Fruit spice detection method and device based on gas chromatography-ion mobility spectrometry algorithm

Abstract

The invention provides a fruit spice detection method and device based on a gas chromatography-ion mobility spectrometry algorithm, and can be applied to the technical field of fruit spice detection. The method comprises the following steps: performing fingerprint extraction on to-be-detected fruit flavor data based on a gas chromatography-ion mobility spectrometry algorithm to obtain a fingerprint; detecting the fingerprint spectrum by using a detection model to obtain a plurality of target volatile components; the multiple target volatile components, preset signal peak difference data and the volatile component markers are analyzed, fruits corresponding to the to-be-detected fruit flavor data are obtained, and the preset signal peak difference data represent sample signal peak data differences of different fruits for the preset volatile components; the volatile component marker is obtained by screening from the plurality of preset volatile components based on a preset projection variable importance threshold and the respective projection variable importance of the plurality of preset volatile components.

Description

Technical Field

[0001] This application relates to the field of fruit flavoring detection technology, specifically to a method and apparatus for detecting fruit flavorings based on gas chromatography ion mobility spectrometry algorithm. Background Technology

[0002] Natural flavoring raw materials are an important component of tobacco flavorings and fragrances. Fruit-based flavoring raw materials have complex compositions and vary in quality. Currently, relevant research is limited to the overall sensory understanding of fruit-based flavoring raw materials, lacking scientific, objective, and unified identification methods based on characteristic chemical components, making it difficult to accurately distinguish between different types of flavoring raw materials. Summary of the Invention

[0003] In view of the above problems, this application provides a method and apparatus for detecting fruit spices based on gas chromatography ion mobility spectrometry algorithm.

[0004] According to the first aspect of this application, a method for detecting fruit flavorings based on gas chromatography ion mobility spectrometry (GC-IMS) algorithm is provided, comprising: extracting fingerprint spectrum from the fruit flavoring data to be tested using GC-IMS algorithm to obtain a fingerprint spectrum, wherein the fingerprint spectrum includes signal peak data corresponding to multiple preset volatile components; detecting the fingerprint spectrum using a detection model to obtain multiple target volatile components, wherein the target volatile components are the volatile components contained in the fruit flavoring data to be tested; analyzing the multiple target volatile components, preset signal peak difference data, and volatile component markers to obtain the fruit corresponding to the fruit flavoring data to be tested, wherein the preset signal peak difference data characterizes the difference in sample signal peak data of different fruit types for preset volatile components, and the volatile component markers are selected from multiple preset volatile components based on a preset projection variable importance threshold and the projection variable importance of each of the multiple preset volatile components, wherein the projection variable importance is determined based on statistical analysis of the sample flavoring data of different fruit types for preset volatile components.

[0005] According to embodiments of this application, multiple target volatile components, preset signal peak difference data, and volatile component markers are analyzed to obtain the fruit category corresponding to the fruit flavoring data to be tested. This includes: extracting matching volatile components from multiple target volatile components that match multiple volatile component markers; processing the signal peak data of the matching volatile components using a target category discrimination model to obtain the fruit category corresponding to the fruit flavoring data to be tested. The target category discrimination model is constructed based on the difference analysis of the preset signal peak difference data.

[0006] According to embodiments of this application, determining preset signal peak difference data includes: extracting two-dimensional spectra from sample fragrance data of different fruits based on a gas chromatography ion mobility spectrometry algorithm to obtain sample two-dimensional spectra, wherein the sample two-dimensional spectra characterize the signal peak difference distribution among sample fragrance data of different fruits; marking signal peaks in the fingerprint spectrum of the target sample to obtain multiple marker boxes, wherein the marker boxes characterize the positions of signal peaks that are higher than a preset signal peak threshold; performing qualitative analysis on the multiple marker boxes to obtain multiple preset volatile components; and extracting difference spectra from sample fragrance data of different fruits based on multiple preset volatile components using a gas chromatography ion mobility spectrometry algorithm to obtain sample difference spectra, wherein the sample difference spectra characterize the signal peak difference distribution data of different fruits for sample fragrance data in the preset volatile components, and the preset signal peak difference data includes signal peak difference distribution data.

[0007] According to an embodiment of this application, the target sample fingerprint spectrum is determined based on the following operation: using a gas chromatography ion mobility spectrometry algorithm to extract fingerprint spectrum of the fragrance data of different fruits based on multiple preset volatile components, the target sample fingerprint spectrum is obtained. The target sample fingerprint spectrum represents the regional aggregation difference distribution data of the fragrance data of different fruits in multiple preset volatile components. The preset signal peak difference data also includes regional aggregation difference distribution data.

[0008] According to an embodiment of this application, a fingerprint spectrum of the fragrance data of different fruits is extracted using a gas chromatography-ion mobility spectrometry algorithm based on multiple preset volatile components to obtain a target sample fingerprint spectrum. This includes: extracting fingerprint spectra of the fragrance data of different fruits based on multiple preset volatile components using a gas chromatography-ion mobility spectrometry algorithm to obtain an initial sample fingerprint spectrum, which includes signal peak data of the fragrance data of different fruits at multiple preset volatile components; sorting the positions of the multiple preset volatile components in the spectrum according to the signal peak data to obtain signal peak aggregation difference distribution data; and dividing the initial sample fingerprint spectrum into regions based on the signal peak aggregation difference distribution data to obtain the target sample fingerprint spectrum.

[0009] According to an embodiment of this application, the importance of projected variables is determined by statistical analysis of sample spice data of different fruits on preset volatile components. This includes: using principal component analysis algorithm to process the signal peak data of different fruits on multiple preset volatile components in the fingerprint spectrum of the target sample to obtain analysis results; and when the analysis results indicate that there are differences between different fruits, using orthogonal partial least squares discriminant algorithm to process the signal peak data of different fruits on multiple preset volatile components to obtain the importance of projected variables corresponding to multiple preset volatile components.

[0010] According to an embodiment of this application, volatile component markers are screened from multiple preset volatile components based on a preset projection variable importance threshold and the respective projection variable importance of multiple preset volatile components, including: determining preset volatile components with a projection variable importance greater than the preset projection variable importance threshold as volatile component markers.

[0011] According to an embodiment of this application, on the instrument on which the gas chromatography ion mobility spectrometry algorithm is deployed, the migration gas flow rate is 75 mL / min, the carrier gas flow rate is initially maintained at 2 mL / min for 20 min, and then linearly increased to 100 mL / min within 20 to 35 min and maintained for 5 min.

[0012] The second aspect of this application provides a fruit flavoring detection device, comprising: an extraction module for extracting a fingerprint spectrum from the fruit flavoring data to be tested based on a gas chromatography ion mobility spectrometry algorithm, wherein the fingerprint spectrum includes signal peak data corresponding to multiple preset volatile components; a detection module for detecting the fingerprint spectrum using a detection model to obtain multiple target volatile components, wherein the target volatile components are the volatile components contained in the fruit flavoring data to be tested; and an identification module for analyzing the multiple target volatile components, preset signal peak difference data, and volatile component markers to obtain the fruit corresponding to the fruit flavoring data to be tested, wherein the preset signal peak difference data characterizes the difference in sample signal peak data of different fruit types for preset volatile components, and the volatile component markers are selected from multiple preset volatile components based on a preset projection variable importance threshold and the projection variable importance of each of the multiple preset volatile components, wherein the projection variable importance is determined based on statistical analysis of the sample flavoring data of different fruit types for preset volatile components.

[0013] A third aspect of this application provides an electronic device comprising: one or more processors; and a memory for storing one or more computer programs, wherein the one or more processors execute the one or more computer programs to implement the steps of the method described above.

[0014] A fourth aspect of this application also provides a computer-readable storage medium having a computer program or instructions stored thereon, which, when executed by a processor, implement the steps of the above-described method.

[0015] The fifth aspect of this application also provides a computer program product, including a computer program or instructions that, when executed by a processor, implement the steps of the above-described method.

[0016] According to the method, apparatus, equipment, medium and product for detecting fruit flavorings based on gas chromatography ion mobility spectrometry algorithm provided in this application, fingerprint spectrum is extracted from the fruit flavoring data to be tested based on gas chromatography ion mobility spectrometry algorithm to obtain fingerprint spectrum, and then a detection model is obtained by training a large model based on YOLO V8 to detect the fingerprint spectrum and identify multiple target volatile components contained in the fruit flavoring data. Due to its core advantages such as high sensitivity, rapid response, and strong anti-interference ability, the gas chromatography ion mobility spectrometry algorithm can achieve efficient separation and accurate quantification of fruit flavoring data, significantly improving the accuracy and reliability of fingerprint extraction. Combined with the detection model, it enables rapid and accurate identification of target volatile components in different types of fruit flavorings. In addition, based on a large number of sample flavoring data from different fruits, volatile component markers that contribute significantly to category discrimination are extracted. Then, the volatile component markers obtained by statistical analysis are compared with the target volatile components. By combining the differences in sample signal peak data of different fruits for preset volatile components, the category accuracy is achieved, providing scientific and effective support for the identification of fruit flavoring varieties, liquid analysis, and quality control. Attached Figure Description

[0017] The above-mentioned contents, other objects, features and advantages of this application will become clearer from the following description of embodiments with reference to the accompanying drawings, in which:

[0018] Figure 1 A flowchart of a fruit flavoring detection method based on a gas chromatography ion mobility spectrometry algorithm according to an embodiment of this application is shown;

[0019] Figure 2 A schematic diagram of regression curves according to an embodiment of this application is shown;

[0020] Figure 3 A schematic diagram of a two-dimensional atlas according to an embodiment of this application is shown;

[0021] Figure 4 A schematic diagram of a difference map according to an embodiment of this application is shown;

[0022] Figure 5 A schematic diagram of a three-dimensional atlas according to an embodiment of this application is shown;

[0023] Figure 6 A schematic diagram of a target sample fingerprint spectrum according to an embodiment of this application is shown;

[0024] Figure 7 A schematic diagram illustrating the analysis results according to an embodiment of this application is shown;

[0025] Figure 8 A schematic diagram showing the order of importance of projection variables according to an embodiment of this application is provided;

[0026] Figure 9 A structural block diagram of a fruit fragrance detection device according to an embodiment of this application is shown; and

[0027] Figure 10 A block diagram of an electronic device suitable for implementing a method for detecting fruit spices based on a gas chromatography ion mobility spectrometry algorithm, according to an embodiment of this application, is shown. Detailed Implementation

[0028] The embodiments of this application will now be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of this application. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of this application for ease of explanation. However, it will be apparent that one or more embodiments may be implemented without these specific details. Furthermore, descriptions of well-known structures and technologies are omitted in the following description to avoid unnecessarily obscuring the concepts of this application.

[0029] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of this application. The terms “comprising,” “including,” etc., as used herein indicate the presence of features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0030] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein are to be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.

[0031] When using expressions such as "at least one of A, B and C", they should generally be interpreted in accordance with the meaning that is commonly understood by those skilled in the art (e.g., "a system having at least one of A, B and C" should include, but is not limited to, a system having A alone, a system having B alone, a system having C alone, a system having A and B, a system having A and C, a system having B and C, and / or a system having A, B and C, etc.).

[0032] In the process of developing this application, it was discovered that while gas chromatography-mass spectrometry (GC-MS) is widely used in the analysis of flavoring raw materials, it suffers from drawbacks such as long detection cycles, complex sample pretreatment procedures, and insufficient sensitivity to trace volatile components. In particular, for characteristic volatile components in fruit flavoring raw materials that are present in extremely low amounts but determine aroma characteristics, as well as some structurally similar isomers, GC-MS struggles to achieve efficient separation and accurate quantification, failing to meet the practical needs of rapid identification of fruit flavoring raw material types and rapid analysis of liquid components in cigarette production.

[0033] In view of this, embodiments of this application provide a method for detecting fruit flavorings based on gas chromatography ion mobility spectrometry (GC-IMS), comprising: extracting fingerprint spectra from the fruit flavoring data to be tested using GC-IMS to obtain fingerprint spectra; detecting the fingerprint spectra using a detection model to obtain multiple target volatile components; analyzing the multiple target volatile components, preset signal peak difference data, and volatile component markers to obtain the fruit corresponding to the fruit flavoring data to be tested, wherein the preset signal peak difference data characterizes the differences in sample signal peak data of different fruit for preset volatile components, and the volatile component markers are selected from multiple preset volatile components based on a preset projection variable importance threshold and the importance of the projection variables of each of the multiple preset volatile components.

[0034] The gas chromatography-ion mobility spectrometry (GC-IMS) algorithm ingeniously combines the high separation efficiency of GC with the core advantages of IMS, such as high sensitivity, rapid response, and strong anti-interference capabilities, significantly improving the accuracy and reliability of analytical results. Even characteristic volatile components with similar structures and boiling points in fruit flavorings (such as isomers of key aroma components like esters, aldehydes, and ketones) can be efficiently separated and accurately identified; its detection sensitivity can reach the ppb level (10⁻⁻¹). 9 The GC-MS technique (g / mL) can accurately capture trace volatile components in fruit flavorings, which are present in extremely low concentrations but play a decisive role in aroma characteristics. This advantage precisely compensates for the shortcomings of traditional GC-MS technology in the detection of trace components. Furthermore, it can construct fingerprint profiles for the complex volatile organic compound (VOC) system in fruit flavorings, enabling rapid differentiation of different types of fruit flavorings by analyzing the differences in their fingerprint profiles. The fingerprint profiles of the samples are then segmented and labeled to form a training dataset. Fine-tuning and validation of a target detection model (You Only Look Once v8, Yolo V8) are then conducted to build a detection model. This model is then deployed using the high-performance inference engine ONNXRuntime to achieve rapid determination of the presence or absence of volatile components in the fingerprint profile.

[0035] In the technical solution of this application, the user information (including but not limited to user personal information, user image information, user device information, such as location information) and data (including but not limited to data used for analysis, stored data, and displayed data) involved are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, storage, use, processing, transmission, provision, disclosure, and application of related data all comply with relevant laws, regulations, and standards, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entry points for users to choose to authorize or refuse.

[0036] In scenarios involving automated decision-making using personal information, the methods, devices, and systems provided in this application all offer users corresponding entry points for choosing to agree to or reject the automated decision-making results. If the user chooses to reject, the process proceeds to the expert decision-making stage. Here, "automated decision-making" refers to the activity of automatically analyzing and evaluating an individual's behavioral habits, interests, or economic, health, and credit status through computer programs, and then making a decision. Here, "expert decision-making" refers to the activity of making decisions by personnel who specialize in a particular field, possess specialized experience, knowledge, and skills, and have reached a certain level of professional expertise.

[0037] Figure 1 A flowchart of a fruit flavoring detection method based on a gas chromatography ion mobility spectrometry algorithm according to an embodiment of this application is shown.

[0038] like Figure 1 As shown, the fruit flavoring detection method based on gas chromatography ion mobility spectrometry algorithm in this embodiment includes operations S110 to S130.

[0039] In operation S110, fingerprint spectrum extraction is performed on the test fruit fragrance data based on gas chromatography ion mobility spectrometry algorithm to obtain fingerprint spectrum.

[0040] During operation of S120, the fingerprint spectrum is detected using a detection model to obtain multiple target volatile components.

[0041] In operation S130, multiple target volatile components, preset signal peak difference data, and volatile component markers are analyzed to obtain the fruit types corresponding to the fruit flavor data to be tested.

[0042] The gas chromatography ion mobility spectrometry algorithm is Gas Chromatography–Ion Mobility Spectrometry (GC-IMS).

[0043] The data for fruit flavorings to be tested refers to the fruit flavoring raw materials or flavor base samples of the category to be tested. For example, grape extracts, plum extracts, etc.

[0044] Two independent instruments, gas chromatograph (GC) and ion mobility spectrometer (IMS), are connected in series to form a separation and detection instrument to process the data of the fruit flavorings to be tested, and output the fingerprint spectrum of the data of the fruit flavorings to be tested. The gas chromatography ion mobility spectrometry algorithm is deployed on the separation and detection instrument.

[0045] The preset volatile components are volatile organic small molecules identified through volatile component detection and qualitative analysis of volatile component data from sample fragrance data of different fruits. There are multiple preset volatile components. The fingerprint spectrum of the fragrance data of the fruits to be tested is a high-dimensional data matrix, which can be divided into multiple grids. Each grid corresponds to the signal peak data of one preset volatile component. The signal peak data can be peak volume data or peak area data.

[0046] The detection model can be built based on the real-time object detection model (You Only Look Once v8, YOLO V8). Sample fingerprints of the spice data from 14 fruit types were divided into grids, each corresponding to a preset volatile component. Each grid was labeled to obtain a sample label: sample label 1 indicates that the spice data of the fruit type contains this preset volatile component, and sample label 0 indicates that the spice data of the fruit type does not contain this preset volatile component. A training dataset was constructed based on the sample fingerprints and sample labels, with 80% of the training dataset used as the training set and 20% as the validation set for fine-tuning and validating the YOLO V8 model.

[0047] The detection model is used to detect the presence or absence of each preset volatile component in the fingerprint spectrum. The high-performance inference engine onnxruntime is used to infer and deploy the detection model, which improves the detection model's ability to quickly distinguish the presence or absence of volatile components in the fingerprint spectrum.

[0048] The target volatile components are the volatile components contained in the fruit flavoring data to be tested.

[0049] For example, if the preset volatile components are isopropanol D, ethyl acetate M, and 2-butylfuran, the fingerprint spectrum of the fruit flavoring data to be tested is divided into 3 grids. The 3 grids correspond to the signal peak data of isopropanol D, ethyl acetate M, and 2-butylfuran, respectively. The detection model is used to detect each grid, and the grid corresponding to isopropanol D is 1, the grid corresponding to ethyl acetate M is 1, and the grid corresponding to 2-butylfuran is 0. Then, the target volatile components are found to be isopropanol D and ethyl acetate M, which means that the fruit flavoring data to be tested contains isopropanol D and ethyl acetate M.

[0050] Preset signal peak difference data characterizes the differences in sample signal peak data of different fruits for preset volatile components. The preset signal peak difference distribution includes multiple preset signal peak difference data, which is obtained based on multi-dimensional spectral extraction and analysis of sample fragrance data of different fruits. For example, the preset signal peak difference data may also include the sample signal peak data differences of each fruit for multiple preset volatile components.

[0051] The volatile component markers are selected from multiple preset volatile components based on a preset importance threshold for projected variables and the importance of the projected variables for each preset volatile component. For example, the preset importance threshold for projected variables can be 1.

[0052] The higher the Variable Importance in Projection (VIP) of a predefined volatile component, the greater its contribution to the significant differences in flavor among different fruits. Volatile component markers characterize volatile components that significantly contribute to fruit flavor data.

[0053] The importance of the projected variables was determined based on statistical analysis of the spice data of different fruit samples at the preset volatile components.

[0054] For example, statistical analysis can include orthogonal partial least squares discriminant analysis, etc.

[0055] The preset signal peak difference data includes the differences in sample signal peak data of different fruits for a certain volatile component, or the differences in sample signal peak data of a certain fruit for different volatile components. Based on the difference analysis of the preset signal peak difference data, a category discrimination model can be constructed based on the orthogonal partial least squares discriminant analysis method. The category discrimination model is trained and validated based on the sample spice data of different fruits to obtain the target category discrimination model.

[0056] The discrimination framework comprises a separation and detection instrument, a detection model, and a target category discrimination model. During the validation process of this application's discrimination framework, the separation and detection instrument was used to analyze samples of 20 kinds of fruit aroma raw materials and 10 kinds of fragrance base samples. The detection model was used to quickly determine the target volatile components of the samples. Volatile components related to volatile component markers were extracted from multiple target volatile components of the samples. The signal peak data of the extracted volatile components were then input into the target category discrimination model for fruit category prediction. The category discrimination results of the discrimination framework were basically consistent with the label types of the sample fruit aroma raw materials and fragrance base samples, with an accuracy of 93%, indicating that the discrimination framework for different kinds of fruit aroma raw materials established in this application is effective.

[0057] In the application of the target category discrimination model, volatile components related to volatile component markers are extracted from multiple target volatile components. Then, the signal peak data of the extracted volatile components are input into the target category discrimination model to obtain the fruit category corresponding to the fruit spice data to be tested.

[0058] For example, fruits include grapes, plums, or apricots.

[0059] According to embodiments of this application, a fingerprint spectrum is extracted from the fruit flavoring data based on a gas chromatography-ion mobility spectrometry (GC-IMS) algorithm. A detection model trained on a YOLO V8 large model is then used to detect the fingerprint spectrum and identify multiple target volatile components contained in the fruit flavoring data. Due to the core advantages of GC-IMS, such as high sensitivity, fast response, and strong anti-interference capabilities, it can achieve efficient separation and accurate quantification of fruit flavoring data, significantly improving the accuracy and reliability of fingerprint spectrum extraction. Combined with the detection model, it enables rapid and accurate identification of target volatile components in different types of fruit flavorings. Furthermore, volatile component markers that significantly contribute to category discrimination are extracted from a large number of sample flavoring data from different fruits. These markers are then compared with the target volatile components based on statistical analysis. The differences in sample signal peak data for preset volatile components across different fruits are used to accurately classify the components, providing scientific and effective support for variety identification, liquid analysis, and quality control of fruit flavorings.

[0060] According to embodiments of this application, multiple target volatile components, preset signal peak difference data, and volatile component markers are analyzed to obtain the fruit category corresponding to the fruit flavoring data to be tested. This includes: extracting matching volatile components from multiple target volatile components that match multiple volatile component markers; processing the signal peak data of the matching volatile components using a target category discrimination model to obtain the fruit category corresponding to the fruit flavoring data to be tested. The target category discrimination model is constructed based on the difference analysis of the preset signal peak difference data.

[0061] The target volatile component and the volatile component markers that coexist are identified as the matching volatile components.

[0062] For example, if the volatile component markers include volatile components A, B, C, and D, and the target volatile components are A, B, and E, then the matching volatile components A and B that match the multiple volatile component markers can be extracted from the multiple target volatile components.

[0063] The sample fragrance data for 14 fruits are as follows: jujube tincture, grape extract, fig extract, plum extract, apricot extract, red date extract, carob extract, fenugreek extract, raisin extract, wolfberry extract, hawthorn extract, apple extract, vanilla extract, and sea buckthorn extract. To obtain the sample fragrance data for each fruit, 0.1g of the sample fragrance data was placed in a 20ml headspace vial, 2g of water was added, the temperature was set to 70℃, and the sample was shaken at 500 rpm for 10 minutes. The injection volume was 300µL, and the sample fragrance data for each fruit were obtained in triplicate (3 replicates).

[0064] Multi-dimensional spectral extraction and chemometric analysis can be performed on the spice data of 14 types of fruits to obtain the preset signal peak difference distribution. By performing principal component analysis on the preset signal peak difference data in the preset signal peak difference distribution, it can be determined whether there are significant differences between fruits. Based on the existence of significant differences, orthogonal partial least squares discriminant analysis is used to obtain the target category discrimination model.

[0065] Figure 2 A schematic diagram of the regression curve according to an embodiment of this application is shown.

[0066] like Figure 2 As shown, the target category discrimination model is validated. The horizontal axis represents signal peak data (X variable), and the vertical axis represents the fruit category (Y variable). The explanatory power R of the X variable is... 2 X = 0.989, the explanatory power R of variable Y 2 Y = 0.989, model prediction parameter Q 2 =0.973, R 2 A value close to 1 indicates a good model fit. The permutation test is used to further validate the target category discrimination model. After 200 permutation tests, two regression curves are obtained, showing that R0... 2 Q 2 The two regression lines have steep slopes and intersect the ordinate at (0.0, 0.269) and (0.0, -0.646) respectively. It can be seen that all the left-hand R... 2 Q 2 The values ​​are all lower than the original point on the right, Q 2 The regression curve intersects the vertical axis below 0.05, and R0... 2 The values ​​are all greater than 0, indicating that the established model has good predictive ability.

[0067] By using a target category discrimination model to process the signal peak data of matching volatile components, the fruit category corresponding to the fruit spice data to be tested can be obtained.

[0068] According to embodiments of this application, a difference analysis is performed on preset signal peak difference data to determine whether there are significant differences between fruit types. Based on the existence of significant differences, orthogonal partial least squares discriminant analysis is used to obtain a target category discrimination model. Thus, the target category discrimination model concentrates category signals into principal components and noise into orthogonal components, reducing overfitting and improving discrimination accuracy. In addition, the target category discrimination model processes signal peak data of matching volatile components related to volatile component markers, further improving discrimination accuracy.

[0069] According to an embodiment of this application, on the instrument on which the gas chromatography ion mobility spectrometry algorithm is deployed, the migration gas flow rate is 75 mL / min, the carrier gas flow rate is initially maintained at 2 mL / min for 20 min, and then linearly increased to 100 mL / min within 20 to 35 min and maintained for 5 min.

[0070] For the separation and detection instrument deployed by the gas chromatography ion mobility spectrometry algorithm, the parameters of the carrier gas flow rate program include the carrier gas flow rate and the duration (as shown in Table 1).

[0071] Table 1 shows the parameter table of the carrier gas flow rate program according to an embodiment of this application.

[0072]

[0073] The instrument parameters include the migration gas flow rate, incubation time, and inlet temperature, and the specific settings are shown in Table 2.

[0074] Table 2 shows the instrument parameter table according to an embodiment of this application.

[0075]

[0076] According to embodiments of this application, determining preset signal peak difference data includes: extracting two-dimensional spectra from sample fragrance data of different fruits based on a gas chromatography ion mobility spectrometry algorithm to obtain sample two-dimensional spectra, wherein the sample two-dimensional spectra characterize the signal peak difference distribution among sample fragrance data of different fruits; marking signal peaks in the fingerprint spectrum of the target sample to obtain multiple marker boxes, wherein the marker boxes characterize the positions of signal peaks that are higher than a preset signal peak threshold; performing qualitative analysis on the multiple marker boxes to obtain multiple preset volatile components; and extracting difference spectra from sample fragrance data of different fruits based on multiple preset volatile components using a gas chromatography ion mobility spectrometry algorithm to obtain sample difference spectra, wherein the sample difference spectra characterize the signal peak difference distribution data of different fruits for sample fragrance data in the preset volatile components, and the preset signal peak difference data includes signal peak difference distribution data.

[0077] Figure 3 A schematic diagram of a two-dimensional atlas according to an embodiment of this application is shown.

[0078] like Figure 3 As shown, the vertical axis represents relative migration time, and the horizontal axis represents retention time (after normalization). The background of the entire graph is blue. The reactant ion peak (RIP) is at a relative migration time of 1.0. Each point to its left and right represents a signal peak; the darker the color, the higher the content of the substance. The two-dimensional spectrum can intuitively show the peak distribution of different types of fruit samples in GC-IMS.

[0079] Fingerprint analysis was performed on the spice data of different fruits to obtain the fingerprint spectrum of the target sample. The signal peaks of the fingerprint spectrum of the target sample were labeled to obtain multiple bounding boxes, for example, 106 bounding boxes were labeled.

[0080] For example, the preset signal peak threshold can be determined based on a combination of factors such as signal-to-noise ratio threshold, peak area threshold, peak shape threshold, or occurrence rate.

[0081] Qualitative analysis was performed on multiple bounding boxes using the software's built-in database, yielding qualitative results and 78 preset volatile components. These preset volatile components were jointly determined from the spice data of 14 fruit samples.

[0082] Qualitative results indicate that a total of 21 compounds exist in monomeric and dimer forms, including 10 ketones (12.82%), 33 esters (42.31%), 13 alcohols (16.67%), 10 aldehydes (12.82%), and 12 other types of compounds (15.38%).

[0083] The gas chromatography ion mobility spectrometry algorithm was used to extract the differential spectrum of the fragrance data of different fruits, and then multiple preset volatile components were added to the sample differential spectrum. This allowed for a direct analysis of the signal peak difference distribution data of the fragrance data of different fruits on each preset volatile component.

[0084] Figure 4 A schematic diagram of a difference map according to an embodiment of this application is shown.

[0085] like Figure 4 As shown, the vertical axis represents the relative migration time, and the horizontal axis represents the retention time (after normalization). Using one spectrum as a control, the other spectra are deducted from the parts that are the same as the control spectrum. The background after deduction is white. Red indicates that the content of the volatile component is higher than that of the control sample, and blue indicates that the content of the volatile component is lower than that of the control sample. The difference spectrum can more clearly compare the differences in volatile components of different kinds of fruit flavor raw materials.

[0086] Three-dimensional spectra were extracted from the spice data of different fruit samples using gas chromatography ion mobility spectrometry algorithm to obtain three-dimensional spectra.

[0087] Figure 5 A schematic diagram of a three-dimensional atlas according to an embodiment of this application is shown.

[0088] like Figure 5 As shown, the vertical axis represents the relative migration time, the horizontal axis represents the retention time (after normalization), and the z-axis represents the signal peak intensity. The background of the entire graph is blue, and the signal peak intensity is red, which allows for a direct observation of the intensity differences between the various signal peak intensities of the spice data from different fruit varieties.

[0089] According to the embodiments of this application, by performing two-dimensional, differential, and three-dimensional spectrum analysis on the sample spice data of 14 kinds of fruits, the differences in sample signal peak data of different fruits for a certain preset volatile component or the differences in sample signal peak data of a certain fruit for different preset volatile components can be observed intuitively, thereby determining the preset signal peak difference distribution and further supporting the construction of the subsequent target category discrimination model.

[0090] According to an embodiment of this application, the target sample fingerprint spectrum is determined based on the following operation: using a gas chromatography ion mobility spectrometry algorithm to extract fingerprint spectrum of the fragrance data of different fruits based on multiple preset volatile components, the target sample fingerprint spectrum is obtained. The target sample fingerprint spectrum represents the regional aggregation difference distribution data of the fragrance data of different fruits in multiple preset volatile components. The preset signal peak difference data also includes regional aggregation difference distribution data.

[0091] The fingerprint spectrum of the fragrance data of different fruits was extracted by gas chromatography ion mobility spectrometry algorithm to obtain multiple sub-fingerprint spectrums. Multiple preset volatile components were added to the multiple sub-fingerprint spectrums, and the multiple sub-fingerprint spectrums were regionally aggregated according to the signal peak intensity of each preset volatile component to obtain the fingerprint spectrum of the target sample.

[0092] Figure 6 A schematic diagram of a target sample fingerprint spectrum according to an embodiment of this application is shown.

[0093] like Figure 6 As shown, the vertical axis represents fruit type, and the horizontal axis represents 78 preset volatile components. The fingerprint spectrum of the target sample is divided into multiple grids, each grid corresponding to a fruit type and a preset volatile component. The presence of a center point in a grid indicates that the sample fragrance data of that fruit type includes that preset volatile component. Figure 6 The darker the color, the higher the signal peak intensity. Figure 6The system allows for a clear view of multiple color-aggregated regions, enabling a direct observation of the differential distribution of spice data from different fruit samples across multiple preset volatile component regions.

[0094] The fruits are gqzjg (goji berry extract), whgtqw (fig extract), xztqw (apricot extract), mztqw (plum extract), sztqw (hawthorn extract), pgtqw (apple extract), hlbtqw (fenugreek extract), ptgtqw (raisin extract), jdtqw (carob extract), ptjzw (grape extract), zd (jujube tincture), sjjg (sea buckthorn extract), xjltqw (vanilla extract), and hztqw (jujube extract).

[0095] The 78 preset volatile components are respectively Acetic acid ethyl ester M, Acetic acid ethylester D, isoamyl formate M, isoamyl formate D, Butanoic acid ethyl ester M, Butanoic acid ethyl ester D, Butanoic acid 3-methyl ethyl ester M, Butanoic acid 3-methyl ethyl ester D, Isobutyl 3-methylbutyrate M, Isobutyl 3-methylbutyrate D, Ethyl caproate M, Ethyl caproate D, Ethyl octanoate M, Ethyl octanoate D, Hexyl 2-methylpropanoate M, Hexyl 2-methylpropanoate D, Ethyl 2-methylpropionate M, Ethyl 2-methylpropionate D, Ethyl pentanoate M, Ethyl pentanoate D, 1-Butanol, 3-methyl-, acetate, Acetic acid butyl ester M, Acetic acid butyl ester D, Isovaleric acid, methyl ester, Isoamyl butyrate, 2-Methylbutanoic acid ethyl ester, Hexanoic acid, methyl ester, 2-methyl-1-propylacetate, hexyl formate, Butanoic acid butyl ester, methyl pentanoate, Ethyl propanoate, 2-Methylbutanoic acid, methyl ester, 2-Butanone M, 2-Butanone D, 2-Heptanone, 2-propanone, 2-hexanone M, 2-hexanone D, 3-Penten-2-one, 1-Octen-3-one, 2,6-dimethyl-4-heptanone (diisobutyl ketone), 3-Octanone, 2-Propanol M, 2-Propanol D, 1-butanol M, 1-butanol D, 1-Pentanol M, 1-Pentanol D, 1-Propanol, 2-methyl M, 1-Propanol, 2-methyl D, 2-Methyl-2-propanol, 1-Propanol M, 1-PropanolD, 2-butoxyethanol, 2-Hexanol, 3-Methyl butanal M, 3-Methyl butanal D, Butanal, 2-Furaldehyde M, 2-Furaldehyde D, (E)-2-Heptenal, (E)-2-hexen-1-al, Heptaldehyde, 1-hexanal, 2-methyl butanal, dimethyl disulfide, Pyridine, 2-ethyl-4-methylthiazole M, 2-ethyl-4-methylthiazole D. triethylamine, 2-Ethylpyridine, 2,4-Dimethylthiazole, 2-butylfuran, 2-Isopropyl-3-methoxy pyrazine, 1-Pyrroline, 2-acetyl, (1-methylethyl) benzene, gamma-Terpinene (γ-terpinene). ,

[0096] According to an embodiment of this application, a fingerprint spectrum of the fragrance data of different fruits is extracted using a gas chromatography-ion mobility spectrometry algorithm based on multiple preset volatile components to obtain a target sample fingerprint spectrum. This includes: extracting fingerprint spectra of the fragrance data of different fruits based on multiple preset volatile components using a gas chromatography-ion mobility spectrometry algorithm to obtain an initial sample fingerprint spectrum, which includes signal peak data of the fragrance data of different fruits at multiple preset volatile components; sorting the positions of the multiple preset volatile components in the spectrum according to the signal peak data to obtain signal peak aggregation difference distribution data; and dividing the initial sample fingerprint spectrum into regions based on the signal peak aggregation difference distribution data to obtain the target sample fingerprint spectrum.

[0097] The initial sample fingerprint spectrum is a fingerprint spectrum processed by gas chromatography ion mobility spectrometry algorithm and with the addition of preset volatile components. The initial sample fingerprint spectrum includes the signal peak data of multiple preset volatile components for the sample fragrance data of different fruits.

[0098] Since the signal peak intensities of multiple preset volatile components differ among the sample spice data of each fruit type, in order to conduct a more intuitive analysis, the positions of multiple preset volatile components in the spectrum are sorted according to the color development degree of the signal peak intensity in the spectrum. The positions of preset volatile components with similar color development degrees are grouped together as much as possible. Then, the positions of the multiple preset volatile components that have been grouped together are sorted in descending order according to the magnitude of the signal peak intensity to obtain the signal peak aggregation difference distribution data.

[0099] The fingerprint spectrum of the initial sample is divided into regions based on the differential distribution data of signal peak aggregation to obtain the fingerprint spectrum of the target sample.

[0100] like Figure 6 As shown, the fingerprint spectrum of the target sample is divided into 17 regions. Region A is the common component region of three fruits: wolfberry extract, fig extract and grape extract. It contains pre-defined volatile components such as dimethyl disulfide, 2-heptanone, pyridine and 2-ethylpyridine. 2-heptanone has a banana aroma and a slight medicinal aroma. Pyridine has a special pungent odor and a spicy taste.

[0101] Regions B and H are the component regions of sea buckthorn extract fruit, containing compounds such as 2-ethylpyridine, isoamyl formate (monomer), 2-butanone (monomer and dimer), methyl isovalerate, methyl 2-methylbutyrate, and acetone. Among them, isoamyl formate has a sweet, fruity aroma reminiscent of purple grapes and plums, with a wine-like fruity flavor; 2-butanone has a distinctive pungent odor, similar to acetone; methyl isovalerate has a strong, phenolic ether-like, apple-like aroma and flavor; and methyl 2-methylbutyrate can modify the fruity and minty aromas in tobacco varieties.

[0102] Region C represents the common component region of apricot and plum extracts, including isoamyl formate (monomer and dimer), 2-furfural (monomer and dimer), ethyl butyrate (monomer), ethyl isovalerate (monomer), and ethyl hexanoate (monomer and dimer). Furfural possesses a sweet, cereal, bread-like, fermented, caramel, and spicy aroma; ethyl butyrate has an aroma similar to banana and pineapple, with a sweet fruity taste; ethyl isovalerate has a strong fruity and wine-like aroma; and ethyl hexanoate has a strong, diffused aroma of apple, pineapple, and wine.

[0103] Region D is the component region of fig extract, containing ethyl hexanoate (monomer and dimer).

[0104] Region E contains 2-ethyl-4-methylthiazole (monomer and dimer), ethyl propionate, 2-methylbutyraldehyde, and 2-hexanol, and is found in high concentrations in plum extract and jujube tincture. 2-ethyl-4-methylthiazole is one of the volatile aroma components of coffee, with aromas of coffee, roasted hazelnuts, and other nuts; ethyl propionate has a strong fruity aroma of banana and pineapple.

[0105] Region F is the component region of hawthorn extract, including 2-hexanol, 3-penten-2-one, and 1-propanol (dimer). Among them, 3-penten-2-one has a pungent, spicy aroma, but it has a fruity taste when diluted.

[0106] Region G is a region of unique components of fenugreek extract, including 2-butanone (monomer and dimer) and methyl isovalerate.

[0107] S16: Compounds in region I are present in high concentrations in plum extract, including acetone, ethyl valerate (monomer and dimer), ethyl isovalerate (dimer), 2-hexanone (monomer and dimer), 2-butoxyethanol, methyl valerate, 2-isopropyl-3-methoxypyrazine, 2-acetyl-1-pyrrolline, diisobutyl ketone, (E)-2-hexenal, and heptanal. Methyl valerate has an aroma and flavor similar to apple and pineapple; 2-isopropyl-3-methoxypyrazine has a fresh pea, potato, and green pepper-like vegetable flavor; diisobutyl ketone has a sweet, fruity aroma; (E)-2-hexenal has a strong, leafy, fruity, and pungent vegetable-like aroma; and heptanal has a strong, rough, pungent, oily odor.

[0108] The compounds in region J are all unknown compounds, representing a common component region for wolfberry extract and fig extract.

[0109] Region K includes isopropanol (monomer and dimer) and tert-butanol, of which isopropanol has a slightly alcoholic odor and a pungent taste.

[0110] Region L contains unique components of wolfberry extract, including isobutanol (monomer and dimer), methyl hexanoate, ethyl 2-methylbutyrate, isobutyl acetate, and 1-pentanol (monomer and dimer). Isobutanol has a characteristic pungent odor of alcohol; methyl hexanoate has a strong aroma of pineapple, apricot, and banana; ethyl 2-methylbutyrate has a strong, fresh, apple-strawberry aroma and flavor; isobutyl acetate has a fruity aroma of redcurrant and pear, a floral aroma of hyacinth and rose, and a distinctive slightly bitter, ethereal flavor; and 1-pentanol has an alcoholic, fusel oil-like odor and a strong flavor.

[0111] Region M includes 1-pentanol (monomer and dimer), cumene, 3-octanone, and butyl butyrate, which are unique components of jujube tincture. Among them, 3-octanone has a strong, pungent fruity aroma with a hint of lavender; butyl butyrate has a pear-pineapple-like fruity aroma.

[0112] Region N includes butyl butyrate, 1-butanol (monomer and dimer), ethyl butyrate (dimer), isobutyl isovalerate (monomer and dimer), ethyl octanoate (monomer and dimer), hexyl isobutyrate (monomer and dimer), ethyl isobutyrate (monomer and dimer), butyl acetate (monomer and dimer), γ-terpinene, and isoamyl butyrate. 1-Butanol has a particularly pungent, wine-like, banana-like aroma with a dry, spicy taste; isobutyl isovalerate has an aroma similar to apples and strawberries; ethyl octanoate has a pleasant floral and fruity aroma with a fruity wine-like scent and a sweet taste similar to apricots and pineapples; hexyl isobutyrate has a strong, slightly rough fruity aroma; ethyl isobutyrate has a sweet, phenolic, ethereal, fruity aroma and taste, as well as an apple-like aftertaste; butyl acetate has a strong fruity aroma with a sweet-then-spicy, pineapple-like taste; γ-terpinene has a lemony aroma; isoamyl butyrate has a strong fruity aroma and sweet taste similar to pears, apricots, bananas, and pineapples.

[0113] Region O contains unique components of apple extract, including triethylamine, hexyl formate, isoamyl acetate, and 1-propanol (monomer). Hexyl formate has a sweet fruity aroma, resembling apple or unripe plum; isoamyl acetate has fruity aromas such as banana, pear, and apple.

[0114] Region P contains components unique to grape extracts, including 1-propanol (monomer), (E)-2-heptenal, 1-octen-3-one, and 2,4-dimethylthiazole. (E)-2-heptenal has a slightly oily, pungent, green aroma. Region Q contains components unique to jujube tincture and hawthorn extracts, including 2,4-dimethylthiazole, 2-butylfuran, and hexanal. Hexanal has a characteristic fruity aroma and flavor.

[0115] The sample fragrance data of 14 kinds of fruits each have different component regions, which can be well distinguished, thus providing strong support for detecting and analyzing the differences in volatile components to identify the types of fragrance data of the fruits to be tested.

[0116] According to an embodiment of this application, the importance of projected variables is determined by statistical analysis of sample spice data of different fruits on preset volatile components. This includes: using principal component analysis algorithm to process the signal peak data of different fruits on multiple preset volatile components in the fingerprint spectrum of the target sample to obtain analysis results; and when the analysis results indicate that there are differences between different fruits, using orthogonal partial least squares discriminant algorithm to process the signal peak data of different fruits on multiple preset volatile components to obtain the importance of projected variables corresponding to multiple preset volatile components.

[0117] Principal Component Analysis (PCA) was used to process the signal peak data corresponding to multiple preset volatile components in the fingerprint spectrum of the target sample for different fruit types. Each fruit type was treated as a point, and the distance between points was calculated to obtain the analysis results. The analysis results represent the degree of difference between fruit types.

[0118] Figure 7 A schematic diagram showing the analysis results according to an embodiment of this application is illustrated.

[0119] like Figure 7 As shown, a three-dimensional visualization analysis of the score map in the principal component analysis algorithm results reveals aggregation in some fruit types and significant differences among most fruit types. For example, the spatial positions of grape extract and apricot extract are far apart, indicating a large difference. PC represents the relative intensity percentage. The sum of the peak areas of all preset volatile components in the fingerprint spectrum of the target sample is 100%, and the percentage of the peak area of ​​a single preset volatile component relative to the total area is the PC value of that peak. For example, PC1 of 31.657% indicates that the peak area of ​​preset volatile component 1 accounts for 31.657% of the total peak area of ​​the entire spectrum; PC2 of 16.820% indicates that the peak area of ​​preset volatile component 2 accounts for 16.820% of the total peak area of ​​the entire spectrum; and PC3 of 15.408% indicates that the peak area of ​​preset volatile component 3 accounts for 15.408% of the total peak area of ​​the entire spectrum.

[0120] When there are differences between different types of fruits, the target category discrimination model constructed based on sample spice data of different fruits and the accurate determination of the importance of the projected variables are more reliable.

[0121] Figure 8 A schematic diagram showing the ordering of the importance of projection variables according to an embodiment of this application is illustrated.

[0122] like Figure 8As shown, the importance of the projected variable for each preset volatile component is calculated based on the signal peak volume corresponding to each of the different fruit types for multiple preset volatile components. The preset volatile components are then sorted in descending order according to the importance of the projected variable. The horizontal axis represents the preset volatile components, and the vertical axis represents the importance of the projected variable. The threshold for the importance of the projected variable can be 1. Volatile component markers with a projected variable importance > 1 are then... Figure 8 The following are marked in red: 38 types in total.

[0123] The 38 volatile component markers are isopropanol D, ethyl acetate M, 2-butylfuran, tert-butanol, isopropanol M, isoamyl acetate, triethylamine, 6, methyl isovalerate, hexyl formate, 1-propanol M, (E)-2-heptenal, 10, isovaleral D, 18, 1-octen-3-one, 16, butyraldehyde, 24, 27, 2-butanone M, 2-butanone D, 9, ethyl hexanoate M, isovaleral M, 1-propanol D, ethyl hexanoate D, furfural M, 2,4-dimethylthiazole, 8, 2-ethylpyridine, methyl 2-methylbutyrate, 11, 1-pentanol M, isobutanol D, 1-pentanol D, isobutanol M, and cumene.

[0124] According to an embodiment of this application, volatile component markers are screened from multiple preset volatile components based on a preset projection variable importance threshold and the respective projection variable importance of multiple preset volatile components, including: determining preset volatile components with a projection variable importance greater than the preset projection variable importance threshold as volatile component markers.

[0125] If the importance of a projected variable is greater than the preset importance threshold, it means that the preset volatile component corresponding to this projected variable has a significant impact on the target category discrimination model. Therefore, it is necessary to select the preset volatile components whose projected variable importance is greater than the preset importance threshold, so as to input the signal peak data of the volatile component markers in the fruit flavor data to be tested into the target category discrimination model, thereby improving the discrimination accuracy of the target category discrimination model.

[0126] Based on the above-mentioned method for detecting fruit flavorings based on gas chromatography ion mobility spectrometry algorithm, this application also provides a device for detecting fruit flavorings. The following will be combined with... Figure 9 The device is described in detail.

[0127] Figure 9 A structural block diagram of a fruit flavoring detection device according to an embodiment of this application is shown.

[0128] like Figure 9 As shown, the fruit spice detection device 900 of this embodiment includes an extraction module 910, a detection module 920, and an identification module 930.

[0129] The extraction module 910 is used to display the database parameter set corresponding to the target version in the visual configuration interface. The database parameter set includes multiple parameter entries, and each parameter entry includes a parameter name, parameter value, and effective scope.

[0130] The detection module 920 is used to detect the fingerprint spectrum using a detection model to obtain multiple target volatile components, wherein the target volatile components are the volatile components contained in the fruit fragrance data to be tested.

[0131] The identification module 930 is used to analyze multiple target volatile components, preset signal peak difference data, and volatile component markers to obtain the fruit category corresponding to the fruit flavor data to be tested. Among them, the preset signal peak difference data represents the difference in sample signal peak data of different fruit categories with respect to preset volatile components. The volatile component markers are selected from multiple preset volatile components based on preset projection variable importance thresholds and the importance of projection variables of each preset volatile component. The importance of projection variables is determined based on statistical analysis of the sample flavor data of different fruit categories with respect to preset volatile components.

[0132] According to an embodiment of this application, the identification module 930 includes a first identification submodule and a second identification submodule.

[0133] The first identification submodule is used to extract matching volatile components from multiple target volatile components that are matched with multiple volatile component markers.

[0134] The second identification submodule is used to process the signal peak data of matching volatile components using the target category discrimination model to obtain the fruit category corresponding to the fruit spice data to be tested. The target category discrimination model is built on the basis of difference analysis of preset signal peak difference data.

[0135] According to an embodiment of this application, the identification module 930 further includes a first extraction submodule, a second extraction submodule, a third extraction submodule, and a fourth extraction submodule.

[0136] The first extraction submodule is used to extract two-dimensional spectra from the spice data of different fruits based on the gas chromatography ion mobility spectrometry algorithm, and obtain sample two-dimensional spectra, wherein the sample two-dimensional spectra represent the signal peak difference distribution between the spice data of different fruits.

[0137] The second extraction submodule is used to mark the signal peaks in the fingerprint spectrum of the target sample to obtain multiple bounding boxes, wherein the bounding boxes represent the positions of signal peaks that are higher than the preset signal peak threshold.

[0138] The third extraction submodule is used to perform qualitative analysis on multiple bounding boxes to obtain multiple preset volatile components.

[0139] The fourth extraction submodule is used to extract differential spectra from the spice data of different fruits based on multiple preset volatile components using a gas chromatography ion mobility spectrometry algorithm, and obtain sample differential spectra. The sample differential spectra represent the signal peak difference distribution data of different fruits for the spice data of the sample in the preset volatile components. The preset signal peak difference data includes the signal peak difference distribution data.

[0140] According to an embodiment of this application, the identification module 930 further includes a fifth extraction submodule.

[0141] The fifth extraction submodule is used to extract fingerprint spectrum of spice data of different fruits based on multiple preset volatile components using gas chromatography ion mobility spectrometry algorithm to obtain the fingerprint spectrum of the target sample. The fingerprint spectrum of the target sample represents the regional aggregation difference distribution data of spice data of different fruits in multiple preset volatile components. The preset signal peak difference data also includes regional aggregation difference distribution data.

[0142] According to an embodiment of this application, the fifth extraction submodule includes a first extraction unit, a second extraction unit, and a third unit.

[0143] The first extraction unit is used to extract fingerprint spectra from the spice data of different fruits based on multiple preset volatile components using a gas chromatography ion mobility spectrometry algorithm, to obtain an initial sample fingerprint spectrum. The initial sample fingerprint spectrum includes the signal peak data of the spice data of different fruits at multiple preset volatile components.

[0144] The second extraction unit is used to sort the positions of multiple preset volatile components in the spectrum according to the signal peak data, and obtain the signal peak aggregation difference distribution data.

[0145] The third extraction unit is used to divide the initial sample fingerprint spectrum into regions based on the signal peak aggregation difference distribution data to obtain the target sample fingerprint spectrum.

[0146] According to an embodiment of this application, the identification module 930 further includes a principal component analysis submodule and a discrimination submodule.

[0147] The principal component analysis submodule is used to process the signal peak data of different fruits in the fingerprint spectrum of the target sample at each of the multiple preset volatile components using the principal component analysis algorithm, and obtain the analysis results.

[0148] The discrimination submodule is used to process the signal peak data corresponding to each of the different fruit types based on the orthogonal partial least squares discrimination algorithm when the analysis results indicate that there are differences between different fruit types, and to obtain the importance of the projection variables corresponding to each of the multiple preset volatile components.

[0149] According to an embodiment of this application, the identification module 930 further includes a determination submodule.

[0150] The determination submodule is used to identify preset volatile components whose importance of the projected variable is greater than the preset importance threshold of the projected variable as volatile component markers.

[0151] According to embodiments of this application, any plurality of modules among the modules, submodules, and units can be combined into one module, or any one of the modules can be split into multiple modules. Alternatively, at least a portion of the functionality of one or more of these modules can be combined with at least a portion of the functionality of other modules and implemented in one module. Optionally, at least one of the modules, submodules, and units can be at least partially implemented as hardware circuitry, such as a field-programmable gate array (FPGA), a programmable logic array (PLA), a system-on-a-chip, a system-on-a-substrate, a system-on-package, an application-specific integrated circuit (ASIC), or implemented in hardware or firmware by any other reasonable means of integrating or packaging circuitry, or implemented in any one of software, hardware, and firmware methods, or in a suitable combination of any of these. Alternatively, at least one of the modules, submodules, and units can be at least partially implemented as a computer program module, which can perform corresponding functions when the computer program module is run.

[0152] Figure 10 A block diagram of an electronic device suitable for implementing a method for detecting fruit spices based on a gas chromatography ion mobility spectrometry algorithm, according to an embodiment of this application, is shown.

[0153] like Figure 10 As shown, an electronic device 1000 according to an embodiment of this application includes a processor 1001, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage portion 1008 into a random access memory (RAM) 1003. The processor 1001 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or an associated chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 1001 may also include onboard memory for caching purposes. The processor 1001 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of this application.

[0154] RAM 1003 stores various programs and data required for the operation of electronic device 1000. Processor 1001, ROM 1002, and RAM 1003 are interconnected via bus 1004. Processor 1001 executes various operations of the method flow according to embodiments of this application by executing programs in ROM 1002 and / or RAM 1003. It should be noted that programs may also be stored in one or more memories other than ROM 1002 and RAM 1003. Processor 1001 may also execute various operations of the method flow according to embodiments of this application by executing programs stored in one or more memories.

[0155] According to embodiments of this application, the electronic device 1000 may further include an input / output (I / O) interface 1005, which is also connected to a bus 1004. The electronic device 1000 may also include one or more of the following components connected to the input / output (I / O) interface 1005: an input section 1006 including a keyboard, mouse, etc.; an output section 1007 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 1008 including a hard disk, etc.; and a communication section 1009 including a network interface card such as a LAN card, modem, etc. The communication section 1009 performs communication processing via a network such as the Internet. A drive 1010 is also connected to the input / output (I / O) interface 1005 as needed. A removable medium 1011, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 1010 as needed so that computer programs read from it can be installed into the storage section 1008 as needed.

[0156] This application also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or it may exist independently and not assembled into the device / apparatus / system. The computer-readable storage medium carries one or more programs, which, when executed, implement the fruit flavoring detection method based on gas chromatography ion mobility spectrometry algorithm according to the embodiments of this application.

[0157] According to embodiments of this application, the computer-readable storage medium can be a non-volatile computer-readable storage medium, such as including but not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, the computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, according to embodiments of this application, the computer-readable storage medium may include ROM 1002 and / or RAM 1003 and / or one or more memories other than ROM 1002 and RAM 1003 described above.

[0158] Embodiments of this application also include a computer program product comprising a computer program containing program code for performing the methods shown in the flowchart. When the computer program product is run on a computer system, the program code enables the computer system to implement the fruit flavoring detection method based on gas chromatography ion mobility spectrometry algorithm provided in the embodiments of this application.

[0159] When the computer program is executed by the processor 1001, it performs the functions defined in the system / apparatus of this application embodiment. According to the embodiments of this application, the systems, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0160] In one embodiment, the computer program may rely on a tangible storage medium such as an optical storage device or a magnetic storage device. In another embodiment, the computer program may also be transmitted and distributed in the form of signals over a network medium, and may be downloaded and installed via the communication section 1009, and / or installed from a removable medium 1011. The program code contained in the computer program can be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination thereof.

[0161] In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 1009, and / or installed from the removable medium 1011. When the computer program is executed by the processor 1001, it performs the functions defined in the system of this application embodiment. According to the embodiments of this application, the systems, devices, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0162] According to embodiments of this application, program code for executing the computer programs provided in the embodiments of this application can be written in any combination of one or more programming languages. Specifically, these computational programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, languages ​​such as Java, C++, Python, "C", or similar programming languages. The program code can be executed entirely on the user's computing device, partially on the user's device, partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0163] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0164] Those skilled in the art will understand that the features described in the various embodiments of this application can be combined and / or combined in various ways, even if such combinations or combinations are not explicitly described in this application. In particular, the features described in the various embodiments of this application can be combined and / or combined in various ways without departing from the spirit and teachings of this application. All such combinations and / or combinations fall within the scope of this application.

Claims

1. A method for detecting fruit flavorings based on gas chromatography ion mobility spectrometry algorithm, characterized in that, The method includes: Fingerprint spectrum extraction is performed on the test fruit fragrance data based on gas chromatography ion mobility spectrometry algorithm to obtain fingerprint spectrum, wherein the fingerprint spectrum includes signal peak data corresponding to each of multiple preset volatile components; The fingerprint spectrum is detected using a detection model to obtain multiple target volatile components, wherein the target volatile components are the volatile components contained in the fruit fragrance data to be tested; The analysis of multiple target volatile components, preset signal peak difference data, and volatile component markers yields the fruit categories corresponding to the tested fruit flavor data. The preset signal peak difference data characterizes the differences in sample signal peak data for different fruit categories with respect to preset volatile components. The volatile component markers are selected from multiple preset volatile components based on a preset projection variable importance threshold and the importance of the projection variables for each preset volatile component. The importance of the projection variables is determined based on statistical analysis of the sample flavor data for different fruit categories with respect to the preset volatile components.

2. The method according to claim 1, characterized in that, The analysis of multiple target volatile components, preset signal peak difference data, and volatile component markers yields the fruits corresponding to the tested fruit flavor data, including: Extract matching volatile components that correspond to the multiple volatile component markers from the plurality of target volatile components; The target category discrimination model is used to process the signal peak data of matching volatile components to obtain the fruit category corresponding to the fruit spice data to be tested. The target category discrimination model is constructed based on the difference analysis of preset signal peak difference data.

3. The method according to claim 1, characterized in that, Determine the preset signal peak difference data, including: Two-dimensional spectra were extracted from the spice data of different fruits based on the gas chromatography ion mobility spectrometry algorithm to obtain sample two-dimensional spectra, wherein the sample two-dimensional spectra represent the signal peak difference distribution between the spice data of different fruits. Signal peaks are marked on the fingerprint spectrum of the target sample to obtain multiple bounding boxes, wherein the bounding boxes represent the positions of signal peaks that are higher than a preset signal peak threshold; Qualitative analysis was performed on multiple of the marked boxes to obtain multiple preset volatile components; The gas chromatography ion mobility spectrometry algorithm is used to extract differential spectra from the spice data of different fruits based on multiple preset volatile components, and the resulting sample differential spectra are obtained. The sample differential spectra represent the signal peak difference distribution data of different fruits for the spice data of the sample volatile components, and the preset signal peak difference data includes signal peak difference distribution data.

4. The method according to claim 3, characterized in that, The target sample fingerprint is determined based on the following operations: The fingerprint spectrum of the fragrance data of different fruits is extracted using a gas chromatography ion mobility spectrometry algorithm based on multiple preset volatile components to obtain the fingerprint spectrum of the target sample. The fingerprint spectrum of the target sample represents the regional aggregation difference distribution data of the fragrance data of different fruits in multiple preset volatile components. The preset signal peak difference data also includes regional aggregation difference distribution data.

5. The method according to claim 4, characterized in that, The method utilizes gas chromatography-ion mobility spectrometry (GC-IMS) to extract fingerprint spectra from the fragrance data of different fruit samples based on multiple preset volatile components, obtaining the fingerprint spectra of the target samples, including: The fingerprint spectrum of the fragrance data of different fruits was extracted using the gas chromatography ion mobility spectrometry algorithm based on multiple preset volatile components to obtain the initial sample fingerprint spectrum. The initial sample fingerprint spectrum includes the signal peak data of the fragrance data of different fruits at multiple preset volatile components. Based on the signal peak data, the positions of multiple preset volatile components in the spectrum are sorted to obtain signal peak aggregation difference distribution data; The fingerprint spectrum of the initial sample is divided into regions based on the differential distribution data of signal peak aggregation to obtain the fingerprint spectrum of the target sample.

6. The method according to claim 4, characterized in that, The importance of projected variables was determined by statistical analysis of the preset volatile components based on sample spice data from different fruits, including: The principal component analysis algorithm was used to process the signal peak data of different fruits in the fingerprint spectrum of the target sample for each of the multiple preset volatile components, and the analysis results were obtained. When the analysis results indicate differences between different fruit types, the signal peak data corresponding to each of the different fruit types in multiple preset volatile components are processed based on the orthogonal partial least squares discriminant algorithm to obtain the importance of the projection variables corresponding to each of the multiple preset volatile components.

7. The method according to claim 1, characterized in that, Based on a preset importance threshold for projected variables and the importance of the projected variables for each of the multiple preset volatile components, the volatile component markers are screened from multiple preset volatile components, including: The volatile components whose importance of the projected variable is greater than a preset importance threshold are identified as the volatile component markers.

8. The method according to any one of claims 1-7, characterized in that, On the instrument where the gas chromatography ion mobility spectrometry algorithm is deployed, the migration gas flow rate is 75 mL / min, the carrier gas flow rate is initially 2 mL / min and maintained for 20 min, then linearly increased to 100 mL / min within 20–35 min and maintained for 5 min.

9. A fruit flavoring detection device, characterized in that, The device includes: The extraction module is used to extract fingerprint spectrum from the fruit fragrance data to be tested based on the gas chromatography ion mobility spectrometry algorithm to obtain fingerprint spectrum, wherein the fingerprint spectrum includes signal peak data corresponding to each of multiple preset volatile components; The detection module is used to detect the fingerprint spectrum using a detection model to obtain multiple target volatile components, wherein the target volatile components are the volatile components contained in the fruit fragrance data to be tested; The identification module is used to analyze multiple target volatile components, preset signal peak difference data, and volatile component markers to obtain the fruit category corresponding to the fruit flavoring data to be tested. The preset signal peak difference data characterizes the differences in sample signal peak data of different fruit categories for preset volatile components. The volatile component markers are selected from multiple preset volatile components based on a preset projection variable importance threshold and the importance of the projection variables of each preset volatile component. The importance of the projection variables is determined based on statistical analysis of the sample flavoring data of different fruit categories for the preset volatile components.

10. An electronic device, comprising: One or more processors; Memory, used to store one or more computer programs. The characteristic feature is that the one or more processors invoke the one or more computer programs to implement the steps of the method according to any one of claims 1 to 8.

Images