8 is a schematic structural diagram of a spectrum reconstruction apparatus according to an embodiment of the present disclosure;

fig. 9 is a schematic structural diagram of an electronic device according to an embodiment of the present application.

Reference numerals: a spectral reconstruction device-1; a data acquisition module-11; a response value acquisition module-12; a data processing module-13; an index calculation module-14; a spectrum reconstruction module-15; a data acquisition unit-121; transformation matrix determination unit-122; a data optimization unit-123; optimization degree verifying unit-124; minimum value determining unit-151; a target order determination unit-152; average calculation unit-153; average value determining unit-154; target number determination unit-155.

Detailed Description

The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application.

In the following description, the terms "first" and "second" are used for descriptive purposes only and are not intended to indicate or imply relative importance. The following description provides embodiments of the present application, where different embodiments may be substituted or combined, and thus the present application is intended to include all possible combinations of the same and/or different embodiments described. Thus, if one embodiment includes feature A, B, C and another embodiment includes feature B, D, then this application should also be considered to include an embodiment that includes one or more of all other possible combinations of A, B, C, D, even though this embodiment may not be explicitly recited in text below.

The following description provides examples, and does not limit the scope, applicability, or examples set forth in the claims. Changes may be made in the function and arrangement of elements described without departing from the scope of the disclosure. Various examples may omit, substitute, or add various procedures or components as appropriate. For example, the described methods may be performed in an order different than the order described, and various steps may be added, omitted, or combined. Furthermore, features described with respect to some examples may be combined into other examples.

Spectral reflectance has become increasingly popular in various industrial applications as a physical attribute of human facial skin color. Including face detection in artificial intelligence, skin pigmentation prediction in the cosmetics industry, skin color modeling in computer graphics, skin color measurement in skin disease diagnosis, and skin color matching in body and maxillofacial soft tissue prostheses, the skin spectrum can also be used to predict skin chromophores, thus providing an opportunity to extract important health-related information. At present, spectrum reconstruction based on a statistical method is to reconstruct human skin reflectivity from prior knowledge of castration version of training set, and generally, device-dependent camera RGB is converted into device-dependent CIEXYZ tristimulus values, and then the camera RGB is converted into skin reflectivity by a polynomial method, or the camera RGB is directly converted into skin reflectivity by a principal component analysis method, a human skin spectrum color chart and skin reflectivity database with skin approximation is used to deduce a basis vector of a principal component, and the camera RGB is converted into skin reflectivity by the polynomial method or is deduced by the human skin color chart and skin reflectivity database, so that the reconstructed spectrum performance is low.

Based on this, the embodiment of the present application provides an entity linking method, which includes obtaining a human skin color database including a training set and a verification set, obtaining an original RGB response value of a sample in the training set and a reflectivity corresponding to the original RGB response value, performing polynomial expansion of different orders on the original RGB response value to obtain polynomial terms corresponding to the polynomial expansion of different orders, obtaining feature vectors of different numbers of reflectivity principal components by using a principal component analysis method for the reflectivity corresponding to the original RGB response value, combining the polynomial terms corresponding to the polynomial expansion of different orders and the different numbers of the feature vectors, determining indexes of spectral reconstruction performance corresponding to each order respectively under the different numbers of the feature vectors, where the indexes include color difference and spectral error, determining a target order and a target number of the feature vectors of the reflectivity principal components based on the indexes, and performing spectrum reconstruction based on the target order and the target number of the feature vectors. The method comprises the steps of obtaining polynomial terms through polynomial expansion of an original RGB response value and obtaining different numbers of characteristic vectors through principal component analysis of reflectivity corresponding to the original RGB response value, combining the polynomial expanded terms and the different numbers of the characteristic vectors, respectively calculating chromatic aberration and spectral errors of spectral reconstruction, determining the target polynomial order and the target number of the characteristic vectors according to the chromatic aberration and the spectral errors, and performing spectral reconstruction, so that the spectral reconstruction performance is improved.

Fig. 1 is a schematic diagram of a spectrum reconstruction system according to an embodiment of the present disclosure.

The system comprises a terminal and a camera, wherein the terminal can comprise but is not limited to a smart phone, a smart interactive tablet, a personal computer, a desktop computer, a tablet computer, a palm computer, a laptop computer, a computer all-in-one machine, vehicle-mounted multimedia and the like. The camera is used for randomly collecting the skin color of the forehead, the cheek, the cheekbone and the neck of the yellow or white person to form a desired image, the terminal takes the image collected by the camera as a sample of a human skin color database, a training set is randomly selected through the sample in the human skin color database, the rest samples in the database are taken as a verification set after the training set is determined, an original RGB response value in the training set and the reflectivity corresponding to the original RGB response value are obtained, polynomial terms corresponding to different orders are obtained through performing polynomial expansion of different orders on the original RGB response value, feature vectors corresponding to different numbers in reflection are obtained through a principal component analysis method on the reflectivity corresponding to the original RGB response value, the polynomial terms and the different numbers of the feature vectors are combined to determine the color difference and the spectral error corresponding to each order under different numbers, determining minimum chromatic aberration and minimum spectral error in chromatic aberration and spectral error respectively corresponding to each order under different numbers of the feature vectors, determining the order corresponding to the minimum error and the minimum spectral error as a target order of polynomial expansion, calculating a first average set of chromatic aberration and a second average set of spectral error corresponding to each order under different numbers of the feature vectors, determining a minimum first average in the first average set, determining a minimum second average in the second average set, determining the number of the feature vectors corresponding to the minimum first average and the minimum second average as the target number of the feature vectors, and performing spectral reconstruction based on the target order and the target number of the feature vectors. By performing polynomial expansion of different orders on the original RGB response values and performing principal component analysis on the reflectivity corresponding to the original RGB response values, the target orders of the polynomial expansion and the target number of the eigenvectors corresponding to the reflectivity are determined to perform spectrum reconstruction, and the spectrum reconstruction performance is improved.

Referring to fig. 2, a schematic flow chart of a spectral reconstruction method is provided in an embodiment of the present application. As shown in fig. 2, the spectral reconstruction method may include the steps of:

s101, obtaining a human skin color database, wherein the human skin color database comprises a training set and a verification set;

specifically, images corresponding to four body parts, namely the forehead, the cheek, the cheekbone and the neck of the yellow and white people are randomly collected by a camera to serve as samples in a human skin color database, a training set is randomly selected from the samples in the human skin color database, and after the training set is determined, the rest samples in the database serve as corresponding verification sets.

The number of samples in the training set is increased from 4 one by one until the samples in the training set reach half of the human skin color database, and the other half of the samples in the human skin color database are used as a verification set.

And performing polynomial expansion of different orders on the original RGB response values of the images by using the images corresponding to the samples in the training set to obtain polynomial terms of different orders, and performing dimensionality reduction on the reflectivity of the original RGB response values corresponding to the images of the samples in the training set by using a principal component analysis method to obtain different numbers of characteristic vectors of the reflectivity.

The order of the polynomial expansion may include, but is not limited to, 1 st order, 2 nd order, 3 rd order, and 4 th order, and when the order of the polynomial expansion is 1 st order, 2 nd order, 3 rd order, and 4 th order, the numbers of terms of the polynomial corresponding to the orders are 4 terms, 10 terms, 20 terms, and 35 terms, respectively.

And the samples in the verification set are used for verifying the spectral reconstruction performance obtained according to the samples in the training set.

S102, acquiring an original RGB response value of a sample in the training set and a reflectivity corresponding to the original RGB response value;

specifically, the samples in the training set are images corresponding to four body parts of the yellow and/or white people, and the original RGB response values corresponding to the samples in the training set are obtained.

The original RGB response value is determined by the reflectivity corresponding to the original RGB response value and the transformation matrix corresponding to the reflectivity.

S103, performing polynomial expansion of different orders on the original RGB response value to obtain polynomial terms corresponding to the polynomial expansion of different orders, and obtaining different numbers of eigenvectors of main components of the reflectivity by adopting a main component analysis method for the reflectivity;

specifically, polynomial expansions of different orders are performed on original RGB response values of an image corresponding to a sample in a training set to obtain polynomial terms corresponding to the polynomial expansions of different orders, and principal component analysis is performed on reflectances corresponding to the original RGB response values to obtain different numbers of feature vectors of corresponding reflectivity principal components.

And performing singular value decomposition on the reflectivity corresponding to the RGB response values of the samples in the training set to obtain the eigenvectors and the vector coefficient matrix of the reflectivity of the samples in the training set.

And arranging the eigenvalues in the principal component matrix from large to small or from small to large, wherein the number of the eigenvectors can comprise the number of eigenvectors corresponding to the eigenvalues arranged from large to small and arranged before the eigenvalues are arranged, or the number of eigenvectors corresponding to the eigenvalues arranged from small to large and arranged after the eigenvalues are arranged, and the number of the eigenvectors is equal to the number of elements in the column vector of the vector coefficient matrix.

As the number of samples in the training set increases, the elements of the column vector of the coefficient vector matrix corresponding to the reflectivity corresponding to the original RGB response value also increase. The number of elements of the column vector of the coefficient vector matrix may include, but is not limited to, 3, 4, 5, 6, 7, 8, 9, 10, and correspondingly, the number of feature vectors may include, but is not limited to, 3, 4, 5, 6, 7, 8, 9, 10.

The order of the polynomial expansion may include, but is not limited to, 1 st order, 2 nd order, 3 rd order, and 4 th order, for 1 st order polynomial expansion, a polynomial expansion may be generated to be 4 terms, for 2 nd order polynomial expansion, a polynomial expansion may be generated to be 10 terms, for 3 rd order polynomial expansion, a polynomial expansion may be generated to be 20 terms, and for 4 th order polynomial expansion, a polynomial expansion may be generated to be 35 terms.

The components of the polynomial expanded RGB response values may be:

A1= [1 R G B]；

A2= [RR GG BB RG RB GB]；

A3= [RRR GGG BBB RRG RRB GGR GGB BBR BBG RGB]；

A4= [RRRR GGGG BBBB RRRG RRRB GGGR GGGB BBBR BBBG RRGG RRBB GGBB RRGB GGRB BBRG]。

the column vectors of the polynomials of different orders after the polynomial expansion may be:

the column vector of the order 1 polynomial is

；

The column vector of the order 2 polynomial is

；

The column vector of the order 3 polynomial is

；

The column vector of the 4 th order polynomial is

；

Wherein the superscript t is the transpose of a vector or a matrix.

S104, combining polynomial terms corresponding to the polynomial extensions with different orders and different numbers of the feature vectors, and determining indexes of spectrum reconstruction performance corresponding to each order under the different numbers of the feature vectors, wherein the indexes comprise chromatic aberration and spectrum error;

specifically, in step 103, polynomial terms corresponding to polynomial extensions of different orders are obtained by performing polynomial extension of different orders on original RGB response values of samples in the training set, different numbers of eigenvectors of the principal component of the reflectivity are obtained by performing singular value decomposition on the reflectivity corresponding to the original RGB response values, and the indexes of spectral reconstruction performance corresponding to each polynomial extension order under different numbers of eigenvectors are determined after combining the polynomial terms and the different numbers of eigenvectors, where the indexes of spectral reconstruction performance include color difference and spectral error.

Where the color difference is denoted by DELab, which is a color difference based on CIE colorimetry, assuming a standard observer (CIE1931 standard 2 degree observer) and a given lighting condition, DELab (CIELAB color difference) is used in this case to predict the 2-norm distance between the LAB coordinates of the skin spectrum and the measured skin spectrum.

The spectral color difference is expressed in RMSE, which only takes into account the difference in spectral consistency between the predicted spectrum and the measured spectrum, which is not affected by the lamp.

For example, the order of the polynomial expansion may include, but is not limited to, 1 st order, 2 nd order, 3 rd order, and 4 th order, and when the order of the polynomial expansion is 1 st order, 2 nd order, 3 rd order, and 4 th order, the numbers of terms of the polynomial corresponding to the polynomial expansion are 4 terms, 10 terms, 20 terms, and 35 terms, respectively.

The number of feature vectors may include, but is not limited to, 3, 4, 5, 6, 7, 8, 9, 10.

And traversing and combining 3-term, 4-term, 10-term, 20-term and 35-term polynomials corresponding to 0-order, 1-order, 2-order, 3-order and 4-order polynomial expansion and 3, 4, 5, 6, 7, 8, 9 and 10 feature vectors respectively to form 40 combinations, and determining the color difference and the spectral error corresponding to each combination.

And S105, determining a target order and the target number of the feature vectors of the reflectivity principal components based on the indexes, and performing spectrum reconstruction based on the target order and the target number of the feature vectors.

Specifically, as in step 104, after combining the polynomial terms corresponding to the polynomial extensions of different orders with different numbers of eigenvectors, determining the color difference and spectral error respectively corresponding to each combination, comparing the color difference and spectral error of each combination respectively, determining the minimum color difference and the minimum spectral error, determining the target order of the polynomial extension according to the polynomial terms corresponding to the minimum color difference and the minimum spectral error, and determining the target number of eigenvectors according to the number of eigenvectors corresponding to the minimum color difference and the minimum spectral error.

And performing spectrum reconstruction according to the determined target order of polynomial expansion and the target number of the eigenvectors.

Referring to fig. 3, a schematic flow chart of a spectral reconstruction method is provided in an embodiment of the present application. As shown in fig. 3, the spectral reconstruction method may include the steps of:

s201, acquiring a human skin color data set, wherein the human skin color data set comprises a training set and a verification set;

the step 101 may be referred to in the detailed implementation of this step, and details are not described herein.

S202, acquiring an original RGB response value and a reflectivity corresponding to the original RGB response value; determining a reflectivity transformation matrix based on the original RGB response value and the reflectivity corresponding to the original RGB response value;

specifically, the original RGB response values of the samples in the training set are determined by the reflectivities corresponding to the original RGB response values and the transformation matrices corresponding to the reflectivities.

In a specific mode, the method comprises the following steps of,

，

to train the response matrix of the raw RGB response values of the samples in the set,

m is a transformation matrix of the reflectivity for responding to the corresponding reflectivity of the matrix.

S203, optimizing the reflectivity sample set;

specifically, the optimization of the transformation matrix is realized by changing the reflectivity sample set corresponding to the original RGB response value, so that the mathematical ill-conditioned degree of the transformation matrix M is reduced.

In a specific mode, the method comprises the following steps of,

，

for the pseudo-inverse matrix after the optimization of M,

is composed of

An RGB response matrix of the original RGB response values,

the corresponding reflectivity for the response matrix;

s204, verifying the optimization degree of the transformation matrix;

specifically, the optimization degree of the transformation matrix needs to be verified through the reflection spectrum matrix in the verification set and the RGB response value matrix corresponding to the samples in the verification set.

In a specific mode, the method comprises the following steps of,

，

for the pseudo-inverse matrix after the optimization of M,

to verify the concentrated reflection spectrum matrix,

to verify the corresponding RGB response value matrix of the samples in the set.

S205, performing polynomial expansion of different orders on the original RGB response value to obtain polynomial terms corresponding to the polynomial expansion of different orders, and obtaining different numbers of eigenvectors of main components of reflectivity by adopting a main component analysis method for the reflectivity;

the step 103 may be referred to in the detailed implementation manner of this step, and details are not described herein.

S206, combining the polynomial terms corresponding to the polynomial extensions with different orders and different numbers of the feature vectors, and determining indexes of spectrum reconstruction performance corresponding to each order under the different numbers of the feature vectors, wherein the indexes comprise chromatic aberration and spectrum error;

the step 104 may be referred to in the present embodiment, and is not described herein again.

And S207, determining a target order and the target number of the feature vectors of the reflectivity principal components based on the indexes, and performing spectrum reconstruction based on the target order and the target number of the feature vectors.

The step 105 may be referred to in the present embodiment, and is not described herein again.

In the embodiment of the application, the original RGB response value and the reflectivity corresponding to the original RGB response value are obtained by obtaining a human skin color database containing a training set and a verification set, a reflectivity transformation matrix is determined based on the original RGB response value and the reflectivity corresponding to the original RGB response value, the reflectivity transformation matrix is determined by polynomial expansion and reflectivity principal components, polynomial expansion corresponding to the polynomial expansion of different orders is obtained by performing polynomial expansion of different orders on the original RGB response value, eigenvectors of different numbers of reflectivity principal components are obtained by adopting a principal component analysis method for the reflectivity corresponding to the original RGB response value, the polynomial expansion corresponding to the polynomial expansion of different orders and the different numbers of eigenvectors are combined, and the index of spectral reconstruction performance corresponding to each order under the different numbers of eigenvectors is determined, the indexes comprise chromatic aberration and spectral errors, the target orders and the target number of the feature vectors of the reflectivity principal components are determined based on the indexes, and spectral reconstruction is carried out based on the target orders and the target number of the feature vectors. The method comprises the steps of obtaining polynomial terms through polynomial expansion of an original RGB response value and conducting principal component analysis on reflectivity corresponding to the original RGB response value to obtain different numbers of eigenvectors, combining the polynomial expanded terms and the different numbers of eigenvectors, respectively calculating chromatic aberration and spectral error of spectral reconstruction, determining target polynomial orders and target numbers of eigenvectors according to the chromatic aberration and the spectral error to conduct spectral reconstruction, and optimizing the reflectivity corresponding to the original RGB response value, so that the mathematical ill-conditioned degree of a transformation matrix is reduced, and the spectral reconstruction performance is improved.

Referring to fig. 4, a schematic flow chart of a spectral reconstruction method according to an embodiment of the present application is provided. As shown in fig. 4, the spectral reconstruction method may include the steps of:

s301, acquiring a human skin color data set, wherein the human skin color data set comprises a training set and a verification set;

S302, acquiring an original RGB response value of the sample in the training set and a reflectivity corresponding to the original RGB response value;

the step 102 may be referred to in the detailed implementation of this step, and details are not described here.

S303, performing polynomial expansion of different orders on the RGB response value to obtain the number of terms corresponding to the polynomial expansion of different orders;

specifically, polynomial expansion of different orders is performed on original RGB response values of images corresponding to samples in a training set, and polynomial terms corresponding to each order are obtained according to the polynomial expansion of different orders.

The order of the polynomial expansion may include, but is not limited to, 1 st order, 2 nd order, 3 rd order, and 4 th order.

After 1-order, 2-order, 3-order and 4-order polynomial expansion is carried out on the original RGB response value, the obtained polynomial terms are 4 terms, 10 terms, 20 terms and 35 terms respectively.

In a specific mode, the method comprises the following steps of,

，

for RGB response value after expanding the polynomial, M is a transformation matrix of reflectivity, and i is the number of terms of polynomial expansion;

s304, performing singular value decomposition on the reflectivity corresponding to the original RGB response to obtain different numbers of eigenvectors of the reflectivity principal component;

specifically, singular value decomposition is performed on the reflectivity corresponding to the RGB response values of the samples in the training set to obtain the eigenvectors and the vector coefficient matrix of the reflectivity of the samples in the training set.

As the number of samples in the training set increases, the number of elements of the column vector of the coefficient vector matrix of the reflectivity corresponding to the original RGB response value also increases, and thus the number of the feature vectors of the reflectivity also increases, for example, the number of elements of the column vector of the coefficient vector matrix may include, but is not limited to, 3, 4, 5, 6, 7, 8, 9, 10, and correspondingly, the number of the feature vectors may include, but is not limited to, 3, 4, 5, 6, 7, 8, 9, 10.

In a specific mode, the method comprises the following steps of,

，

the reflectivity for the original RGB response value,

is a matrix of the principal components,

each row of vectors of (a) is a eigenvector of the singular value decomposition reflectivity, alpha is a coefficient vector matrix, j is the number of eigenvectors corresponding to larger eigenvalues contained in the columns of the principal component matrix, and the size of j is equal to the element number of the column vectors in the coefficient vector matrix.

S305, combining the polynomial terms corresponding to the polynomial extensions with different orders and different numbers of the feature vectors, and determining indexes of spectrum reconstruction performance corresponding to each order under the different numbers of the feature vectors, wherein the indexes comprise chromatic aberration and spectrum error;

the detailed implementation of this step can refer to step 104, which is not described herein again

S306, determining a target order and a target number of the feature vectors of the reflectivity principal components based on the indexes, and performing spectrum reconstruction based on the target order and the target number of the feature vectors.

In the embodiment of the application, the original RGB response value and the reflectivity corresponding to the original RGB response value are obtained by obtaining a human skin color database comprising a training set and a verification set, a reflectivity transformation matrix is determined based on the original RGB response value and the reflectivity corresponding to the original RGB response value, the reflectivity transformation matrix is determined by polynomial expansion and reflectivity principal components, polynomial expansion of different orders is carried out on the original RGB response value to obtain polynomial terms corresponding to the polynomial expansion of different orders, singular value decomposition is carried out on the reflectivity corresponding to the original RGB response value to obtain eigenvectors of different numbers of the reflectivity principal components, the polynomial terms corresponding to the polynomial expansion of different orders and the different numbers of the eigenvectors are combined to determine indexes of spectral reconstruction performance corresponding to each order under the different numbers of the eigenvectors, the indexes comprise chromatic aberration and spectral errors, the target orders and the target number of the feature vectors of the reflectivity principal components are determined based on the indexes, and spectral reconstruction is carried out based on the target orders and the target number of the feature vectors. The method comprises the steps of obtaining polynomial terms by performing polynomial expansion on an original RGB response value, obtaining a characteristic vector and a coefficient vector matrix by performing singular value decomposition on reflectivity corresponding to the original RGB response value, wherein the number of the characteristic vectors is the same as the number of elements of a column vector of the coefficient vector matrix, increasing the number of the column vector elements of the coefficient vector of the reflectivity corresponding to the original RGB response value with the increase of the sample number of a training set, generating different numbers of characteristic vectors, combining the terms after the polynomial expansion and the different numbers of the characteristic vectors, calculating chromatic aberration and spectral error of spectral reconstruction respectively, determining the order of a target polynomial and the target number of the characteristic vectors according to the chromatic aberration and the spectral error, and performing spectral reconstruction, so that the spectral reconstruction performance is improved.

Referring to fig. 5, a schematic flow chart of a spectral reconstruction method according to an embodiment of the present application is provided. As shown in fig. 5, the spectral reconstruction method may include the steps of:

s401, acquiring a human skin color data set, wherein the human skin color data set comprises a training set and a verification set;

S402, acquiring an original RGB response value of the sample in the training set and a reflectivity corresponding to the original RGB response value;

S403, performing polynomial expansion of different orders on the original RGB response value to obtain polynomial terms corresponding to the polynomial expansion of different orders, and obtaining different numbers of eigenvectors of main components of reflectivity by adopting a main component analysis method for the reflectivity;

S404, combining polynomial terms corresponding to the polynomial extensions with different orders and different numbers of the feature vectors, and determining indexes of spectrum reconstruction performance corresponding to each order under the different numbers of the feature vectors, wherein the indexes comprise chromatic aberration and spectrum error;

S405, determining the minimum chromatic aberration in the chromatic aberration respectively corresponding to each order of the feature vectors with different numbers, and determining the minimum spectral error in the spectral errors respectively corresponding to each order of the feature vectors with different numbers; taking the order corresponding to the minimum chromatic aberration and the minimum spectral error as a target order;

specifically, different numbers of feature vectors and the order generated after polynomial expansion are combined, the chromatic aberration and the spectral error of each combination are determined, the minimum chromatic aberration and the minimum spectral error are respectively determined in all the combined chromatic aberration and spectral errors, the polynomial terms corresponding to the minimum chromatic aberration and the minimum spectral error are determined, and the order of the polynomial expansion corresponding to the polynomial terms is determined as the target order.

For example, the number of the feature vectors is 3, 4, 5, 6, 7, 8, 9, 10, and the polynomial terms with the order of 0, 1, 2, 3, 4 of the polynomial expansion are 3, 4, 10, 20, 35.

Combining 3, 4, 5, 6, 7, 8, 9, 10 and 3, 4, 10, 20, 35 to generate 40 different combinations, calculating and determining the color difference and the spectral error corresponding to each combination, determining the polynomial term number corresponding to the minimum color difference from the 40 color differences, determining the polynomial term number corresponding to the minimum spectral error from the 40 spectral errors, determining the order of polynomial expansion according to the polynomial term number corresponding to the minimum color difference and the minimum spectral error, and determining the order of polynomial expansion as the target order.

S406, calculating a first average set of chromatic aberration corresponding to each order under different numbers of the feature vectors, and calculating a second average set of spectral error corresponding to each order under different numbers of the feature vectors; determining a minimum first mean set among the first means, and determining a minimum second mean among the second mean set; taking the number of the feature vectors corresponding to the minimum first average value and the minimum second average value as the target number of the feature vectors;

specifically, different numbers of feature vectors are combined with orders generated after polynomial expansion, color difference and spectral error of each combination are determined, a first average set corresponding to the color difference of each number of the feature vectors under each order is calculated, a second average set corresponding to the spectral error of each number of the feature vectors under each order is calculated, a minimum first average value is determined in the first average set, a minimum second average value is determined in the second average set, the numbers of the feature vectors corresponding to the minimum first average value and the minimum second average value are determined, and the number of the feature vectors is determined as a target number.

For example, the number of the feature vectors is 3, 4, and 5, and the polynomial terms corresponding to the order of the polynomial expansion being 0, 1, 2, 3, and 4 are 3, 4, 10, 20, and 35.

Combining 3, 4, 5 and 3, 4, 10, 20, 35 yields 15 different combinations, each of which is calculated to determine the color difference and the spectral error for each combination.

When the number of the feature vectors is 3, color difference and spectral error corresponding to items with polynomial terms of 3, 4, 10, 20 and 35 are respectively obtained, and average values of the obtained color difference and spectral error are respectively calculated to obtain a color difference average value and a spectral error average value when the number of the feature vectors is 3.

When the number of the feature vectors is 4, color difference and spectral error corresponding to items with polynomial terms of 3, 4, 10, 20 and 35 are respectively obtained, and average values of the obtained color difference and spectral error are respectively calculated to obtain a color difference average value and a spectral error average value when the number of the feature vectors is 4.

When the number of the feature vectors is 5, color difference and spectral error corresponding to items with polynomial terms of 3, 4, 10, 20 and 35 are respectively obtained, and average values of the obtained color difference and spectral error are respectively calculated to obtain a color difference average value and a spectral error average value when the number of the feature vectors is 5.

And when the number of the contrast characteristic vectors is 3, 4 and 5, calculating the obtained color difference and the spectrum error, determining the minimum color difference average value and the spectrum error average value from the color difference and the spectrum error, and determining the number of the characteristic vectors corresponding to the color difference average value and the spectrum error average value as the target number.

S407, spectrum reconstruction is carried out on the basis of the target order and the target number of the feature vectors.

In the embodiment of the application, the original RGB response value and the reflectivity corresponding to the original RGB response value are obtained by obtaining a human skin color database comprising a training set and a verification set, a reflectivity transformation matrix is determined based on the original RGB response value and the reflectivity corresponding to the original RGB response value, the reflectivity transformation matrix is determined by polynomial expansion and reflectivity principal components, polynomial expansion of different orders is carried out on the original RGB response value to obtain polynomial terms corresponding to the polynomial expansion of different orders, singular value decomposition is carried out on the reflectivity corresponding to the original RGB response value to obtain eigenvectors of different numbers of the reflectivity principal components, the polynomial terms corresponding to the polynomial expansion of different orders and the different numbers of the eigenvectors are combined to determine indexes of spectral reconstruction performance corresponding to each order under the different numbers of the eigenvectors, the indexes include chromatic aberration and spectral error, chromatic aberration and spectral error corresponding to each order under different numbers of a feature vector are determined, the minimum chromatic aberration and the minimum spectral error and the polynomial term number corresponding to the minimum chromatic aberration and the minimum spectral error are determined in all combined chromatic aberration and spectral error, the order of polynomial expansion corresponding to the polynomial term number is determined as a target order, a first average set of chromatic aberration corresponding to each order under different numbers of the feature vector is calculated, a second average set of spectral error corresponding to each order under different numbers of the feature vector is calculated, the minimum first average set is determined in the first average, the minimum second average set is determined in the second average set, the minimum first average set and the number of the feature vector corresponding to the minimum second average are used as the target number of the feature vector, and performing spectrum reconstruction based on the target order and the target number of the feature vectors. The method comprises the steps of determining the color difference of each order of the feature vectors under different numbers and the minimum color difference and the minimum spectrum error in the spectrum errors, determining the polynomial expansion order corresponding to the minimum color difference and the minimum spectrum error as a target order, determining the average value of the minimum color difference and the average value of the minimum spectrum error in the average values of the color difference and the spectrum errors corresponding to the orders under different numbers, determining the number of the feature vectors corresponding to the average value of the minimum color difference and the average value of the minimum spectrum error as a target number, and carrying out spectrum reconstruction based on the target order of the polynomial expansion and the target number of the feature vectors, thereby improving the spectrum reconstruction performance.

The spectral reconstruction device provided by the embodiment of the present application will be described in detail below with reference to fig. 6. It should be noted that. The spectrum reconstruction apparatus in fig. 6 is used for executing the method of the embodiment shown in fig. 2 to 5, and for convenience of illustration, only the relevant portions of the embodiment are shown, and for the purpose of disclosure, please refer to the method embodiment shown in fig. 2 to 5.

Fig. 6 is a schematic structural diagram of a spectrum reconstruction apparatus according to the present application. As shown in fig. 6, the spectrum reconstruction apparatus 1 according to the embodiment of the present application includes: the system comprises a database acquisition module 11, a response value acquisition module 12, a data processing module 13, an index calculation module 14 and a spectrum reconstruction module 15.

A data set obtaining module 11, configured to obtain a human skin color data set, where the human skin color data set includes a training set and a verification set;

a response value obtaining module 12, configured to obtain an original RGB response value of the sample in the training set and a reflectivity corresponding to the original RGB response value;

the data processing module 13 is configured to perform polynomial expansion of different orders on the RGB response values to obtain polynomial terms corresponding to the polynomial expansion of different orders, and obtain different numbers of eigenvectors of principal components of reflectivity by using a principal component analysis method for the reflectivity;

an index calculation module 14, configured to combine polynomial terms corresponding to the polynomial extensions of different orders and different numbers of each feature vector, and determine an index of spectrum reconstruction performance corresponding to each order under different numbers, where the index includes a color difference and a spectrum error;

and a spectrum reconstruction module 15, configured to determine a target order and a target number of the feature vectors of the reflectivity principal component based on the index, and perform spectrum reconstruction based on the target order and the target number of the feature vectors.

Optionally, as shown in fig. 7, the response value obtaining module 12 includes:

a data obtaining unit 121, configured to obtain an original RGB response value and a reflectivity corresponding to the original RGB response value;

a transformation matrix determining unit 122, configured to determine a transformation matrix of the reflectivity based on the response value of the original RGB and the reflectivity corresponding to the response value of the original RGB;

wherein the content of the first and second substances,

said

A response matrix for the raw RGB response values of the samples in the training set, the

The reflectivity corresponding to the response matrix, and M is a transformation matrix of the reflectivity;

a data optimization unit 123, configured to optimize the reflectivity sample set;

wherein the content of the first and second substances,

said

For the M optimized pseudo inverse matrix, said

Is that it is

An RGB response matrix of raw RGB response values of, the

The corresponding reflectivity for the response matrix;

an optimization degree verifying unit 124 for verifying the optimization degree of the transformation matrix;

wherein the content of the first and second substances,

said

For the M optimized pseudo inverse matrix, the

To validate a concentrated reflectance spectrum matrix, the

Optionally, the data processing module 13 is specifically configured to:

said

For the RGB response value after the expansion of the polynomial, M is a transformation matrix of the reflectivity, and i is the number of terms of the polynomial expansion.

Optionally, the data processing module 13 is specifically configured to:

obtaining different numbers of eigenvectors of the main components of the reflectivity by performing singular value decomposition on the reflectivity corresponding to the original RGB response;

wherein the content of the first and second substances,

said

Reflectivity corresponding to the original RGB response values, said

Is a principal component matrix, said

The vectors of each row of the method are eigenvectors of the reflectivity decomposed by singular values, alpha is a coefficient vector, j is the number of eigenvectors, and j is equal to the element number of the column vector in the alpha.

Optionally, as shown in fig. 8, the spectrum reconstruction module 15 includes:

a minimum index determining unit 151, configured to determine a minimum color difference from color differences corresponding to each order under different numbers of the feature vectors, and determine a minimum spectral error from spectral errors corresponding to each order under different numbers of the feature vectors;

a target order determining unit 152, configured to use the minimum color difference and the order corresponding to the minimum spectral error as a target order;

an average value calculating unit 153, configured to calculate a first average value of color differences corresponding to each order under different numbers of the feature vectors, and calculate a second average value of spectral errors corresponding to each order under different numbers of the feature vectors;

an average value determining unit 154 for determining a minimum first average value among the first average values and determining a minimum second average value among the second average values;

the target number determining unit 155 is configured to use the number of feature vectors corresponding to the minimum first average value and the minimum second average value as the target number of feature vectors.

Fig. 9 is a schematic structural diagram of an electronic device according to an embodiment of the present application. As shown in fig. 9, the terminal device may include: the system comprises at least one processor, at least one network interface, at least one input/output interface, at least one display unit, at least one memory and at least one communication bus. The processor may include one or more processing cores, among others. The processor connects various parts within the overall electronic device using various interfaces and lines, performs various functions of the terminal and processes data by executing or executing instructions, programs, code sets, or instruction sets stored in the memory, and calling data stored in the memory. The memory may be a high-speed RAM memory or a non-volatile memory (e.g., at least one disk memory). The memory may optionally be at least one memory device located remotely from the processor. The network interface may optionally include a standard wired interface or a wireless interface (e.g., WI-FI interface or bluetooth interface). The communication bus is used to enable connection communication between these components. The display unit may be a touch panel. As shown in fig. 9, the memory as a storage medium may include therein an operating system, a network communication module, an input-output interface module, and a spectrum reconstruction program.

In the electronic device shown in fig. 9, the input/output interface is mainly used for providing an interface for the user and the access device, and acquiring data input by the user and the access device.

In one embodiment, the processor may be configured to invoke an entity linker stored in the memory and specifically perform the following operations:

acquiring a human skin color data set, wherein the human skin color data set comprises a training set and a verification set;

performing polynomial expansion of different orders on the original RGB response value to obtain polynomial terms corresponding to the polynomial expansion of different orders, and obtaining different numbers of eigenvectors of main components of the reflectivity by adopting a main component analysis method for the reflectivity;

combining polynomial terms corresponding to the polynomial extensions with different orders and different numbers of the feature vectors to determine indexes of spectrum reconstruction performance corresponding to each order under the different numbers of the feature vectors, wherein the indexes comprise chromatic aberration and spectrum error;

In an embodiment, when the processor obtains an original RGB response value of a sample in the training set and a reflectivity corresponding to the original RGB response value, the following operations are specifically performed:

acquiring an original RGB response value and a reflectivity corresponding to the original RGB response value;

determining a transformation matrix of the reflectivity based on the response value of the original RGB and the reflectivity corresponding to the response value of the original RGB;

wherein the content of the first and second substances,

said

optimizing the reflectivity;

wherein the content of the first and second substances,

said

For the M optimized pseudo inverse matrix, said

Is that it is

An RGB response matrix of raw RGB response values of, the

The corresponding reflectivity for the response matrix;

verifying the optimization degree of the transformation matrix;

wherein the content of the first and second substances,

said

Pseudo-inverse optimized for said MA matrix of

To validate a concentrated reflectance spectrum matrix, the

In an embodiment, when performing polynomial expansion of different orders on the original RGB response value to obtain polynomial terms corresponding to the polynomial expansion of different orders, the processor specifically performs the following operations:

said

In one embodiment, when the processor performs principal component analysis on the reflectivity to obtain different numbers of eigenvectors of the reflectivity principal component, the following operations are specifically performed:

wherein the content of the first and second substances,

said

Reflectivity corresponding to the original RGB response values, said

Is a principal component matrix, said

The vector of each row of the matrix is a characteristic vector of the reflectivity decomposed by singular values, the alpha is a coefficient vector matrix, the j is the number of the characteristic vectors, and the j is equal to the element number of the column vector in the alpha.

In one embodiment, the processor, when performing the determining of the target order and the target number of the feature vectors of the reflectivity principal components based on the index, specifically performs the following operations:

determining the minimum chromatic aberration in the chromatic aberration corresponding to each order under different numbers, and determining the minimum spectral error in the spectral error corresponding to each order under different numbers;

taking the order corresponding to the minimum chromatic aberration and the minimum spectral error as a target order;

calculating a first average value of chromatic aberration corresponding to each order under different numbers, and calculating a second average value of spectral error corresponding to each order under different numbers;

determining a minimum first average value among the first average values, and determining a minimum second average value among the second average values;

and taking the number of the feature vectors corresponding to the minimum first average value and the minimum second average value as the target number of the feature vectors.

It is clear to a person skilled in the art that the solution of the present application can be implemented by means of at least one of software and hardware. The "unit" and "module" in this specification refer to at least one of software and hardware, such as a Field-ProgrammaBLE Gate Array (FPGA), an Integrated Circuit (IC), and the like, capable of performing a specific function independently or in cooperation with other components.

It should be noted that, for simplicity of description, the above-mentioned method embodiments are described as a series of acts or combination of acts, but those skilled in the art will recognize that the present application is not limited by the order of acts described, as some steps may occur in other orders or concurrently depending on the application. Further, those skilled in the art should also appreciate that the embodiments described in the specification are preferred embodiments and that the acts and modules referred to are not necessarily required in this application.

In the foregoing embodiments, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.

In the embodiments provided in the present application, it should be understood that the disclosed apparatus may be implemented in other manners. For example, the above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one type of division of logical functions, and there may be other divisions when actually implementing, for example, a plurality of units or components may be combined or may be integrated into another system, or some features may be omitted, or not implemented. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of some service interfaces, devices or units, and may be an electrical or other form.

The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.

The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable memory. Based on such understanding, the technical solution of the present application may be substantially implemented or contributed to by the prior art, or all or part of the technical solution may be embodied in a software product, which is stored in a memory and includes instructions for causing an electronic device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned memory comprises: various media capable of storing program codes, such as a usb disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic disk, or an optical disk.

Those skilled in the art will appreciate that all or part of the steps in the methods of the above embodiments may be implemented by a program, which is stored in a computer-readable memory, and the memory may include: flash disks, Read-Only memories (ROMs), Random Access Memories (RAMs), magnetic or optical disks, and the like.

The above description is only an exemplary embodiment of the present disclosure, and the scope of the present disclosure should not be limited thereby. That is, all equivalent changes and modifications made in accordance with the teachings of the present disclosure are intended to be included within the scope of the present disclosure. Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure herein. This application is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.

Claims

1. A method of spectral reconstruction, the method comprising:

2. The method of claim 1, wherein obtaining the raw RGB response values of the samples in the training set and the reflectivities corresponding to the raw RGB response values comprises:

determining a reflectivity transformation matrix based on the original RGB response value and the reflectivity corresponding to the original RGB response value;

wherein the content of the first and second substances,

said

optimizing the set of reflectance samples;

wherein the content of the first and second substances,

said

For the M optimized pseudo inverse matrix, said

Is that it is

An RGB response matrix of raw RGB response values of, the

The corresponding reflectivity for the response matrix;

verifying the optimization degree of the transformation matrix;

wherein the content of the first and second substances,

said

For the M optimized pseudo inverse matrix, the

To validate a concentrated reflectance spectrum matrix, the

3. The method as claimed in claim 1, wherein the performing polynomial expansion of different orders on the RGB response values to obtain the number of terms corresponding to the polynomial expansion of different orders includes:

said

4. The method of claim 3, wherein said obtaining different numbers of eigenvectors of the principal component of reflectance from said reflectance using principal component analysis comprises:

wherein the content of the first and second substances,

said

Reflectivity corresponding to the original RGB response values, said

Is a principal component matrix, said

5. The method of claim 1, wherein determining a target order and a target number of eigenvectors for the reflectivity principal component based on the index comprises:

determining the minimum chromatic aberration in the chromatic aberration corresponding to each order under different numbers of the eigenvectors, and determining the minimum spectral error in the spectral error corresponding to each order under different numbers of the eigenvectors;

calculating a first average value of chromatic aberration corresponding to each order under different numbers of the feature vectors, and calculating a second average value of spectral error corresponding to each order under different numbers of the feature vectors;

6. A spectral reconstruction apparatus, characterized in that said apparatus comprises:

7. The apparatus of claim 6, wherein the data acquisition module comprises:

the data acquisition unit is used for acquiring an original RGB response value and the reflectivity corresponding to the original RGB response value;

the conversion matrix determining unit is used for determining a reflectivity conversion matrix based on the original RGB response value and the reflectivity corresponding to the original RGB response value;

wherein the content of the first and second substances,

said

a data optimization unit for optimizing the reflectance sample set;

wherein the content of the first and second substances,

said

For the M optimized pseudo inverse matrix, said

Is that it is

An RGB response matrix of raw RGB response values of, the

The corresponding reflectivity for the response matrix;

the optimization degree verifying unit is used for verifying the optimization degree of the transformation matrix;

wherein the content of the first and second substances,

said

For the M optimized pseudo inverse matrix, the

To validate a concentrated reflectance spectrum matrix, the

8. The apparatus according to claim 6, wherein the data processing module is specifically configured to:

said

9. The apparatus according to claim 6, wherein the data processing module is specifically configured to:

wherein the content of the first and second substances,

said

Reflectivity corresponding to the original RGB response values, said

Is a principal component matrix, said

10. An electronic device, comprising: a processor and a memory; wherein the memory stores a computer program adapted to be loaded by the processor and to perform the method steps of any of claims 1 to 5.