CN116933056B - Method and system for determining peak area of characteristic peak of Raman spectrum without subtracting Raman background - Google Patents

Abstract

A method and a system for determining the peak area of a Raman spectrum characteristic without deducting the Raman background belong to the technical field of Raman spectrum detection and spectrum data characteristic extraction. The method aims to solve the problems that the effect of an automatic background subtraction algorithm is large in difference and non-uniform in parameters when the peak area of the spectrum characteristic is calculated in the existing quantitative detection system based on the Raman method, and the influence of subjective factors of manually subtracting the background is large. The method comprises the steps of firstly determining accurate center peak position information of a characteristic peak, calculating accurate coordinates of intersection points at two sides of the characteristic peak based on a set confidence peak position ratio p value, then calculating horizontal distance between the center peak position and the two intersection points, moving from a left intersection point to two sides, calculating accurate coordinates of tangential points at two sides of the characteristic peak based on the horizontal distance, and finally calculating the area of a region surrounded by a connecting line of a Raman characteristic peak and the two tangential points based on the coordinates of the left tangential point and the right tangential point. The method is used for determining the characteristic peak area of the Raman spectrum.

Description

Method and system for determining peak area of characteristic peak of Raman spectrum without subtracting Raman background

Technical Field

The invention belongs to the technical field of Raman spectrum detection and spectrum data feature extraction, and in particular relates to a method and system for calculating the peak area of a characteristic peak without deducting the background interference of the Raman spectrum.

Background technique

Raman spectroscopy has the advantages of extremely high detection sensitivity, simple and fast detection process, low detection cost, and the ability to provide a "fingerprint" of the substance to be tested. It has become the preferred technical solution for many detection and analysis instruments. Raman spectroscopy instruments have been widely used in many fields such as food pesticide residue detection, water environment heavy metal ion detection, and environmental biological pollutant detection.

Raman spectroscopy data contains fingerprint information of the object to be tested, and it is relatively easy to achieve qualitative detection of the object to be tested by Raman method; however, the actual detection needs not only need to confirm the type of the object to be tested, but also need to achieve quantitative detection of the object to be tested, which requires obtaining characteristic information associated with the concentration of the object to be tested from the Raman spectrum. At present, the quantitative detection of the object to be tested by Raman spectroscopy technology is mainly achieved by establishing a corresponding relationship between the peak area of the characteristic peak of the Raman spectrum of the object to be tested and the concentration of the substance. The calculation of the peak area needs to avoid the background interference of the Raman spectrum, but the Raman spectrum actually obtained contains unpredictable background information, and the existing methods cannot directly calculate the peak area of the characteristic peak of the Raman spectrum. At present, the calculation of the peak area of the characteristic peak of the Raman spectrum is to first deduct the background interference of the Raman spectrum by various methods, such as polynomial fitting, penalized adaptive partial least squares fitting and other algorithms to automatically deduct the background information of the spectrum or manually deduct the background information of the spectrum, and then calculate the peak area of the characteristic peak to achieve quantitative detection.

However, the effects of the automatic background interference subtraction method vary greatly, and sometimes the fitting parameters for the same background interference are inconsistent; the manual background interference subtraction method is greatly affected by the subjective factors of the experimenter, and the existing method for calculating the peak area of the characteristic peak of the Raman spectrum is not completely reasonable. Therefore, the development of a method that is not interfered by the background information of the Raman spectrum and can calculate the peak area of the characteristic peak without subtracting the background is of great significance to improving the stability and reliability of the Raman spectroscopy quantitative detection system.

Summary of the invention

The purpose of the present invention is to solve the problem that the automatic background subtraction algorithm has large differences and inconsistent parameters when calculating the peak area of spectral characteristic peaks in the existing Raman method-based quantitative detection system, and the manual background subtraction is greatly influenced by subjective factors, which leads to poor result accuracy.

The method for determining the peak area of a characteristic peak of a Raman spectrum without subtracting the Raman background comprises the following steps:

Step S1, calculating the accurate central peak position information of the characteristic peak:

First, read the entire Raman spectrum data and store the raman shift and intensity in arrays x and y respectively; raman shift refers to Raman displacement and intensity refers to Raman intensity;

Then, based on the Raman frequency shift of the characteristic peak to be extracted and the center peak search range, the maximum Raman intensity within the center peak search range sequence length is determined as the accurate center peak position of the characteristic peak to be extracted, and the corresponding index value is the center peak index value center_peak_raman_shift_index;

Step S2, based on the set confidence peak position ratio p value, calculate the exact coordinates of the intersection points on both sides of the characteristic peak, wherein the intersection points are reference points determined based on the confidence peak position ratio p value; the process of calculating the exact coordinates of the intersection points on both sides of the characteristic peak includes the following steps:

Step S21, obtaining the left and right intersection range of the characteristic peak, that is, the peak search range search_peak_length;

Step S22, setting the starting values of the left and right intersection index values xlnum_index and xrnum_index to be equal to the accurate center peak index value determined in step S1;

Step S23: Based on the center peak position, the index values of the left and right intersections are calculated respectively:

Step S231, the length of the sequence for finding the left intersection point is search_peak_length/2;

When y[xlnum_index]>(1-p)*y[center_peak_raman_shift_index], the left intersection point x[xlnum_index] moves one step to the left, that is, the index value decreases by 1, and it is determined whether x[xlnum_index] is within the sequence length range of the left intersection point. If it is within the range, it continues to move to the left to find the intersection point;

Among them, x[xlnum_index] represents the Raman frequency shift corresponding to the index xlnum_index, y[xlnum_index] represents the Raman intensity corresponding to the index xlnum_index, and y[center_peak_raman_shift_index] represents the Raman intensity corresponding to the index center_peak_raman_shift_index;

When y[xlnum_index]<=(1-p)*y[center_peak_raman_shift_index] or x[xlnum_index] at this moment has exceeded the sequence length range of the left intersection, stop, and xlnum_index at this moment is the index value of the left intersection;

Step S232, the length of the sequence for finding the right intersection point is search_peak_length/2;

When y[xrnum_index]>(1-p)*y[center_peak_raman_shift_index], the right intersection point x[xrnum_index] moves one step to the right, that is, the index value is increased by 1, and it is judged whether x[xrnum_index] at this time is within the sequence length range of the right intersection point. If it is within the range, it continues to move to the right to find the intersection point;

When y[xrnum_index]<=(1-p)*y[center_peak_raman_shift_index] or x[xrnum_index] at this moment has exceeded the sequence length range of the right intersection, stop, and xrnum_index at this moment is the index value of the right intersection;

Step S3, calculating the accurate coordinates of the tangent points on both sides of the characteristic peak based on the intersection points on both sides of the characteristic peak, the specific process includes the following steps:

Step S31, calculating the horizontal distances △x _L and △x _R between the central peak and the two intersection points;

Step S32, the starting values of the left and right tangent point coordinate index values xllnum_index and xrrnum_index are respectively equal to xlnum_index and xrnum_index; the left and right tangent points are the left and right endpoints for calculating the peak area of the characteristic peak of the Raman spectrum;

Step S33, the left tangent point x[xllnum_index] moves to the left by one step each time, that is, the index value decreases by 1, and when x[xllnum_index]=x[center_peak_raman_shift_index]-(1/p)*△x _L , the index value xllnum_index of the left tangent point of the characteristic peak is found;

Step S34, the right tangent point x[xrrnum_index] moves rightward by one step each time, that is, the index value is increased by 1, and when x[xrrnum_index]=x[center_peak_raman_shift_index]+(1/p)*△x _R , the right tangent point index value xrrnum_index of the characteristic peak is found;

Step S35, the coordinates of the left and right tangent points are (x[xllnum_index], y[xllnum_index]), (x[xrrnum_index], y[xrrnum_index]);

Step S4: Calculate the area of the region enclosed by the Raman characteristic peak and the line connecting the two tangent points based on the coordinates of the left and right tangent points.

Further, the process of determining the central peak position and finding the maximum Raman intensity within the range sequence length includes the following steps:

First, the step length divis of the Raman shift is obtained, and then the maximum Raman intensity is determined within the sequence length of the central peak position search range according to the step length divis, which is used as the accurate central peak position of the characteristic peak to be extracted.

Furthermore, the process of calculating the area of the region enclosed by the Raman characteristic peak and the line connecting the two tangent points based on the coordinates of the left and right tangent points includes the following steps:

Step S41, taking the left tangent point of the characteristic peak as the starting point, the Raman intensity values y[xllnum_index] and y[xllnum_index+1] of two adjacent points as the upper and lower bases, and the Raman step length as the height, calculate the area of each small trapezoid until the right tangent point is reached, and _{the sum} of the areas of the small trapezoids is recorded as Stotal;

Step S42, determining the area _Sback of the Raman background based on the left and right tangent points of the characteristic peak;

Step S43, the area obtained by subtracting _{Sback from Stotal} is the peak area of the Raman characteristic peak.

Further, the process of determining the area _Sback of the Raman background based on the left and right tangent points of the characteristic peak includes the following steps:

Taking the Raman intensity values y[xllnum_index] and y[xrrnum_index] of the left and right tangent points of the characteristic peak as the upper and lower bases, and the horizontal distance between the two tangent points as the height, the area of the trapezoid is calculated and recorded as _Sback , which is the area of the Raman background.

A system for determining the peak area of characteristic peaks in Raman spectra without subtracting Raman background, including:

Central peak position information determination unit: used to calculate the accurate central peak position information of the characteristic peak, the specific process includes the following steps:

First, read the entire Raman spectrum data and store the raman shift and intensity in arrays x and y respectively; raman shift refers to Raman displacement and intensity refers to Raman intensity;

Then, based on the Raman frequency shift of the characteristic peak to be extracted and the center peak search range, the maximum Raman intensity within the center peak search range sequence length is determined as the accurate center peak position of the characteristic peak to be extracted, and the corresponding index value is the center peak index value center_peak_raman_shift_index;

The intersection point determination unit on both sides of the characteristic peak: based on the set confidence peak position ratio p value, calculates the accurate coordinates of the intersection points on both sides of the characteristic peak, wherein the intersection points are reference points determined based on the confidence peak position ratio p value; the process of calculating the accurate coordinates of the intersection points on both sides of the characteristic peak includes the following steps:

Step S21, obtaining the left and right intersection range of the characteristic peak, that is, the peak search range search_peak_length;

Step S22, setting the starting values of the left and right intersection index values xlnum_index and xrnum_index to be equal to the accurate center peak index value;

Step S23: Based on the center peak position, the index values of the left and right intersections are calculated respectively:

Step S231, the length of the sequence for finding the left intersection point is search_peak_length/2;

When y[xlnum_index]>(1-p)*y[center_peak_raman_shift_index], the left intersection point x[xlnum_index] moves one step to the left, that is, the index value decreases by 1, and it is determined whether x[xlnum_index] is within the sequence length range of the left intersection point. If it is within the range, it continues to move to the left to find the intersection point;

Among them, x[xlnum_index] represents the Raman frequency shift corresponding to the index xlnum_index, y[xlnum_index] represents the Raman intensity corresponding to the index xlnum_index, and y[center_peak_raman_shift_index] represents the Raman intensity corresponding to the index center_peak_raman_shift_index;

When y[xlnum_index]<=(1-p)*y[center_peak_raman_shift_index] or x[xlnum_index] at this moment has exceeded the sequence length range of the left intersection, stop, and xlnum_index at this moment is the index value of the left intersection;

Step S232, the length of the sequence for finding the right intersection point is search_peak_length/2;

When y[xrnum_index]>(1-p)*y[center_peak_raman_shift_index], the right intersection point x[xrnum_index] moves one step to the right, that is, the index value is increased by 1, and it is judged whether x[xrnum_index] at this time is within the sequence length range of the right intersection point. If it is within the range, it continues to move to the right to find the intersection point;

When y[xrnum_index]<=(1-p)*y[center_peak_raman_shift_index] or x[xrnum_index] at this moment has exceeded the sequence length range of the right intersection, stop, and xrnum_index at this moment is the index value of the right intersection;

Characteristic peak two-side tangent point determination unit: based on the intersection points on both sides of the characteristic peak, the accurate coordinates of the tangent points on both sides of the characteristic peak are calculated. The specific process includes the following steps:

Step S31, calculating the horizontal distances △x _L and △x _R between the central peak and the two intersection points;

Step S32, the starting values of the left and right tangent point coordinate index values xllnum_index and xrrnum_index are respectively equal to xlnum_index and xrnum_index; the left and right tangent points are the left and right endpoints for calculating the peak area of the characteristic peak of the Raman spectrum;

Step S33, the left tangent point x[xllnum_index] moves to the left by one step each time, that is, the index value decreases by 1, and when x[xllnum_index]=x[center_peak_raman_shift_index]-(1/p)*△x _L , the index value xllnum_index of the left tangent point of the characteristic peak is found;

Step S34, the right tangent point x[xrrnum_index] moves rightward by one step each time, that is, the index value is increased by 1, and when x[xrrnum_index]=x[center_peak_raman_shift_index]+(1/p)*△x _R , the right tangent point index value xrrnum_index of the characteristic peak is found;

Step S35, the coordinates of the left and right tangent points are (x[xllnum_index], y[xllnum_index]), (x[xrrnum_index], y[xrrnum_index]);

Raman characteristic peak area calculation unit: calculates the area of the region enclosed by the Raman characteristic peak and the line connecting the two tangent points based on the coordinates of the left and right tangent points.

Further, the process of determining the central peak position and finding the maximum Raman intensity within the range sequence length includes the following steps:

First, the step length divis of the Raman shift is obtained, and then the maximum Raman intensity is determined within the sequence length of the central peak position search range according to the step length divis, which is used as the accurate central peak position of the characteristic peak to be extracted.

Furthermore, the process of calculating the area of the region enclosed by the Raman characteristic peak and the line connecting the two tangent points based on the coordinates of the left and right tangent points includes the following steps:

Step S41, taking the left tangent point of the characteristic peak as the starting point, the Raman intensity values y[xllnum_index] and y[xllnum_index+1] of two adjacent points as the upper and lower bases, and the Raman step length as the height, calculate the area of each small trapezoid until the right tangent point is reached, and _{the sum} of the areas of the small trapezoids is recorded as Stotal;

Step S42, determining the area _Sback of the Raman background based on the left and right tangent points of the characteristic peak;

Step S43, the area obtained by subtracting _{Sback from Stotal} is the peak area of the Raman characteristic peak.

Further, the process of determining the area _Sback of the Raman background based on the left and right tangent points of the characteristic peak includes the following steps:

Taking the Raman intensity values y[xllnum_index] and y[xrrnum_index] of the left and right tangent points of the characteristic peak as the upper and lower bases, and the horizontal distance between the two tangent points as the height, the area of the trapezoid is calculated and recorded as _Sback , which is the area of the Raman background.

A computer storage medium stores at least one instruction, which is loaded and executed by a processor to implement the method for determining the peak area of a characteristic peak of a Raman spectrum without subtracting the Raman background.

A device for determining the peak area of a characteristic peak of a Raman spectrum without subtracting the Raman background, the device comprising a processor and a memory, the memory storing at least one instruction, the at least one instruction being loaded and executed by the processor to implement the method for determining the peak area of a characteristic peak of a Raman spectrum without subtracting the Raman background.

Beneficial effects of the present invention:

1. In the process of automatically calculating the peak area of the characteristic peak of the Raman spectrum, the method of the present invention does not need to deduct the background information of the Raman spectrum. Compared with the existing method for calculating the peak area of the characteristic peak of the Raman spectrum, it avoids the problems of large differences in the effect of the automatic background deduction algorithm, inconsistent parameters, and large influence of subjective factors on manual background deduction.

Second, the method of the present invention introduces two optimizable parameters, namely, the confidence peak position ratio and the intersection range, to ensure the accuracy of calculating the Raman peak area. Under the interference of different background function parameters and different background function combinations, for Raman characteristic peaks of different half-peak widths and different intensities, the calculated characteristic area ratio with background and without background changes less than 5%.

3. The method of the present invention can improve the stability and reliability of quantitative analysis of the Raman detection system without changing the Raman instrument.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a schematic diagram of a method for determining the peak area of a characteristic peak of a Raman spectrum without subtracting the Raman background.

FIG2 is a simulated Raman spectrum without background and a simulated Raman spectrum with four backgrounds, namely, Gaussian function, Sigmoid function, Exponential function and Ploynomial function, superimposed simultaneously, in the case where the background intensity is less than the Raman signal intensity in the first experiment of the present invention; the dots in the figure are the accurate central peak position, intersection points on both sides and tangent points on both sides of the characteristic peak calculated by the present invention.

FIG3 is a simulated Raman spectrum without background and a simulated Raman spectrum with four backgrounds, namely, Gaussian function, Sigmoid function, Exponential function and Ploynomial function, superimposed simultaneously, when the background intensity is greater than the Raman signal intensity in Experiment 2 of the present invention; the dots in the figure are the accurate central peak position, intersection points on both sides and tangent points on both sides of the characteristic peak calculated by the present invention.

FIG4 shows a simulated Raman spectrum without background and a simulated Raman spectrum with four backgrounds, namely, Gaussian function, Sigmoid function, Exponential function and Ploynomial function, superimposed simultaneously, when the background intensity of Experiment 3 of the present invention is equivalent to the intensity of some Raman signals and greater than the intensity of the remaining Raman signals; the dots in the figure are the accurate central peak position, intersection points on both sides and tangent points on both sides of the characteristic peak calculated by the present invention.

Detailed ways

The present invention first calculates the accurate central peak position coordinates of the characteristic peak to be extracted according to the input peak position of the characteristic peak to be extracted; then calculates the coordinate values of the left and right intersections according to the input confidence peak position ratio and the left and right intersection search range sequence length parameters; further, calculates the coordinates of the left and right tangent points of the characteristic peak to be extracted; finally, calculates the area of the region enclosed by the line connecting the two tangent points and the Raman characteristic peak, thereby completing the calculation of the peak area of the characteristic peak.

It has been verified that for Raman spectral signals with different types and different parameter background combinations, the ratio of feature areas with and without background extracted by the method of the present invention varies by less than 5%.

The present invention will be further described below in conjunction with the accompanying drawings and specific implementation methods. The following examples are used to illustrate the present invention but are not used to limit the scope of the present invention.

Specific implementation method 1: This implementation method is described in conjunction with Figure 1.

This embodiment is a method for determining the peak area of a characteristic peak of a Raman spectrum without subtracting the Raman background, comprising the following steps:

Step S1, calculating the accurate central peak position information of the characteristic peak:

Step S11, read the entire Raman spectrum data, and store the raman shift and intensity in arrays x and y respectively; raman shift refers to the Raman displacement, and intensity refers to the Raman intensity;

Step S12, determining the Raman frequency shift of the characteristic peak to be extracted and the search range of the central peak position;

Step S13, based on the Raman frequency shift of the characteristic peak to be extracted and the center peak search range, determine the maximum Raman intensity within the center peak search range sequence length as the accurate center peak position of the characteristic peak to be extracted, and the corresponding index value is the center peak index value center_peak_raman_shift_index;

The process of determining the central peak position and finding the maximum Raman intensity within the range sequence length includes the following steps:

First, the step length divis of the Raman shift is obtained, and then the maximum Raman intensity is determined within the sequence length of the central peak position search range according to the step length divis, which is used as the accurate central peak position of the characteristic peak to be extracted.

Step S2, based on the set confidence peak position ratio p value, calculate the exact coordinates of the intersection points on both sides of the characteristic peak, wherein the intersection points are reference points determined based on the confidence peak position ratio p value; the process of calculating the exact coordinates of the intersection points on both sides of the characteristic peak includes the following steps:

Step S21, obtaining the left and right intersection range of the characteristic peak, that is, the peak search range search_peak_length;

Step S22, setting the starting values of the left and right intersection index values xlnum_index and xrnum_index to be equal to the accurate center peak index value;

Step S23: Based on the center peak position, the index values of the left and right intersections are calculated respectively:

Step S231, the sequence length for finding the left intersection is search_peak_length/2, that is, half of the peak search range. When y[xlnum_index]>(1-p)*y[center_peak_raman_shift_index], the left intersection x[xlnum_index] moves one step to the left, that is, the index value is reduced by 1, and it is determined whether x[xlnum_index] at this time is within the sequence length range of the left intersection. If it is within the range, continue to move left to find the intersection.

Among them, x[xlnum_index] represents the Raman frequency shift corresponding to the index xlnum_index, y[xlnum_index] represents the Raman intensity corresponding to the index xlnum_index, and similarly y[center_peak_raman_shift_index] represents the Raman intensity corresponding to the index center_peak_raman_shift_index;

The index value is a mapping of the horizontal and vertical coordinates of the Raman spectrum (similar to computer addressing), which corresponds to the coordinate value one by one. If the index value is added/subtracted by 1, the horizontal coordinate will move right/left by one step. Adjusting the index value can control the horizontal and vertical coordinates at the same time. The relationship between the index value and the horizontal and vertical coordinates can be understood through the following example:

Horizontal axis Raman shift: 400, 401, 402, 403, 404, 405

Vertical axis Raman intensity: 50, 70, 90, 100, 120, 150

Index: 1, 2, 3, 4, 5, 6

The current horizontal coordinate is x[3]=402, and the corresponding y[3]=90; when moving one step to the right, that is, the index value is 3+1=4, then x[4]=403 must be present, and the corresponding y[4]=100.

When y[xlnum_index]<=(1-p)*y[center_peak_raman_shift_index] or x[xlnum_index] at this moment has exceeded the sequence length range of the left intersection, stop, and xlnum_index at this moment is the index value of the left intersection;

Step S232, the sequence length for finding the right intersection is search_peak_length/2, which is half of the peak search range. When y[xrnum_index]>(1-p)*y[center_peak_raman_shift_index], the right intersection x[xrnum_index] moves one step to the right, that is, the index value is increased by 1, and it is determined whether x[xrnum_index] at this time is within the sequence length range of the right intersection. If so, continue to move right to find the intersection.

When y[xrnum_index]<=(1-p)*y[center_peak_raman_shift_index] or x[xrnum_index] at this moment has exceeded the sequence length range of the right intersection, stop, and xrnum_index at this moment is the index value of the right intersection.

Step S3, calculate the exact coordinates of the tangent points on both sides of the characteristic peak:

Step S31, calculating the horizontal distances △x _L and △x _R between the central peak and the two intersection points;

Step S32, the starting values of the left and right tangent point coordinate index values xllnum_index and xrrnum_index are respectively equal to xlnum_index and xrnum_index; the left and right tangent points are the left and right endpoints for calculating the peak area of the characteristic peak of the Raman spectrum;

Step S33, the left tangent point x[xllnum_index] moves to the left by one step each time, that is, the index value decreases by 1, and when x[xllnum_index]=x[center_peak_raman_shift_index]-(1/p)*△x _L , the index value xllnum_index of the left tangent point of the characteristic peak is found;

Step S34, the right tangent point x[xrrnum_index] moves rightward by one step each time, that is, the index value is increased by 1, and when x[xrrnum_index]=x[center_peak_raman_shift_index]+(1/p)*△x _R , the right tangent point index value xrrnum_index of the characteristic peak is found;

Step S35, the coordinates of the left and right tangent points are (x[xllnum_index], y[xllnum_index]), (x[xrrnum_index], y[xrrnum_index]) respectively.

Step S4, calculating the area of the region enclosed by the Raman characteristic peak and the line connecting the two tangent points:

Step S41, taking the left tangent point of the characteristic peak as the starting point, the Raman intensity values y[xllnum_index] and y[xllnum_index+1] of two adjacent points as the upper and lower bases, and the Raman step length as the height, calculate the area of each small trapezoid until the right tangent point is reached, and _{the sum} of the areas of the small trapezoids is recorded as Stotal;

Step S42, taking the Raman intensity values y[xllnum_index] and y[xrrnum_index] of the left and right tangent points of the characteristic peak as the upper and lower bases, and the horizontal distance between the two tangent points as the height, the area of the trapezoid is calculated and recorded as _Sback , that is, the area of the Raman background;

Step S43, the area obtained by subtracting _{Sback from Stotal} is the peak area of the Raman characteristic peak.

It should be noted that, in the method for determining the peak area of a characteristic peak of a Raman spectrum without subtracting Raman background of the present invention, the term "no subtraction of Raman background" refers to a method relative to the existing method for calculating the peak area of a characteristic peak of a Raman spectrum, in which the Raman background needs to be subtracted in advance. That is, the method does not need to subtract the background according to the existing method, but does not mean that the present invention does not need to subtract the area of the Raman background.

Specific implementation method 2:

This embodiment is a system for determining the peak area of a characteristic peak of a Raman spectrum without subtracting the Raman background. The system is a program corresponding to the method for determining the peak area of a characteristic peak of a Raman spectrum without subtracting the Raman background described in the first embodiment. The system includes:

Central peak position information determination unit: used to calculate the accurate central peak position information of the characteristic peak, the specific process includes the following steps:

First, read the entire Raman spectrum data and store the raman shift and intensity in arrays x and y respectively; raman shift refers to Raman displacement, and intensity refers to Raman intensity;

Then, based on the Raman frequency shift of the characteristic peak to be extracted and the center peak search range, the maximum Raman intensity within the center peak search range sequence length is determined as the accurate center peak position of the characteristic peak to be extracted, and the corresponding index value is the center peak index value center_peak_raman_shift_index;

The intersection point determination unit on both sides of the characteristic peak: based on the set confidence peak position ratio p value, calculates the accurate coordinates of the intersection points on both sides of the characteristic peak, wherein the intersection points are reference points determined based on the confidence peak position ratio p value; the process of calculating the accurate coordinates of the intersection points on both sides of the characteristic peak includes the following steps:

Step S21, obtaining the left and right intersection range of the characteristic peak, that is, the peak search range search_peak_length;

Step S22, setting the starting values of the left and right intersection index values xlnum_index and xrnum_index to be equal to the index value of the accurate center peak position;

Step S23: Based on the center peak position, the index values of the left and right intersections are calculated respectively:

Step S231, the length of the sequence for finding the left intersection point is search_peak_length/2;

When y[xlnum_index]>(1-p)*y[center_peak_raman_shift_index], the left intersection point x[xlnum_index] moves one step to the left, that is, the index value decreases by 1, and it is determined whether x[xlnum_index] is within the sequence length range of the left intersection point. If it is within the range, it continues to move to the left to find the intersection point;

Among them, x[xlnum_index] represents the Raman frequency shift corresponding to the index xlnum_index, y[xlnum_index] represents the Raman intensity corresponding to the index xlnum_index, and y[center_peak_raman_shift_index] represents the Raman intensity corresponding to the index center_peak_raman_shift_index;

When y[xlnum_index]<=(1-p)*y[center_peak_raman_shift_index] or x[xlnum_index] at this moment has exceeded the sequence length range of the left intersection, stop, and xlnum_index at this moment is the index value of the left intersection;

Step S232, the length of the sequence for finding the right intersection point is search_peak_length/2;

When y[xrnum_index]>(1-p)*y[center_peak_raman_shift_index], the right intersection point x[xrnum_index] moves one step to the right, that is, the index value is increased by 1, and it is judged whether x[xrnum_index] at this time is within the sequence length range of the right intersection point. If it is within the range, it continues to move to the right to find the intersection point;

When y[xrnum_index]<=(1-p)*y[center_peak_raman_shift_index] or x[xrnum_index] at this moment has exceeded the sequence length range of the right intersection, stop, and xrnum_index at this moment is the index value of the right intersection;

Characteristic peak two-side tangent point determination unit: based on the intersection points on both sides of the characteristic peak, the accurate coordinates of the tangent points on both sides of the characteristic peak are calculated. The specific process includes the following steps:

Step S31, calculating the horizontal distances △x _L and △x _R between the central peak and the two intersection points;

Step S32, the starting values of the left and right tangent point coordinate index values xllnum_index and xrrnum_index are respectively equal to xlnum_index and xrnum_index; the left and right tangent points are the left and right endpoints for calculating the peak area of the characteristic peak of the Raman spectrum;

Step S33, the left tangent point x[xllnum_index] moves to the left by one step each time, that is, the index value decreases by 1, and when x[xllnum_index]=x[center_peak_raman_shift_index]-(1/p)*△x _L , the index value xllnum_index of the left tangent point of the characteristic peak is found;

Step S34, the right tangent point x[xrrnum_index] moves rightward by one step each time, that is, the index value is increased by 1, and when x[xrrnum_index]=x[center_peak_raman_shift_index]+(1/p)*△x _R , the right tangent point index value xrrnum_index of the characteristic peak is found;

Step S35, the coordinates of the left and right tangent points are (x[xllnum_index], y[xllnum_index]), (x[xrrnum_index], y[xrrnum_index]);

Raman characteristic peak area calculation unit: based on the coordinates of the left and right tangent points, the area of the Raman characteristic peak and the line connecting the two tangent points is calculated. The specific process includes the following steps:

Step S41, taking the left tangent point of the characteristic peak as the starting point, the Raman intensity values y[xllnum_index] and y[xllnum_index+1] of two adjacent points as the upper and lower bases, and the Raman step length as the height, calculate the area of each small trapezoid until the right tangent point is reached, and _{the sum} of the areas of the small trapezoids is recorded as Stotal;

Step S42, taking the Raman intensity values y[xllnum_index] and y[xrrnum_index] of the left and right tangent points of the characteristic peak as the upper and lower bases, and the horizontal distance between the two tangent points as the height, the area of the trapezoid is calculated and recorded as _Sback , that is, the area of the Raman background;

Step S43, the area obtained by subtracting _{Sback from Stotal} is the peak area of the Raman characteristic peak.

Specific implementation method three:

This embodiment is a computer storage medium, in which at least one instruction is stored. The at least one instruction is loaded and executed by a processor to implement the method for determining the peak area of a characteristic peak of a Raman spectrum without subtracting the Raman background.

It should be understood that the instructions include computer program products, software or computerized methods corresponding to any method described in the present invention; the instructions can be used to program a computer system, or other electronic device. Computer storage media may include readable media on which instructions are stored, which may include but are not limited to magnetic storage media, optical storage media; magneto-optical storage media include read-only memory ROM, random access memory RAM, erasable programmable memory (e.g., EPROM and EEPROM) and flash memory layers, or other types of media suitable for storing electronic instructions.

Specific implementation method four:

This embodiment is a device for determining the peak area of characteristic peaks of Raman spectrum without subtracting Raman background, and the device includes a processor and a memory. It should be understood that the device includes any device including a processor and a memory described in the present invention, and the device may also include other units and modules that perform display, interaction, processing, control, etc. and other functions through signals or instructions;

At least one instruction is stored in the memory, and the at least one instruction is loaded and executed by the processor to implement the method for determining the peak area of a characteristic peak of a Raman spectrum without subtracting the Raman background.

Example

Since the actual collected Raman spectrum background peaks are irregular and unpredictable, it is impossible to obtain Raman spectrum data without any background interference. For Raman spectrum data with background interference, it is not reasonable to use the existing method to deduct the background and calculate the peak area of the characteristic peak as the standard peak area to evaluate the beneficial effects of the present invention. Therefore, the present invention adopts the common method of Raman spectrum simulation to superimpose multiple Lorentz peaks to generate Raman spectrum data without background.

The Raman background signal is generated by four background functions, namely, Gaussian function, Sigmoid function, Exponential function and Ploynomial function, either alone or in combination. By superimposing the background interference with the simulated Raman spectrum data, we can obtain background Raman spectrum data of different types and parameters.

The following examples are used to verify the beneficial effects of the present invention:

In the embodiment, the optimizable parameters used when calculating the peak areas of different characteristic peaks of the same simulated Raman spectrum are consistent, and only the central peak position information of the characteristic peak to be calculated is changed; when comparing the peak area changes with and without background interference, only the background function combination is changed, and the background function parameters and the optimizable parameters are not changed.

Experiment 1: Calculation of Raman peak area when background signal intensity is less than Raman signal intensity is completed according to the following steps:

The simulated Raman spectrum parameters generated in this experiment are as follows: the Raman center positions are 544, 739, 1009, 1329, and 1690, the Raman half-peak widths are 36, 25, 30, 33, and 57, the Raman peak areas are 7200, 6020, 9589, 10345, and 11300, and the Raman step size is 0.1.

The background function parameters of this experiment are as follows: Gaussian parameters are 1000, 128, 550; Sigmoid parameters are 830, 25, 100; Exponential parameter is 21; Ploynomial parameter is 20.

Step S1, calculating the accurate central peak position information of the characteristic peak to be extracted, the specific steps are as follows:

Step S11, storing the raman shift and intensity of the generated Raman spectrum in arrays x and y respectively; in particular, by changing the combination of different types of background functions, Raman spectra containing different background interferences can be generated;

Step S12, input the characteristic peaks to be extracted. In this experiment, the characteristic peaks to be extracted are 535, 750, 1000, 1320, and 1700, respectively, and the input center peak position search range is set to 80;

Step S13, finding the maximum value of the Raman intensity within the length of the input center peak sequence to be extracted as the accurate center peak position of the characteristic peak to be extracted, and its index value is the center peak index value center_peak_raman_shift_index, and the specific steps are as follows:

The simulated Raman spectrum raman shift step divis is 0.1, the sequence length of the central peak search range is 80, and the maximum Raman intensity within the range of 40 sequence lengths on the left and right of the input central peak is found as the accurate central peak position, and the corresponding index value is the accurate central peak index value.

Step S2, calculating the left and right intersection coordinates of the central peak position of the characteristic peak to be extracted, the specific steps are as follows:

Step S21, in this experiment, the left and right intersection search range sequence length parameter is set to 200, the background-free Raman spectrum confidence peak ratio is set to 0.9, and the background Raman spectrum confidence peak ratio is set to 0.35;

Step S22, setting the starting values of the index values xlnum_index and xrnum_index of the left and right intersection points to be equal to the central peak index value determined in step S13;

Step S23, determining the index values of the two intersection points respectively to the left and to the right according to the calculation formula, the specific steps are as follows:

Step S231, the sequence length range for searching the left intersection is half of the sequence length parameter for searching the left and right intersections set in step S21. When y[xlnum_index]>(1-p)*y[center_peak_raman_shift_index], the left intersection x[xlnum_index] moves one step to the left each time, that is, the index value is reduced by 1, and it is determined whether the current x[xlnum_index] is within the set left intersection sequence length range. If it is within the range, it continues to move to the left to search for the intersection. When y[xlnum_index]<=(1-p)*y[center_peak_raman_shift_index] or the current x[xlnum_index] has exceeded the sequence length range of the left intersection, it stops, and the current xlnum_index is the left intersection index value;

Step S232, the sequence length of the right intersection is half of the sequence length parameter of the left and right intersection search range set in step S21. When y[xrnum_index]>(1-p)*y[center_peak_raman_shift_index], the right intersection x[xrnum_index] moves one step to the right each time, that is, the index value is increased by 1, and it is determined whether x[xrnum_index] at this time is within the set sequence length range of the right intersection. If it is within the range, continue to move to the right to find the intersection. When y[xrnum_index]<＝(1-p)*y[center_peak_raman_shift_index] or x[xrnum_index] at this moment has exceeded the sequence length range of the right intersection, stop, and xrnum_index at this time is the index value of the right intersection.

Step S3, calculating the left and right tangent point coordinates of the central peak position of the characteristic peak to be extracted, the specific steps are as follows:

Step S31, calculating the horizontal distances △x _L and △x _R between the central peak to be extracted and the two intersection points;

Step S32, the starting values of the left and right tangent point coordinate index values xllnum_index and xrrnum_index are respectively equal to xlnum_index and xrnum_index;

Step S33, the left tangent point x[xllnum_index] moves to the left by one step each time, that is, the index value decreases by 1, and when x[xllnum_index]=x[center_peak_raman_shift_index]-(1/p)*△x _L , the index value xllnum_index of the left tangent point of the characteristic peak is found;

Step S34, the right tangent point x[xrrnum_index] is moved rightward by one step each time, that is, the index value is increased by 1, and when x[xrrnum_index]=x[center_peak_raman_shift_index]+(1/p)*△x _R , the right tangent point index value xrrnum_index of the characteristic peak is found;

Step S35, the coordinates of the left and right tangent points are (x[xllnum_index], y[xllnum_index]), (x[xrrnum_index], y[xrrnum_index]) respectively.

Step S4, calculating the peak area of the characteristic peak to be extracted, the specific steps are as follows:

Step S41, taking the left tangent point of the characteristic peak as the starting point, the Raman intensity values of the two adjacent points y[xllnum_index] and y[xllnum_index+1] as the upper and lower bases, and the Raman step length divis as the height, calculate the area of each small trapezoid until reaching the right tangent point, and _{the sum} of the areas of the small trapezoids is recorded as Stotal;

Step S42, taking the Raman intensity values y[xllnum_index] and y[xrrnum_index] corresponding to the left and right tangent points of the characteristic peak as the upper and lower bases, and the horizontal distance between the two tangent points as the height, the area of the Raman background is calculated and recorded as _Sback ;

Step S43, the area obtained by _{subtracting Sback} from Stotal is the peak area of the Raman characteristic peak.

FIG2 is a simulated Raman spectrum of four background functions and a simulated Raman spectrum without background when the background intensity is less than the Raman signal intensity in Experiment 1, wherein the dots in the figure are the accurate central peak position, the intersection point on both sides, and the tangent point position on both sides of the characteristic peak calculated by the present invention;

Table 1 is a statistical diagram of the characteristic area ratios with and without background of different characteristic peaks calculated for different background function combinations in the present invention when the background intensity is less than the Raman signal intensity in the first test.

Table 1

Note: Raman spectrum background is generated by the following four functions, where G represents Gaussian function, S represents Sigmoid function, E represents Exponential function, and P represents Ploynomial function; the letter combination indicates that the spectrum background is composed of the corresponding functions.

It can be seen from the results in Figure 2 and Table 1 that when the background intensity is less than the Raman signal intensity, under the interference of different background function parameters and different background function combinations, the calculated characteristic area ratio with background and without background for Raman characteristic peaks with different half-peak widths and different intensities in the present invention changes by less than 5%.

Test 2: Calculate the Raman peak area when the background signal intensity is greater than the Raman signal intensity. The difference between this test and test 1 is:

The background function parameters used in this experiment are as follows: Gaussian parameters are 400, 739, and 1000; Sigmoid parameters are 530, 100, and 500; Exponential parameter is 300; and Ploynomial parameter is 200.

Step S21, the left and right intersection search range sequence length parameter is 140, the background-free Raman spectrum confidence peak ratio is set to 0.93, and the background Raman spectrum confidence peak ratio is set to 0.67; other steps and parameters are the same as those in Experiment 1.

FIG3 is a simulated Raman spectrum graph of four background functions and a simulated Raman spectrum graph without background when the background intensity is greater than the Raman signal intensity in Experiment 2, wherein the dots in the graph are the accurate central peak position, the intersection point on both sides, and the tangent point position on both sides of the characteristic peak calculated by the present invention;

Table 2 is a statistical diagram of the characteristic area ratios with and without background of different characteristic peaks calculated for different background function combinations in the present invention when the background intensity in Experiment 2 is greater than the Raman signal intensity.

Table 2

Note: Raman spectrum background is generated by the following four functions, where G represents Gaussian function, S represents Sigmoid function, E represents Exponential function, and P represents Ploynomial function; the letter combination indicates that the spectrum background is composed of the corresponding functions.

It can be seen from the results in Figure 3 and Table 2 that when the background intensity is greater than the Raman signal intensity, under the interference of different background function parameters and different background function combinations, the calculated characteristic area ratio with background and without background for Raman characteristic peaks with different half-peak widths and different intensities in the present invention changes by less than 5%.

Test 3: Calculate the Raman peak area when the background signal intensity is equal to the intensity of some Raman signals and greater than the intensity of the remaining Raman signals. The difference between this test and test 1 is:

The simulated Raman spectrum parameters of this experiment are as follows: the Raman center positions are 544, 739, 1009, 1329, and 1690, the Raman half-peak widths are 20, 30, 38, 50, and 42, the Raman peak areas are 17600, 29800, 26589, 20345, and 10000, and the Raman step size is 0.1.

The background function parameters of this experiment are as follows: Gaussian parameters are 2400, 750, 800; Sigmoid parameters are 600, 260, 1000; Exponential parameter is 100; Ploynomial parameter is 200.

Step S21, the left and right intersection search range sequence length parameter is 120, the background-free Raman spectrum confidence peak ratio is set to 0.88, and the background Raman spectrum confidence peak ratio is set to 0.6; the remaining steps and parameters are the same as those in Experiment 1.

FIG4 is a simulated Raman spectrum with four background functions and a simulated Raman spectrum without background when the background intensity of Experiment 3 is equivalent to the intensity of some Raman signals and greater than the intensity of other Raman signals. The dots in the figure are the accurate central peak position, intersection points on both sides, and tangent points on both sides of the characteristic peak calculated by the present invention;

Table 3 is a statistical diagram of the characteristic area ratios with and without background of different characteristic peaks calculated for different background function combinations in the present invention when the background intensity of Experiment 3 is equivalent to the intensity of some Raman signals and greater than the intensity of other Raman signals.

table 3

Note: Raman spectrum background is generated by the following four functions, where G represents Gaussian function, S represents Sigmoid function, E represents Exponential function, and P represents Ploynomial function; the letter combination indicates that the spectrum background is composed of the corresponding functions.

It can be seen from the results in Figure 4 and Table 3 that when the background intensity is equivalent to the intensity of some Raman signals and greater than the intensity of other Raman signals, under the interference of different background function parameters and different background function combinations, for Raman characteristic peaks with different half-peak widths and different intensities, the calculated characteristic area ratio with background and without background changes less than 5%.

From the above three test results, it can be seen that under the interference of different background function parameters and different background function combinations, without changing the confidence peak ratio parameter and the left and right intersection range parameter, the calculated characteristic area ratio of different characteristic peaks with background and without background changes less than 5%. The calculation results of the method of the present invention have low variability and are not affected by subjective factors. It is superior to other methods for calculating Raman peak area at this stage, and plays an important role in improving the stability and reliability of Raman spectroscopy quantitative detection systems.

Claims (10)

1. The method for determining the peak area of the characteristic peak of the Raman spectrum without subtracting the Raman background is characterized by comprising the following steps:

s1, calculating accurate center peak position information of a characteristic peak:

Firstly, reading the whole Raman spectrum data, and respectively storing RAMAN SHIFT and intension in an array x and an array y; RAMAN SHIFT, raman shift, intensity, raman intensity;

Then, based on the Raman frequency shift of the characteristic peak to be extracted and the searching range of the central peak position, determining the maximum value of Raman intensity in the sequence length of the searching range of the central peak position, and taking the maximum value as the accurate central peak position of the characteristic peak to be extracted, wherein the corresponding index value is the central peak position index value center_peak_ raman _shift_index;

s2, calculating accurate coordinates of intersection points on two sides of the characteristic peak based on a set confidence peak position ratio p value, wherein the intersection points are reference points determined based on the confidence peak position ratio p value; the process for calculating the accurate coordinates of the intersection points at the two sides of the characteristic peak comprises the following steps:

S21, acquiring a left-right intersection point range of a characteristic peak, namely a peak searching range search_peak_length;

Step S22, setting the initial values of left and right intersection point index values xlnum _index and xrnum_index to be equal to the accurate center peak position index value determined in step S1;

step S23, calculating index values of the left intersection point and the right intersection point respectively leftwards and rightwards based on the central peak position:

step S231, searching for a sequence length of a left intersection point to be search_peak_length/2;

When y [ xlnum _index ] > (1-p) ×y [ center_peak_ raman _shift_index ], each time the left intersection point x [ xlnum _index ] moves leftwards by one step length, namely the index value is subtracted by 1, judging whether the x [ xlnum _index ] is in the sequence length range of the left intersection point at the moment, if so, continuing to move leftwards to search for the intersection point;

Where x [ xlnum _index ] represents the corresponding raman shift at index xlnum _index, y [ xlnum _index ] represents the corresponding raman intensity at index xlnum _index, and y [ center_peak_ raman _shift_index ] represents the corresponding raman intensity at index center_peak_ raman _shift_index;

Stopping when y [ xlnum _index ] < = (1-p) [ center_peak_ raman _shift_index ] or x [ xlnum _index ] at the moment exceeds the sequence length range of the left intersection point, wherein xlnum _index is the left intersection point index value;

step S232, searching a right intersection point, wherein the sequence length is search_peak_length/2;

When y [ xrnum _index ] > (1-p) ×y [ center_peak_ raman _shift_index ], every time the right intersection point x [ xrnum _index ] moves to the right by one step length, namely, adding 1 to the index value, judging whether the x [ xrnum _index ] is in the range of the sequence length of the right intersection point at the moment, if so, continuing to move to the right to search for the intersection point;

Stopping when y [ xrnum _index ] < = (1-p) [ center_peak_ raman _shift_index ] or x [ xrnum _index ] at the moment already exceeds the sequence length range of the right intersection point, wherein xrnum _index is the index value of the right intersection point;

step S3, calculating accurate coordinates of tangential points of two sides of the characteristic peak based on the intersection points of the two sides of the characteristic peak, wherein the specific process comprises the following steps:

S31, calculating a horizontal distance Deltax _L、△x_R between the central peak position and the two intersection points;

Step S32, the starting values of the left and right tangent point coordinate index values xllnum _index and xrrnum_index are respectively equal to xlnum _index and xrnum_index; the left and right tangent points are left and right end points for calculating the peak area of the characteristic peak of the Raman spectrum;

Step S33, moving the left tangent point x [ xllnum _index ] one step to the left each time, i.e. subtracting 1 from the index value, and finding the left tangent point index value xllnum _index of the characteristic peak when x [ xllnum _index ] =x [ center_peak_ raman _shift_index ] - (1/p) ×Δx _L;

step S34, right tangent point x [ xrrnum _index ] is shifted one step to the right each time, i.e. index value is added by 1, when x [ xrrnum _index ] =x [ center_peak_ raman _shift_index ] + (1/p) Δx _R, finding the index value xrrnum _index of right tangent point of characteristic peak;

The coordinates of the left and right tangent points in step S35 are (x [ xllnum _index ], y [ xllnum _index ]), (x [ xrrnum _index ], y [ xrrnum _index ]);

And S4, calculating the area of an area surrounded by the Raman characteristic peak and the connecting line of the two tangent points based on the coordinates of the left tangent point and the right tangent point.

2. The method for determining peak areas of raman spectrum features without subtracting raman background according to claim 1 wherein the process of determining the maximum value of raman intensity within the length of the central peak finding range sequence comprises the steps of:

firstly, a step divis of RAMAN SHIFT is obtained, and then a Raman intensity maximum value is determined in the center peak position searching range sequence length according to the step divis and is used as the accurate center peak position of the characteristic peak to be extracted.

3. The method for determining the peak area of a characteristic peak of a raman spectrum without subtracting a raman background according to claim 1 or 2, wherein the process of calculating the area of a region surrounded by a raman characteristic peak and a line connecting two tangent points based on the coordinates of the two tangent points comprises the following steps:

Step S41, taking a left tangent point of a characteristic peak as a starting point, taking Raman intensity values y [ xllnum _index ] and y [ xllnum _index+1] of two adjacent points as upper and lower bottoms, taking a Raman step length as high, and calculating the area of each small trapezoid until reaching a right tangent point to be cut off, wherein the sum of the areas of the small trapezoids is recorded as S _{Total (S)} ;

step S42, determining the area S _{Back of body} of the Raman background based on the left and right tangential points of the characteristic peak;

The area obtained by subtracting S _{Back of body} from steps S43 and S _{Total (S)} is the peak area of the raman characteristic peak.

4. A method for determining the characteristic peak area of a raman spectrum without subtracting a raman background according to claim 3, wherein the process of determining the area S _{Back of body} of the raman background based on the tangential points of the characteristic peak comprises the steps of:

The Raman intensity values y [ xllnum _index ] and y [ xrrnum _index ] of the left and right tangential points of the characteristic peak are taken as the upper bottom and the lower bottom, the horizontal distance between the two tangential points is high, and the calculated area of the trapezoid is recorded as S _{Back of body} , namely the area of the Raman background.

5. A system for determining peak areas of raman spectral features without subtracting raman background, comprising:

center peak position information determination unit: the specific process for calculating the accurate center peak position information of the characteristic peak comprises the following steps:

Firstly, reading the whole Raman spectrum data, and respectively storing RAMAN SHIFT and intension in an array x and an array y; RAMAN SHIFT, raman shift, intensity, raman intensity;

Then, based on the Raman frequency shift of the characteristic peak to be extracted and the searching range of the central peak position, determining the maximum value of Raman intensity in the sequence length of the searching range of the central peak position, and taking the maximum value as the accurate central peak position of the characteristic peak to be extracted, wherein the corresponding index value is the central peak position index value center_peak_ raman _shift_index;

Characteristic peak both sides intersection point determining unit: calculating accurate coordinates of intersection points on two sides of the characteristic peak based on a set confidence peak position ratio p value, wherein the intersection points are reference points determined based on the confidence peak position ratio p value; the process for calculating the accurate coordinates of the intersection points at the two sides of the characteristic peak comprises the following steps:

S21, acquiring a left-right intersection point range of a characteristic peak, namely a peak searching range search_peak_length;

S22, setting a starting value of left and right intersection point index values xlnum _index and xrnum_index to be equal to an accurate center peak position index value;

step S23, calculating index values of the left intersection point and the right intersection point respectively leftwards and rightwards based on the central peak position:

step S231, searching for a sequence length of a left intersection point to be search_peak_length/2;

When y [ xlnum _index ] > (1-p) ×y [ center_peak_ raman _shift_index ], each time the left intersection point x [ xlnum _index ] moves leftwards by one step length, namely the index value is subtracted by 1, judging whether the x [ xlnum _index ] is in the sequence length range of the left intersection point at the moment, if so, continuing to move leftwards to search for the intersection point;

Where x [ xlnum _index ] represents the corresponding raman shift at index xlnum _index, y [ xlnum _index ] represents the corresponding raman intensity at index xlnum _index, and y [ center_peak_ raman _shift_index ] represents the corresponding raman intensity at index center_peak_ raman _shift_index;

Stopping when y [ xlnum _index ] < = (1-p) [ center_peak_ raman _shift_index ] or x [ xlnum _index ] at the moment exceeds the sequence length range of the left intersection point, wherein xlnum _index is the left intersection point index value;

step S232, searching a right intersection point, wherein the sequence length is search_peak_length/2;

When y [ xrnum _index ] > (1-p) ×y [ center_peak_ raman _shift_index ], every time the right intersection point x [ xrnum _index ] moves to the right by one step length, namely, adding 1 to the index value, judging whether the x [ xrnum _index ] is in the range of the sequence length of the right intersection point at the moment, if so, continuing to move to the right to search for the intersection point;

Stopping when y [ xrnum _index ] < = (1-p) [ center_peak_ raman _shift_index ] or x [ xrnum _index ] at the moment already exceeds the sequence length range of the right intersection point, wherein xrnum _index is the index value of the right intersection point;

characteristic peak both sides tangential point determining unit: the accurate coordinates of tangential points on two sides of the characteristic peak are calculated based on the intersection points on two sides of the characteristic peak, and the specific process comprises the following steps:

S31, calculating a horizontal distance Deltax _L、△x_R between the central peak position and the two intersection points;

Step S32, the starting values of the left and right tangent point coordinate index values xllnum _index and xrrnum_index are respectively equal to xlnum _index and xrnum_index; the left and right tangent points are left and right end points for calculating the peak area of the characteristic peak of the Raman spectrum;

Step S33, moving the left tangent point x [ xllnum _index ] one step to the left each time, i.e. subtracting 1 from the index value, and finding the left tangent point index value xllnum _index of the characteristic peak when x [ xllnum _index ] =x [ center_peak_ raman _shift_index ] - (1/p) ×Δx _L;

step S34, right tangent point x [ xrrnum _index ] is shifted one step to the right each time, i.e. index value is added by 1, when x [ xrrnum _index ] =x [ center_peak_ raman _shift_index ] + (1/p) Δx _R, finding the index value xrrnum _index of right tangent point of characteristic peak;

The coordinates of the left and right tangent points in step S35 are (x [ xllnum _index ], y [ xllnum _index ]), (x [ xrrnum _index ], y [ xrrnum _index ]);

Raman characteristic peak area calculation unit: and calculating the area of an area surrounded by the Raman characteristic peak and the connecting line of the two tangent points based on the coordinates of the left tangent point and the right tangent point.

6. The system for determining peak areas of features of a raman spectrum without subtracting raman background according to claim 5 wherein the process of determining the maximum value of raman intensity within the length of the central peak position search range sequence comprises the steps of:

firstly, a step divis of RAMAN SHIFT is obtained, and then a Raman intensity maximum value is determined in the center peak position searching range sequence length according to the step divis and is used as the accurate center peak position of the characteristic peak to be extracted.

7. The system for determining the peak area of a characteristic peak of a raman spectrum without subtracting a raman background according to claim 5 or 6, wherein the process of calculating the area of a region surrounded by a raman characteristic peak and a line connecting two tangential points based on coordinates of the left tangential point and the right tangential point comprises the following steps:

Step S41, taking a left tangent point of a characteristic peak as a starting point, taking Raman intensity values y [ xllnum _index ] and y [ xllnum _index+1] of two adjacent points as upper and lower bottoms, taking a Raman step length as high, and calculating the area of each small trapezoid until reaching a right tangent point to be cut off, wherein the sum of the areas of the small trapezoids is recorded as S _{Total (S)} ;

step S42, determining the area S _{Back of body} of the Raman background based on the left and right tangential points of the characteristic peak;

The area obtained by subtracting S _{Back of body} from steps S43 and S _{Total (S)} is the peak area of the raman characteristic peak.

8. The system for determining the characteristic peak-to-peak area of a raman spectrum without subtracting the raman background according to claim 7, wherein the process for determining the area S _{Back of body} of the raman background based on the tangential points of the left and right of the characteristic peak comprises the steps of:

The Raman intensity values y [ xllnum _index ] and y [ xrrnum _index ] of the left and right tangential points of the characteristic peak are taken as the upper bottom and the lower bottom, the horizontal distance between the two tangential points is high, and the calculated area of the trapezoid is recorded as S _{Back of body} , namely the area of the Raman background.

9. A computer storage medium having stored therein at least one instruction loaded and executed by a processor to implement the method of determining raman spectral feature peak areas without subtracting raman background according to any one of claims 1 to 4.

10. An apparatus for determining a raman spectral feature peak area without a subtracted raman background, the apparatus comprising a processor and a memory having stored therein at least one instruction loaded and executed by the processor to implement the method for determining a raman spectral feature peak area without a subtracted raman background as claimed in any one of claims 1 to 4.