Quality classification method of kummerowia striata based on infrared spectrum clustering analysis

Abstract

The invention discloses an infrared spectrum clustering analysis-based quality classification method for kummerowia striata, which comprises the following steps of: acquiring original spectrum data of kummerowia striata based on an infrared spectrometer; carrying out smoothing processing and correction processing on the original spectral data to obtain corrected spectral data; identifying the position and the intensity of an absorption peak from the corrected spectral data, and judging whether an overlapping peak or an acromion exists or not; if overlapping peaks or shoulder peaks exist, separating the overlapping peaks by adopting a peak separation algorithm, and extracting independent absorption peak characteristics; carrying out clustering analysis on the extracted absorption peak characteristics to obtain a clustering analysis result; based on the clustering analysis result, establishing a classification model; and classifying the newly collected spectral data based on the classification model to obtain the quality grade of the kummerowia striata. According to the method, the infrared spectrum technology is combined with the data processing and machine learning methods, so that objective, accurate and efficient evaluation of the quality of the kummerowia striata is realized, and powerful support is provided for quality control of the kummerowia striata industry.

Description

Technical Field

[0001] The invention belongs to the technical field of quality classification of traditional Chinese medicines, and in particular relates to a quality classification method of herbaceous vine based on infrared spectroscopy cluster analysis. Background Art

[0002] Traditional methods for the quality classification of traditional Chinese medicinal materials (TCMs) often rely on sensory evaluation or chemical analysis, which suffer from drawbacks such as low efficiency, subjectivity, and time-consuming nature. In recent years, with the development of spectroscopy technology, infrared spectroscopy, due to its rapidity, non-destructive nature, and high information content, has gradually been applied to the identification and classification of TCMs. Infrared spectroscopy can reveal the characteristic absorption peaks of chemical components in TCMs, and mathematical methods such as cluster analysis can be used to rapidly classify TCMs.

[0003] However, for Chinese herbal medicines with unique medicinal value, such as Herba Lycopodii, efficient and accurate technical methods for quality classification remain lacking. Existing research has primarily focused on origin identification or component analysis of common Chinese herbal medicines, while relatively little research has focused on the quality classification of Herba Lycopodii. Therefore, developing a Herba Lycopodii quality classification method based on infrared spectroscopy cluster analysis is crucial for improving Herba Lycopodii's quality control and safeguarding its medicinal value. Summary of the Invention

[0004] In order to solve the above technical problems, the present invention proposes a quality classification method of herbaceous grass based on infrared spectroscopy cluster analysis to solve the problems existing in the above-mentioned prior art.

[0005] To achieve the above object, the present invention provides a quality classification method of herbaceous herbaceous grass based on infrared spectroscopy cluster analysis, comprising the following steps:

[0006] Based on infrared spectrometer, the original spectrum data of herringbone grass is obtained;

[0007] performing smoothing and correction processing on the original spectral data to obtain corrected spectral data;

[0008] From the corrected spectral data, identify the position and intensity of the absorption peaks and determine whether there are overlapping peaks or shoulders;

[0009] If there are overlapping peaks or shoulder peaks, the peak separation algorithm is used to separate the overlapping peaks and extract independent absorption peak features;

[0010] Perform cluster analysis on the extracted absorption peak features to obtain cluster analysis results;

[0011] Based on the cluster analysis results, establishing a classification model;

[0012] The newly collected spectral data were classified based on the classification model to obtain the quality grade of herringbone grass.

[0013] Optionally, the process of smoothing the original spectral data includes:

[0014] The random noise in the original spectral data is smoothed based on the preset window width to obtain the denoised spectral data.

[0015] Optionally, the process of performing correction processing on the original spectral data includes:

[0016] The baseline correction algorithm is used to process the denoised spectral data to eliminate the baseline drift and obtain the corrected spectral data.

[0017] Optionally, the absorption peak characteristics include peak height, peak width and peak area.

[0018] Optionally, cluster analysis is performed on the extracted absorption peak features, and a process of obtaining cluster analysis results includes:

[0019] Based on the preset quality indicators, it is judged whether the extracted peak height, peak width and peak area meet the standard range; if they meet the standard range, the classification method is used to classify and manage the absorption peak characteristics to generate structured data; based on the structured data, the support vector machine algorithm is used to identify the absorption peak marching mode to obtain the spectral characteristic pattern; according to the spectral characteristic pattern, the preset clustering algorithm is selected to perform cluster analysis on the absorption peak characteristics, determine the number of clusters, and then obtain the cluster analysis results.

[0020] Optionally, the process of classifying the newly collected spectral data based on the classification model to obtain the quality grade of the herringbone grass includes:

[0021] The classification model is used to process the newly collected spectral data and extract the spectral features. According to the spectral features, the preset quality grades of the herringbone grass are matched to obtain the classification results.

[0022] The present invention also provides a quality classification system of herbaceous grass based on infrared spectrum cluster analysis, which is used to implement the method described above, comprising: a data acquisition module, a data processing module, a feature extraction module, a cluster analysis module and a quality classification module;

[0023] The data acquisition module is used to obtain original spectrum data of the herringbone grass based on an infrared spectrometer;

[0024] The data processing module is used to perform smoothing and correction processing on the original spectral data to obtain corrected spectral data;

[0025] The feature extraction module is used to identify the position and intensity of the absorption peak from the corrected spectral data and determine whether there are overlapping peaks or shoulder peaks; if there are overlapping peaks or shoulder peaks, a peak separation algorithm is used to separate the overlapping peaks and extract independent absorption peak features;

[0026] The cluster analysis module is used to perform cluster analysis on the extracted absorption peak features to obtain cluster analysis results;

[0027] The quality classification module is used to establish a classification model based on the cluster analysis results; classify the newly collected spectral data based on the classification model to obtain the quality grade of the herringbone grass.

[0028] The present invention also provides a computer device, comprising a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method.

[0029] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the method when executed by a processor.

[0030] The present invention also provides a computer program product, comprising a computer program, characterized in that the computer program implements the steps of the method when executed by a processor.

[0031] Compared with the prior art, the present invention has the following advantages and technical effects:

[0032] The present invention discloses a quality classification method for herbaceous herb based on infrared spectral cluster analysis. The method first collects infrared spectral data of herbaceous herb samples and controls the spectrometer to obtain raw data through preset parameters. The raw data is then denoised and baseline corrected to extract key absorption peak features. A peak separation algorithm is used to separate overlapping peaks and shoulder peaks. Cluster analysis is used to establish a correspondence between the quality grade of herbaceous herb and the spectral characteristics, and a classification model is constructed. Finally, the newly collected spectral data is input into the model to achieve rapid identification of the quality grade of herbaceous herb. The present invention combines infrared spectroscopy technology with data processing and machine learning methods to achieve objective, accurate and efficient evaluation of the quality of herbaceous herb, providing strong support for quality control of the herbaceous herb industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The accompanying drawings, which constitute part of this application, are intended to provide a further understanding of this application. The exemplary embodiments and descriptions of this application are intended to explain this application and do not constitute an improper limitation on this application. In the accompanying drawings:

[0034] Figure 1 Schematic diagram of a method flow in an embodiment of the present invention. DETAILED DESCRIPTION

[0035] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in this application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0036] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0037] Example 1

[0038] like Figure 1 As shown, this embodiment provides a quality classification method of herbaceous grass based on infrared spectroscopy cluster analysis, comprising the following steps:

[0039] Based on infrared spectrometer, the original spectrum data of herringbone grass is obtained;

[0040] performing smoothing and correction processing on the original spectral data to obtain corrected spectral data;

[0041] From the corrected spectral data, identify the position and intensity of the absorption peaks and determine whether there are overlapping peaks or shoulders;

[0042] If there are overlapping peaks or shoulder peaks, the peak separation algorithm is used to separate the overlapping peaks and extract independent absorption peak features;

[0043] Perform cluster analysis on the extracted absorption peak features to obtain cluster analysis results;

[0044] Based on the cluster analysis results, establishing a classification model;

[0045] The newly collected spectral data were classified based on the classification model to obtain the quality grade of herringbone grass.

[0046] As an implementable approach, the following steps are included:

[0047] S101. According to a preset wavelength range and resolution value, the number of scans of the infrared spectrometer is set to obtain original spectrum data of the herringbone grass.

[0048] Infrared spectrometers are important analytical instruments used to study the molecular structure and chemical composition of substances. When using an infrared spectrometer, the appropriate wavelength range and resolution must be determined based on the sample characteristics and analytical objectives. The resolution chosen depends on the desired level of detail. After determining the measurement parameters, a scanning operation is performed to acquire the spectral signal. During this process, the instrument illuminates the sample with infrared light and then measures the intensity of the transmitted or reflected light.

[0049] S102 : Smoothing the random noise in the original spectral data using a preset window width to obtain denoised spectral data.

[0050] Infrared spectral data processing is a complex task involving multiple steps and techniques. First, smoothing the raw spectral data is crucial for reducing the effects of random noise. Common smoothing methods include moving average and Savitsky-Golay smoothing. For example, using the moving average method, an appropriate window width, such as five data points, is selected. Each data point and its surrounding data are averaged to produce a smoothed spectral curve.

[0051] S103 , applying a baseline correction algorithm to the denoised spectral data to eliminate baseline drift and obtain corrected spectral data.

[0052] Spectral data processing is a complex process involving multiple steps and techniques. First, baseline correction algorithms are used to eliminate baseline drift in the spectrum. For example, in Raman spectroscopy, fluorescence background can cause baseline shifts, affecting peak identification. Methods such as polynomial fitting or wavelet transforms can effectively remove this interference, making spectral peaks more prominent.

[0053] S104 , identifying the position and intensity of the absorption peak from the corrected spectral data, and determining whether there are overlapping peaks or shoulder peaks.

[0054] The corrected spectral data is processed using a peak identification algorithm to obtain the position and intensity of the absorption peaks. Based on the absorption peak position, the presence of overlapping peaks or shoulder peaks in the spectrum is determined.

[0055] S105. If there are overlapping peaks or shoulder peaks, use a peak separation algorithm to separate the overlapping peaks and extract independent absorption peak features.

[0056] Peak identification algorithms are an important technique in spectral analysis, used to accurately locate the position and intensity of absorption peaks. Common peak identification methods include derivative methods and wavelet transform methods. The derivative method determines the peak position by calculating the first or second derivative of the spectral curve, while the wavelet transform method uses wavelet functions of different scales to decompose the signal to achieve accurate peak positioning. In practical applications, spectral data often have overlapping peaks or shoulder peaks. Overlapping peaks refer to the partial overlap of two or more absorption peaks, while shoulder peaks refer to smaller peaks that appear next to the main peak. In order to solve the problem of overlapping peaks, peak separation algorithms are widely used. Common peak separation methods include Gaussian fitting and Fourier self-convolution. The Gaussian fitting method assumes that each single peak conforms to the Gaussian distribution and fits the overlapping peaks through iterative optimization. The Fourier self-convolution method uses the properties of the Fourier transform to convert overlapping peaks into the frequency domain, and then obtains the separated single peaks through inverse transformation.

[0057] Furthermore, extracting independent absorption peak features is a key step in spectral analysis. Peak height reflects absorption intensity and can be used for quantitative analysis; peak width is related to molecular structure and environment and can provide information on sample status; peak area is proportional to the substance content and is often used for quantitative determination. The setting of quality indicators is crucial to ensuring the reliability of analytical results. Generally, the relative standard deviation (RSD) of peak height should be less than 5%, the RSD of peak width should be less than 10%, and the RSD of peak area should be less than 3%. These indicators can be adjusted according to the specific application scenario.

[0058] S106 , selecting a preset clustering algorithm to perform cluster analysis based on the extracted absorption peak features, and determining the number of clusters.

[0059] Specifically, peak height, peak width, and peak area are extracted from independent absorption peak features. Pre-set quality indicators are used to determine whether the extracted peak height, peak width, and peak area meet standard ranges. If the characteristic parameters meet the standard ranges, a classification method is used to classify and manage the absorption peak data to generate structured data. Based on the structured data, a support vector machine algorithm is used to perform pattern recognition on the absorption peaks to obtain spectral characteristic patterns. Based on the spectral characteristic patterns, a preset clustering algorithm is selected to perform cluster analysis on the absorption peaks and determine the number of clusters.

[0060] Extracting peak height, peak width, and peak area from independent absorption peaks is a crucial step in spectral analysis. These parameters need to be compared with pre-set quality indicators to ensure data reliability. Absorption peak data that meets the standards is classified and managed to generate structured data, laying the foundation for subsequent pattern recognition. For example, characteristic peaks of different functional groups can be classified and stored, such as the OH peak of alcohols and the C=O peak of ketones. This structured data facilitates the application of machine learning algorithms, such as support vector machines (SVM). SVM algorithms are widely used in spectral analysis, especially in the identification of complex mixtures.

[0061] Furthermore, cluster analysis is an unsupervised learning method used to group similar data points. In spectral analysis, clustering can help identify spectra with similar characteristics. Determining the number of clusters is a key step, and methods such as the silhouette coefficient or the elbow method can be used. Taking the elbow method as an example, by plotting the sum of squared errors for different numbers of clusters, the optimal number of clusters can be determined at the inflection point of the curve.

[0062] S107. Establish a classification model based on the cluster analysis results, and correspond the clustering results to the quality grades of the herringbone grass.

[0063] Based on the determined number of clusters, a classification model can be established using algorithms such as support vector machines (SVMs) or random forests. These models can learn the relationship between spectral features and quality levels. For example, using SVMs, the optimal separating hyperplane can be found in a high-dimensional feature space to separate samples of different quality levels.

[0064] S108. Apply the classification model to classify the newly collected spectral data to obtain the quality grade of the herringbone grass.

[0065] It is feasible to use a classification model to process the newly collected spectral data and extract spectral features. According to the spectral features, the preset herringbone grass quality grade is matched to obtain the classification result.

[0066] This spectral-based quality assessment method not only enables rapid and non-destructive analysis, but also captures subtle differences that may be missed by traditional chemical analysis. Through continued optimization and data accumulation, this method is expected to become a powerful tool for quality control of herringbone grass, providing a scientific basis for industrial development.

[0067] This embodiment also provides a quality classification system for herbaceous grass based on infrared spectroscopy cluster analysis, which is used to implement the method described above, including: a data acquisition module, a data processing module, a feature extraction module, a cluster analysis module and a quality classification module;

[0068] The data acquisition module is used to obtain original spectrum data of the herringbone grass based on an infrared spectrometer;

[0069] The data processing module is used to perform smoothing and correction processing on the original spectral data to obtain corrected spectral data;

[0070] The feature extraction module is used to identify the position and intensity of the absorption peak from the corrected spectral data and determine whether there are overlapping peaks or shoulder peaks; if there are overlapping peaks or shoulder peaks, a peak separation algorithm is used to separate the overlapping peaks and extract independent absorption peak features;

[0071] The cluster analysis module is used to perform cluster analysis on the extracted absorption peak features to obtain cluster analysis results;

[0072] The quality classification module is used to establish a classification model based on the cluster analysis results; classify the newly collected spectral data based on the classification model to obtain the quality grade of the herringbone grass.

[0073] Example 2

[0074] This embodiment further provides a computer device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method.

[0075] Example 3

[0076] This embodiment also provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the method when executed by a processor.

[0077] Example 4

[0078] This embodiment also provides a computer program product, including a computer program, characterized in that the steps of the method are implemented when the computer program is executed by a processor.

[0079] The above are merely preferred embodiments of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A quality classification method for herbaceous grass based on infrared spectroscopy cluster analysis, characterized in that: The following steps are involved: Based on infrared spectrometer, the original spectrum data of herringbone grass is obtained; performing smoothing and correction processing on the original spectral data to obtain corrected spectral data; From the corrected spectral data, identify the position and intensity of the absorption peaks and determine whether there are overlapping peaks or shoulders; If there are overlapping peaks or shoulder peaks, the peak separation algorithm is used to separate the overlapping peaks and extract independent absorption peak features; Perform cluster analysis on the extracted absorption peak features to obtain cluster analysis results; Based on the cluster analysis results, establishing a classification model; The newly collected spectral data were classified based on the classification model to obtain the quality grade of herringbone grass.

2. The method according to claim 1, characterized in that The process of smoothing the raw spectral data includes: The random noise in the original spectral data is smoothed based on the preset window width to obtain the denoised spectral data.

3. The method according to claim 2, characterized in that The process of correcting the original spectral data includes: The baseline correction algorithm is used to process the denoised spectral data to eliminate the baseline drift and obtain the corrected spectral data.

4. The method according to claim 1, wherein The absorption peak characteristics include peak height, peak width and peak area.

5. The method according to claim 4, characterized in that The process of performing cluster analysis on the extracted absorption peak features and obtaining cluster analysis results includes: Based on the preset quality indicators, it is judged whether the extracted peak height, peak width and peak area meet the standard range; if they meet the standard range, the classification method is used to classify and manage the absorption peak characteristics to generate structured data; based on the structured data, the support vector machine algorithm is used to identify the absorption peak marching mode to obtain the spectral characteristic pattern; according to the spectral characteristic pattern, the preset clustering algorithm is selected to perform cluster analysis on the absorption peak characteristics, determine the number of clusters, and then obtain the cluster analysis results.

6. The method according to claim 1, characterized in that The process of classifying the newly collected spectral data based on the classification model and obtaining the quality grade of herringbone grass includes: The classification model is used to process the newly collected spectral data and extract the spectral features. According to the spectral features, the preset quality grades of the herringbone grass are matched to obtain the classification results.

7. A quality classification system for Herba Lycopodii based on infrared spectroscopy cluster analysis, characterized in that: Used to implement the method according to any one of claims 1 to 6, comprising: a data acquisition module, a data processing module, a feature extraction module, a cluster analysis module and a quality classification module; The data acquisition module is used to obtain original spectrum data of the herringbone grass based on an infrared spectrometer; The data processing module is used to perform smoothing and correction processing on the original spectral data to obtain corrected spectral data; The feature extraction module is used to identify the position and intensity of the absorption peak from the corrected spectral data and determine whether there are overlapping peaks or shoulder peaks; if there are overlapping peaks or shoulder peaks, a peak separation algorithm is used to separate the overlapping peaks and extract independent absorption peak features; The cluster analysis module is used to perform cluster analysis on the extracted absorption peak features to obtain cluster analysis results; The quality classification module is used to establish a classification model based on the cluster analysis results; classify the newly collected spectral data based on the classification model to obtain the quality grade of the herringbone grass.

8. A computer device comprising a memory, a processor, and a computer program stored in the memory, wherein: The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images