CN109632761B - Processing method and system of Raman spectrum data - Google Patents
Processing method and system of Raman spectrum data Download PDFInfo
- Publication number
- CN109632761B CN109632761B CN201811533229.6A CN201811533229A CN109632761B CN 109632761 B CN109632761 B CN 109632761B CN 201811533229 A CN201811533229 A CN 201811533229A CN 109632761 B CN109632761 B CN 109632761B
- Authority
- CN
- China
- Prior art keywords
- raman
- spectrum data
- data
- raman spectrum
- processing
- 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.)
- Active
Links
- 238000001237 Raman spectrum Methods 0.000 title claims abstract description 131
- 238000003672 processing method Methods 0.000 title claims abstract description 7
- 238000001069 Raman spectroscopy Methods 0.000 claims abstract description 133
- 238000001228 spectrum Methods 0.000 claims abstract description 110
- 230000003595 spectral effect Effects 0.000 claims abstract description 34
- 238000000034 method Methods 0.000 claims abstract description 29
- 239000011159 matrix material Substances 0.000 claims description 84
- 238000004364 calculation method Methods 0.000 claims description 31
- 230000006870 function Effects 0.000 claims description 16
- 230000003416 augmentation Effects 0.000 claims description 15
- 230000009466 transformation Effects 0.000 claims description 9
- 238000010606 normalization Methods 0.000 claims description 6
- 238000011156 evaluation Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 abstract description 4
- 238000001514 detection method Methods 0.000 description 9
- 230000003190 augmentative effect Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 230000003321 amplification Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 239000011324 bead Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 239000002131 composite material Substances 0.000 description 2
- 238000006073 displacement reaction Methods 0.000 description 2
- 238000003199 nucleic acid amplification method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 229940079593 drug Drugs 0.000 description 1
- 239000007791 liquid phase Substances 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 239000012071 phase Substances 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/62—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light
- G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
- G01N21/65—Raman scattering
Landscapes
- Health & Medical Sciences (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating, Analyzing Materials By Fluorescence Or Luminescence (AREA)
Abstract
The invention discloses a processing method and a system of Raman spectrum data, wherein the method comprises the following steps: collecting first Raman spectrum data of a test sample through a Raman spectrometer; background removing processing is carried out on the first Raman spectrum data to obtain second Raman spectrum data; and identifying the components in the test sample according to the Raman spectrum database of the standard sample and the second Raman spectrum data to obtain an identification result. According to the method, background removal processing is carried out on the first Raman spectrum data of the test sample to obtain second Raman spectrum data, then components in the test sample are identified according to the Raman spectrum database of the standard sample and the second Raman spectrum data to obtain an identification result, and through the background removal processing, the influence of a background spectrum on the test result in the identification process is avoided, and the accuracy of the identification result is ensured. The invention can be widely applied to the field of spectral data processing.
Description
Technical Field
The invention relates to the field of spectral data processing, in particular to a Raman spectral data processing method and system.
Background
The Raman spectrum analysis technology is characterized in that a tested sample is stimulated by laser, the tested sample is stimulated to radiate Raman signals, and due to different internal molecular structures of different substances, the sample can be stimulated to generate a specific Raman scattering spectrum, so that the molecular structure and the composition can be researched by analyzing the Raman signals. With the progress of laser technology and photoelectric detection technology, the application of raman technology in the field of component detection is more and more extensive. In recent years, the domestic strength of food and drug safety supervision is getting bigger and bigger, and compared with the traditional gas-phase liquid-phase analysis equipment, the Raman spectrum detection equipment has the advantages of short detection period, low cost and convenient carrying when being used for component detection. However, because the raman signal is very weak, especially the laser energy of the small and light-weight device is weak, the collection capability of the raman signal to the scattered light is also limited, so that the raman signal obtained by the detection device is easily interfered by the background spectrum, noise and stray light, and if the data collected by the detector is directly processed and analyzed, the accuracy of component judgment on the test sample can be affected.
Disclosure of Invention
To solve the above technical problems, the present invention aims to: provided are a processing method and system capable of effectively detecting whether a test sample contains a Raman spectrum data of a standard sample component.
The first technical scheme adopted by the invention is as follows: a method of processing raman spectral data, comprising the steps of:
collecting first Raman spectrum data of a test sample through a Raman spectrometer;
background removing processing is carried out on the first Raman spectrum data to obtain second Raman spectrum data;
and identifying the components in the test sample according to the Raman spectrum database of the standard sample and the second Raman spectrum data to obtain an identification result.
Further, the background removing processing is performed on the first raman spectrum data to obtain second raman spectrum data, which specifically includes:
background removing processing is carried out on the first Raman spectrum data through a high-pass filter matrix and a penalty function to obtain second Raman spectrum data, and the calculation formula of the second Raman spectrum data is as follows:
wherein,
is the second Raman spectrum data, y is the first Raman spectrum data, x is the iterative spectrum data, F (x) is the evaluation function, H is the finite field high pass filter matrix, phi is the penalty function, lambda is the weight coefficient, DiIs the i-th order differential, M is the total order of the differential, NiIs the i-th order differential data length.
Further, the finite field high-pass filter matrix is obtained by the following steps:
acquiring a first differential operator and a truncation operator;
obtaining a first augmentation matrix and a second augmentation matrix according to the first differential operator and the truncation operator;
acquiring an extension mode transformation matrix;
obtaining a truncation matrix according to the first augmentation matrix and the extension mode transformation matrix;
obtaining a differential matrix according to the second augmentation matrix and the extension mode transformation matrix;
and obtaining a finite field high-pass filter matrix according to the truncation matrix and the differential matrix.
Further, before background removing processing is carried out on the first Raman spectrum data, the method further comprises the following steps:
calculating the minimum resolution of the wave number according to the wavelength resolution of the Raman spectrometer to obtain a wave number division value, wherein the calculation formula of the wave number division value is as follows:
wherein k isΔIs the division value of wave number, lambdamaxIs the maximum of the measurable wavelength, λ, in the first Raman spectral dataΔIs the wavelength resolution of the raman spectrometer.
Further, the Raman spectrum database of the standard sample is obtained by the following steps:
acquiring third Raman spectrum data of the standard sample;
background removing processing is carried out on the third Raman spectrum data to obtain fourth Raman spectrum data;
performing interpolation processing on the fourth Raman spectrum data according to the wave number scale values to obtain first Raman shift spectrum data;
taking the maximum value in the first Raman shift spectrum data as a normalization parameter, and carrying out normalization processing on the first Raman shift spectrum data to obtain second Raman shift spectrum data;
obtaining a mask array according to the magnitude relation between the first Raman shift spectrum data and a first threshold;
and performing background interference removal processing on the second Raman shift spectrum data according to the mask module group to obtain third Raman shift spectrum data, wherein a calculation formula of the third Raman shift spectrum data is as follows:
yk＝yk0οαk
wherein, ykAs third Raman shift spectral data, αkAs a mask array, yk0Performing interpolation processing on Raman spectrum data with a sequence number of k in a Raman spectrum database by using a wave number dividing value to obtain first Raman shift spectrum data with a dimension of n;
and marking and storing the third Raman shift spectrum data to obtain a Raman spectrum database of the standard sample.
Further, the identifying components in the test sample specifically includes:
performing interpolation processing on the second Raman spectrum data according to the wave number scale values to obtain fourth Raman shift spectrum data;
and identifying the components in the test sample according to the Raman spectrum database of the standard sample and the fourth Raman shift spectrum data.
Further, the identifying the components in the test sample according to the raman spectrum database and the fourth raman shift spectrum data of the standard sample specifically comprises:
removing background interference from the fourth Raman shift spectrum data according to the mask module group to obtain fifth Raman shift spectrum data;
calculating the energy ratio of the fifth Raman shift spectrum data in the test sample, wherein the calculation formula of the energy ratio is as follows:
wherein eta iskN is the data dimension of the fourth Raman shift spectrum data, zkProjection of fourth Raman shift spectrum data on a vector subspace spanned by a k mask of a standard sample is carried out, and z is the fourth Raman shift spectrum data;
and judging the relation between the energy ratio and the second threshold value to obtain a first judgment result.
Further, the identifying the components in the test sample according to the Raman spectrum database of the standard sample and the fourth Raman shift spectrum data further comprises the following steps:
calculating a correlation coefficient of the third Raman shift spectrum data and the fifth Raman shift spectrum data, wherein the calculation formula of the correlation coefficient is as follows:
where ρ iskIs the correlation coefficient, ykAs third Raman shift spectral data, zkFifth raman shift spectral data;
and judging the magnitude relation between the correlation coefficient and the third threshold value to obtain a second judgment result.
Further, the identifying the components in the test sample according to the Raman spectrum database of the standard sample and the fourth Raman shift spectrum data further comprises the following steps:
if the first judgment result is that the energy ratio is greater than a second threshold value and the second judgment result is that the correlation coefficient is greater than a third threshold value, determining that the test sample contains the components of the standard sample; otherwise, the test sample is judged to contain no components of the standard sample.
The second technical scheme adopted by the invention is as follows:
a system for processing raman spectral data, comprising:
at least one memory for storing a program;
at least one processor for loading the program to implement the method for processing raman spectral data.
The invention has the beneficial effects that: the method comprises the steps of removing a background spectrum from first Raman spectrum data of a test sample to obtain second Raman spectrum data, and identifying components in the test sample according to a Raman spectrum database of a standard sample and the second Raman spectrum data to obtain an identification result.
Drawings
FIG. 1 is a flow chart of a method of processing Raman spectral data according to the present invention;
fig. 2 is a system block diagram of a raman spectrum data processing system according to the present invention.
Detailed Description
The invention is described in further detail below with reference to the figures and the specific embodiments. The step numbers in the following embodiments are provided only for convenience of illustration, the order between the steps is not limited at all, and the execution order of each step in the embodiments can be adapted according to the understanding of those skilled in the art.
Referring to fig. 1, a method for processing raman spectrum data includes the following steps:
s101, collecting first Raman spectrum data of a test sample through a Raman spectrometer;
specifically, the first raman spectrum data is initial spectrum data of the test sample collected by the raman spectrometer, and the data dimension is N and the unit is wavelength.
S102, background removing processing is carried out on the first Raman spectrum data to obtain second Raman spectrum data;
specifically, the second raman spectrum data is target spectrum data of the test sample obtained after background removal processing is performed on the initial spectrum data of the test sample.
S103, identifying components in the test sample according to the Raman spectrum database of the standard sample and the second Raman spectrum data to obtain an identification result.
Specifically, identifying the components of the test sample refers to identifying the same components in the test sample as the standard sample. And performing interpolation and mask processing on the second Raman spectrum data to obtain Raman shift spectrum data of the test sample, comparing and identifying the Raman shift spectrum data of the test sample with Raman shift spectrum data in a Raman spectrum database of the standard sample, and judging whether the test sample contains components of the standard sample according to an identification result.
Specifically, background spectrum in the first Raman spectrum data is removed through background removing processing of the first Raman spectrum data of the test sample, second Raman spectrum data is obtained, and then components in the test sample are identified according to a Raman spectrum database of the standard sample and the second Raman spectrum data, so that an identification result is obtained.
Further, as a preferred embodiment, the background removing processing is performed on the first raman spectrum data to obtain second raman spectrum data, specifically:
background removing processing is carried out on the first Raman spectrum data through a high-pass filter matrix and a penalty function to obtain second Raman spectrum data, and the calculation formula of the second Raman spectrum data is as follows:
wherein,
is the second Raman spectrum data, y is the first Raman spectrum data, x is the iterative spectrum data, F (x) is the evaluation function, H is the finite field high pass filter matrix, phi is the penalty function, lambda is the weight coefficient, DiIs the i-th order differential, M is the total order of the differential, NiIs the i-th order differential data length.
Specifically, the expression of the finite field high-pass filter matrix is as follows:
H＝BA-1
wherein H is a high-pass filter matrix, and B is a differential matrix; a is a truncated matrix, and H, B and A are both N × N dimensional matrices.
The expression of the penalty function is:
φ＝cx2
where φ is a penalty function, c is a coefficient, and x is iterative spectral data.
The iterative spectrum data is obtained by processing the first Raman spectrum data by adopting a BEADS algorithm, and the processing steps are as follows:
step one, acquiring first Raman spectrum data y, a truncation matrix A, a differential matrix B and an ith order differential DiThe first derivative of the penalty function phi' (x), and a weighting factor lambdaiWherein i is 0,1, … M, and M is the total order of differentiation;
step two, obtaining first intermediate data according to the truncation matrix A and the differential matrix B, wherein a calculation formula of the first intermediate data is as follows:
b＝BTBA-1x；
step three, setting iterative spectrum data x as y;
step four, according to the ith order differential DiAnd the first derivative phi' (x) of the penalty function to obtain second intermediate dataThe calculation formula of the second intermediate data is as follows:
step five, according to the weight coefficient lambdaiThe ith order differential DiAnd obtaining third intermediate data from the second intermediate data, wherein the calculation formula of the third intermediate data is as follows:
sixthly, obtaining fourth intermediate data according to the truncation matrix A, the differential matrix B and the third intermediate data D, wherein a calculation formula of the fourth intermediate data is as follows:
Q＝BTB+ATDA；
seventhly, obtaining first iteration spectrum data g according to the truncation matrix A, the first intermediate data b and the fourth intermediate data Q, wherein a calculation formula of the first iteration spectrum data g is as follows:
g＝AQ-1b；
step eight, judging whether the first iteration spectrum data g reaches a preset iteration frequency, if not, making x equal to g, and returning to the step four; otherwise, let x be g, end iteration, and output x.
Specifically, the background spectrum and the interference noise in the first raman spectrum data are separated by processing the first raman spectrum data with the BEADS algorithm. The penalty function is used to sparsify the initial spectral data. And after background removal processing is carried out on the first Raman spectrum data through a high-pass filter and a penalty function, second Raman spectrum data with the background spectrum removed are obtained.
Further, as a preferred embodiment, the finite field high-pass filter matrix is obtained by:
acquiring a first differential operator and a truncation operator;
specifically, the first differential operator is a 2d order differential operator, which can be obtained by:
with P [ -1,2, -1] as the second order differential operator, the 2d order differential operator can be obtained by d convolutions of P, so the formula for the d convolutions of P is:
Pd＝[ad......a0......ad]
wherein, a0、adAnd d are both constants.
Specifically, the truncation operator is obtained by:
setting the cut-off frequency to fcThen the truncation operator is QdSo as to truncate operator QdThe calculation formula of (2) is as follows:
Qd＝convd([-1 2 -1])+αconvd([1 2 1])
wherein d is a constant, and d is a constant,
obtaining a first augmentation matrix and a second augmentation matrix according to the first differential operator and the truncation operator;
specifically, the first augmentation matrix is AEAnd a second amplification matrix BEThe amplification matrix AEAnd BEIs obtained by the following steps:
convolution of P d timesdAnd truncation operator QdConstructing an N x (N + d) -dimensional augmented matrix A of a truncated matrix A and a differential matrix BEAnd BEFirst augmentation matrix AEThe expression of (a) is:
second augmentation matrix BEThe expression of (a) is:
wherein, a0、ad、b0、bdAnd d are both constants, AEIs a first augmented matrix, BEIs a second augmented matrix.
Acquiring an extension mode transformation matrix;
specifically, the expression of the boundary extension matrix is:
wherein G is a boundary extension matrix, E is an NxN dimensional identity matrix, and L and R are a dxN dimensional left boundary and a right boundary, respectively.
Obtaining a truncation matrix according to the first augmentation matrix and the extension mode transformation matrix;
specifically, the expression of the truncation matrix is:
A＝AEG
where A is a truncation matrix, AEIs the first augmented matrix and G is the boundary extended matrix.
Obtaining a differential matrix according to the second augmentation matrix and the extension mode transformation matrix;
specifically, the expression of the differential matrix is:
B＝BEG
wherein B is a differential matrix, BEIs the second augmented matrix and G is the boundary extended matrix.
And obtaining a finite field high-pass filter matrix according to the truncation matrix and the differential matrix.
Specifically, the calculation formula of the finite field high-pass filter matrix is as follows:
H＝BA-1
wherein H is a high-pass filter matrix, B is a differential matrix, A is a truncation matrix, and H, B and A are both NxN dimensional matrices.
Specifically, by introducing a boundary extension matrix into a finite field high-pass filter matrix, the baseline shift phenomenon caused by the fact that the differential value of the boundary position is a non-zero value in the prior art is eliminated.
As a further preferred embodiment, before background removing processing is performed on the first raman spectrum data, the method further comprises the following steps:
calculating the minimum resolution of the wave number according to the wavelength resolution of the Raman spectrometer to obtain a wave number division value, wherein the calculation formula of the wave number division value is as follows:
wherein k isΔIs the division value of wave number, lambdamaxIs the maximum of the measurable wavelength, λ, in the first Raman spectral dataΔIs the wavelength resolution of the raman spectrometer.
Specifically, the wavenumber division value is wavenumber resolution accurate to a finite fraction of a digit, which is calculated from parameters of the raman spectrometer. And a uniform wave number division value is provided for interpolation processing of Raman spectrum data in subsequent steps, and the uniqueness of variables is ensured.
Further, as a preferred embodiment, the raman spectrum database of the standard sample is obtained by:
acquiring third Raman spectrum data of the standard sample;
specifically, the third raman spectrum data is initial spectrum data of the standard sample obtained by acquiring raman spectrum data of the standard sample by the raman spectrometer.
Background removing processing is carried out on the third Raman spectrum data to obtain fourth Raman spectrum data;
specifically, the fourth raman spectrum data is standard spectrum data of the standard sample obtained after background removal processing is performed on the standard sample, and each standard spectrum data of the standard sample is provided with a serial number, so that processing in subsequent steps is facilitated. The step of background removal of the standard sample and the step of background removal of the test sample are the same, i.e. the step of background removal of the third raman spectral data and the step of background removal of the first raman spectral data are the same.
Performing interpolation processing on the fourth Raman spectrum data according to the wave number scale values to obtain first Raman shift spectrum data;
specifically, each raman spectrum data in the fourth raman spectrum data is interpolated by using the wave number scale values to obtain global raman shift spectrum data, that is, the first raman shift spectrum data. For example, the raman spectrum data of the standard sample with number k in the fourth raman spectrum data is selected, and interpolation processing is performed using the wave number division value, so as to obtain the raman shift spectrum data with dimension n and unit of wave number.
Taking the maximum value in the first Raman shift spectrum data as a normalization parameter, and carrying out normalization processing on the first Raman shift spectrum data to obtain second Raman shift spectrum data;
obtaining a mask array according to the magnitude relation between the first Raman shift spectrum data and a first threshold;
specifically, the first threshold is a raman shift spectrum data threshold, which is set according to actual needs. The dimensions of the mask array and the dimensions of the first raman shift spectral data are the same. The calculation formula of the mask array is as follows:
wherein alpha iskIs a mask array, ε is a first threshold value, yk0The method is characterized in that after the Raman spectrum data with the serial number of k in a Raman spectrum database is subjected to interpolation processing by using a wave number dividing value, first Raman shift spectrum data with the dimensionality of n is obtained.
And performing background interference removal processing on the second Raman shift spectrum data according to the mask module group to obtain third Raman shift spectrum data, wherein a calculation formula of the third Raman shift spectrum data is as follows:
yk＝yk0οαk"omicron" is the Hadamard product;
wherein, ykAs third Raman shift spectral data, αkAs a mask array, yk0To use the wave number scale value to perform Raman spectrum data with the serial number of k in the Raman spectrum databaseAfter interpolation processing, obtaining first Raman shift spectrum data with the dimensionality of n;
specifically, the calculation formula of the third raman shift spectral data represents the third raman shift spectral data ykEqual to the first Raman shift spectrum data yk0And mask array alphakAnd carrying out composite operation to obtain an operation result. The background interference removing processing is zero setting processing of tiny values in the second Raman shift spectrum data, and the scope of the tiny values is set according to actual needs.
And marking and storing the third Raman shift spectrum data to obtain a Raman spectrum database of the standard sample.
Specifically, a Raman spectrum database of the standard sample is constructed, standard reference data are provided for component detection of the test sample, and accuracy of a test result is guaranteed.
Further, as a preferred embodiment, the identifying the components in the test sample specifically includes:
performing interpolation processing on the second Raman spectrum data according to the wave number scale values to obtain fourth Raman shift spectrum data;
and identifying the components in the test sample according to the Raman spectrum database of the standard sample and the fourth Raman shift spectrum data.
Specifically, the fourth raman shift spectrum data is raman shift spectrum data of the test sample having a dimension n and a unit of wavenumber, which is obtained by interpolating the second raman spectrum data using the wavenumber scale. Interpolation processing is carried out on the second Raman spectrum data by using the wave number scale values, so that the singularity of variables in the detection process can be ensured, and the influence of other factors on the detection result is avoided.
Further as a preferred embodiment, the identifying the components in the test sample according to the raman spectrum database and the fourth raman shift spectrum data of the standard sample specifically includes:
removing background interference from the fourth Raman shift spectrum data according to the mask module group to obtain fifth Raman shift spectrum data;
specifically, the background interference removal processing on the fourth raman shift spectrum data is specifically to perform zeroing processing on a tiny value in the fourth raman shift spectrum data, where the tiny value is set according to actual needs. The calculation formula of the fifth raman shift spectrum data is as follows:
zk＝zοαk"omicron" is the Hadamard product;
wherein z iskFor the projection of the fourth Raman shift spectrum data z on the vector subspace spanned by the k mask of the standard sample, alphakTo mask array, z is the fourth raman shift spectral data.
The calculation formula of the fifth Raman shift spectrum data represents the fifth Raman shift spectrum data zkEquals fourth raman shift spectral data z and mask array alphakAnd carrying out composite operation to obtain an operation result.
Calculating the energy ratio of the fifth Raman shift spectrum data in the test sample, wherein the calculation formula of the energy ratio is as follows:
wherein eta iskN is the data dimension of the fourth Raman shift spectrum data, zkProjection of fourth Raman shift spectrum data on a vector subspace spanned by a k mask of a standard sample is carried out, and z is the fourth Raman shift spectrum data;
and judging the relation between the energy ratio and the second threshold value to obtain a first judgment result.
Specifically, the second threshold is an energy-to-ratio threshold, which is obtained by performing a test on a standard sample. The mask number sequence is used for avoiding the influence of a non-correlation peak value on a calculation result in the correlation coefficient calculation process while peak searching operation is carried out, and the calculation complexity is reduced.
Further as a preferred embodiment, the identifying the components in the test sample according to the raman spectrum database and the fourth raman shift spectrum data of the standard sample further comprises the following steps:
calculating a correlation coefficient of the third Raman shift spectrum data and the fifth Raman shift spectrum data, wherein the calculation formula of the correlation coefficient is as follows:
where ρ iskIs the correlation coefficient, ykAs third Raman shift spectral data, zkFifth raman shift spectral data;
and judging the magnitude relation between the correlation coefficient and the third threshold value to obtain a second judgment result.
In particular, the third threshold is a correlation coefficient threshold, which is derived from a large amount of experimental data. By calculating the correlation coefficient of the displacement spectrum data of the test sample and the displacement spectrum of the standard sample, whether the test sample contains the standard sample can be further judged, and the accuracy of the judgment result is improved.
Further as a preferred embodiment, the identifying the components in the test sample according to the raman spectrum database and the fourth raman shift spectrum data of the standard sample further comprises the following steps:
if the first judgment result is that the energy ratio is greater than a second threshold value and the second judgment result is that the correlation coefficient is greater than a third threshold value, determining that the test sample contains the components of the standard sample; otherwise, the test sample is judged to contain no components of the standard sample.
Specifically, whether the components of the standard sample are contained in the test sample or not is judged through the first judgment result and the second judgment result together, and the accuracy of the judgment result is ensured.
Referring to fig. 2, an embodiment of the present invention further provides a system for processing raman spectrum data corresponding to the method in fig. 1, including:
at least one memory for storing a program;
at least one processor for loading the program to implement the method for processing raman spectral data.
The contents in the above method embodiments are all applicable to the embodiment of the present system, the functions specifically implemented by the embodiment of the present system are the same as those in the above method embodiments, and the beneficial effects achieved by the embodiment of the present system are also the same as those achieved by the above method.
In summary, according to the processing method and system for raman spectrum data of the present invention, the background spectrum in the first raman spectrum data of the test sample is removed by performing background removal processing on the first raman spectrum data to obtain the second raman spectrum data, and then the components in the test sample are identified according to the raman spectrum database of the standard sample and the second raman spectrum data to obtain the identification result, and the present invention avoids the influence of the background spectrum on the test result in the identification process by performing the background removal processing, and ensures the accuracy of the identification result; further, a high-pass filter boundary extension matrix is used for eliminating a baseline shift phenomenon caused by the fact that a boundary position differential value is a nonzero value in the prior art; further, the influence of baseline background spectrum and interference noise is removed by performing background removal processing on the initial spectrum data; furthermore, by introducing the mask sequence, the influence of the uncorrelated peak value on the calculation result during peak searching operation is avoided, and the calculation complexity is reduced.
While the preferred embodiments of the present invention have been illustrated and described, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (5)
1. A processing method of Raman spectrum data is characterized in that: the method comprises the following steps:
collecting first Raman spectrum data of a test sample through a Raman spectrometer;
background removing processing is carried out on the first Raman spectrum data to obtain second Raman spectrum data;
identifying components in the test sample according to the Raman spectrum database of the standard sample and the second Raman spectrum data to obtain an identification result;
wherein before background removal processing is performed on the first Raman spectrum data, the method further comprises the following steps:
calculating the minimum resolution of the wave number according to the wavelength resolution of the Raman spectrometer to obtain a wave number division value, wherein the calculation formula of the wave number division value is as follows:
wherein k isΔIs the division value of wave number, lambdamaxIs the maximum of the measurable wavelength, λ, in the first Raman spectral dataΔThe wavelength resolution of the Raman spectrometer;
the Raman spectrum database of the standard sample is obtained by the following steps:
acquiring third Raman spectrum data of the standard sample;
background removing processing is carried out on the third Raman spectrum data to obtain fourth Raman spectrum data;
performing interpolation processing on the fourth Raman spectrum data according to the wave number scale values to obtain first Raman shift spectrum data;
taking the maximum value in the first Raman shift spectrum data as a normalization parameter, and carrying out normalization processing on the first Raman shift spectrum data to obtain second Raman shift spectrum data;
obtaining a mask array according to the magnitude relation between the first Raman shift spectrum data and a first threshold;
and performing background interference removal processing on the second Raman shift spectrum data according to the mask module group to obtain third Raman shift spectrum data, wherein a calculation formula of the third Raman shift spectrum data is as follows:
wherein, ykAs third Raman shift spectral data, αkAs a mask array, yk0Interpolation of Raman spectrum data with number k in Raman spectrum database using wave number scale division valueAfter processing, obtaining first Raman shift spectrum data with the dimensionality of n;
marking and storing the third Raman shift spectrum data to obtain a Raman spectrum database of the standard sample;
the identifying of the components in the test sample comprises:
performing interpolation processing on the second Raman spectrum data according to the wave number scale values to obtain fourth Raman shift spectrum data;
removing background interference from the fourth Raman shift spectrum data according to the mask module group to obtain fifth Raman shift spectrum data;
calculating the energy ratio of the fifth Raman shift spectrum data in the test sample, wherein the calculation formula of the energy ratio is as follows:
wherein eta iskN is the data dimension of the fourth Raman shift spectrum data, zkProjection of fourth Raman shift spectrum data on a vector subspace spanned by a k mask of a standard sample is carried out, and z is the fourth Raman shift spectrum data;
judging the relation between the energy ratio and the second threshold value to obtain a first judgment result;
the method for identifying the components in the test sample further comprises the following steps:
calculating a correlation coefficient of the third Raman shift spectrum data and the fifth Raman shift spectrum data, wherein the calculation formula of the correlation coefficient is as follows:
where ρ iskIs the correlation coefficient, ykAs third Raman shift spectral data, zkFifth raman shift spectral data;
and judging the magnitude relation between the correlation coefficient and the third threshold value to obtain a second judgment result.
2. A method of processing raman spectral data according to claim 1, wherein: the background removing processing is carried out on the first Raman spectrum data to obtain second Raman spectrum data, and the method specifically comprises the following steps:
background removing processing is carried out on the first Raman spectrum data through a high-pass filter matrix and a penalty function to obtain second Raman spectrum data, and the calculation formula of the second Raman spectrum data is as follows:
wherein,
is the second Raman spectrum data, y is the first Raman spectrum data, x is the iterative spectrum data, F (x) is the evaluation function, H is the finite field high pass filter matrix, phi is the penalty function, lambda is the weight coefficient, DiIs the i-th order differential, M is the total order of the differential, NiIs the i-th order differential data length.
3. A method of processing raman spectral data according to claim 2, wherein: the finite field high-pass filter matrix is obtained by the following steps:
acquiring a first differential operator and a truncation operator;
obtaining a first augmentation matrix and a second augmentation matrix according to the first differential operator and the truncation operator;
acquiring an extension mode transformation matrix;
obtaining a truncation matrix according to the first augmentation matrix and the extension mode transformation matrix;
obtaining a differential matrix according to the second augmentation matrix and the extension mode transformation matrix;
and obtaining a finite field high-pass filter matrix according to the truncation matrix and the differential matrix.
4. A method of processing raman spectral data according to claim 1, wherein: the method for identifying the components in the test sample further comprises the following steps:
if the first judgment result is that the energy ratio is greater than a second threshold value and the second judgment result is that the correlation coefficient is greater than a third threshold value, determining that the test sample contains the components of the standard sample; otherwise, the test sample is judged to contain no components of the standard sample.
5. A system for processing raman spectral data, comprising: the method comprises the following steps:
at least one memory for storing a program;
at least one processor for loading the program to perform a method of processing raman spectral data according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811533229.6A CN109632761B (en) | 2018-12-14 | 2018-12-14 | Processing method and system of Raman spectrum data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811533229.6A CN109632761B (en) | 2018-12-14 | 2018-12-14 | Processing method and system of Raman spectrum data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109632761A CN109632761A (en) | 2019-04-16 |
| CN109632761B true CN109632761B (en) | 2021-11-09 |
Family
ID=66074199
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811533229.6A Active CN109632761B (en) | 2018-12-14 | 2018-12-14 | Processing method and system of Raman spectrum data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109632761B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110532285B (en) * | 2019-07-16 | 2022-06-24 | 河北凌析科技有限公司 | Associated case determining method and system, electronic device and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101852734A (en) * | 2010-06-01 | 2010-10-06 | 中国人民解放军第二军医大学 | Fake medicine discrimination and analysis device, system and method |
| CN102360502A (en) * | 2011-09-07 | 2012-02-22 | 中国科学院武汉物理与数学研究所 | Automatic baseline correction method |
| CN103217409A (en) * | 2013-03-22 | 2013-07-24 | 重庆绿色智能技术研究院 | Raman spectral preprocessing method |
| CN103913764A (en) * | 2014-02-24 | 2014-07-09 | 东华理工大学 | NaI (TI) scintillation detector gamma energy spectrum high-resolution inversion analysis process and method based on gauss response matrix |
| CN106645091A (en) * | 2017-02-15 | 2017-05-10 | 西派特（北京）科技有限公司 | Raman spectrum based substance qualitative detection method |
| CN108241845A (en) * | 2016-12-26 | 2018-07-03 | 同方威视技术股份有限公司 | Method for deducting spectrogram background and the method by Raman mass spectrum database substance |
-
2018
- 2018-12-14 CN CN201811533229.6A patent/CN109632761B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101852734A (en) * | 2010-06-01 | 2010-10-06 | 中国人民解放军第二军医大学 | Fake medicine discrimination and analysis device, system and method |
| CN102360502A (en) * | 2011-09-07 | 2012-02-22 | 中国科学院武汉物理与数学研究所 | Automatic baseline correction method |
| CN103217409A (en) * | 2013-03-22 | 2013-07-24 | 重庆绿色智能技术研究院 | Raman spectral preprocessing method |
| CN103913764A (en) * | 2014-02-24 | 2014-07-09 | 东华理工大学 | NaI (TI) scintillation detector gamma energy spectrum high-resolution inversion analysis process and method based on gauss response matrix |
| CN108241845A (en) * | 2016-12-26 | 2018-07-03 | 同方威视技术股份有限公司 | Method for deducting spectrogram background and the method by Raman mass spectrum database substance |
| CN106645091A (en) * | 2017-02-15 | 2017-05-10 | 西派特（北京）科技有限公司 | Raman spectrum based substance qualitative detection method |
Non-Patent Citations (5)
| Title |
|---|
| "基于凸优化的激光诱导击穿光谱基线校正方法";柯轲 等;《光谱学与光谱分析》;20180731;第38卷(第7期);第2256-2261页 * |
| "基于稀疏表述及字典学习遥感图像处理关键技术研究";秦振涛;《中国博士学位论文全文数据库 基础科技辑》;20160415;全文 * |
| "拉曼光谱仪器标准化的算法研究及其在药品快检中的应用";陈辉;《中国优秀硕士学位论文全文数据库 医药卫生科技辑》;20160115;全文 * |
| "模型传递及背景扣除的方法研究及其软件实现";詹德坚;《中国优秀硕士学位论文全文数据库 工程科技Ⅰ辑》;20140515;全文 * |
| "稀疏凸优化理论及其在激光诱导击穿光谱定量分析中的应用";珂轲;《中国优秀硕士学位论文全文数据库 工程科技Ⅰ辑》;20181015;论文正文第34-37页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109632761A (en) | 2019-04-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP0156877B1 (en) | Multicomponent quantitative analytical method and apparatus | |
| JP6091493B2 (en) | Spectroscopic apparatus and spectroscopy for determining the components present in a sample | |
| CN113109317B (en) | Raman spectrum quantitative analysis method and system based on background subtraction extraction peak area | |
| CN109738413B (en) | Mixture Raman spectrum qualitative analysis method based on sparse nonnegative least square | |
| CN109409350B (en) | PCA modeling feedback type load weighting-based wavelength selection method | |
| CN110763913B (en) | Derivative spectrum smoothing processing method based on signal segmentation classification | |
| CN104656100A (en) | Line-scanning hyperspectral real-time anomaly detection method and system | |
| CN113008805A (en) | Radix angelicae decoction piece quality prediction method based on hyperspectral imaging depth analysis | |
| CN109632761B (en) | Processing method and system of Raman spectrum data | |
| CN111474117B (en) | Method and device for monitoring crop diseases | |
| CN114878544B (en) | Method for identifying target component from mixture SERS spectrum | |
| CN108195817B (en) | Raman spectrum detection method for removing solvent interference | |
| CN113109318B (en) | Raman spectrum quantitative analysis method and system based on spectral peak height direct extraction | |
| CN114002162A (en) | Soil organic carbon content estimation method, apparatus, storage medium, and program product | |
| JP2021148776A (en) | Peak analysis method and waveform processing device | |
| EP0298398B1 (en) | Analyzer of partial molecular structures | |
| JPS60143765A (en) | Multichannel simultaneous detection data processor for chromatograph | |
| CN113610817B (en) | Characteristic peak identification method, computing device and storage medium | |
| CN109030452A (en) | A kind of Raman spectrum data noise-reduction method based on 5 points of smoothing algorithms three times | |
| CN116106289A (en) | Complex system substance multicomponent analysis method and system based on Raman signal | |
| CN112683816B (en) | Spectrum identification method for spectrum model transmission | |
| CN116399836A (en) | Cross-talk fluorescence spectrum decomposition method based on alternating gradient descent algorithm | |
| CN117929347B (en) | Quick pesticide residue detection method and system for traditional Chinese medicine safflower | |
| CN112229816A (en) | Wood elastic modulus prediction method based on OPLS-SPA-MIX-PLS | |
| CN116660207B (en) | Method for determining characteristic spectrum in oil product quick detection and octane content detection system |
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 |