Dendrobium officinale extraction purity detection method based on artificial intelligence

Abstract

The invention discloses a dendrobium officinale extraction purity detection method based on artificial intelligence, and relates to the technical field of purity detection, and the method comprises the following steps: carrying out hyperspectral data acquisition and component classification on a dendrobium officinale extract to obtain a classification result; performing content prediction according to the classification result to obtain a content prediction result; performing chromatographic analysis verification according to the content prediction result to obtain a verification result; performing purity prediction according to the verification result to obtain a purity prediction result; performing component spatial distribution identification according to the purity prediction result to obtain component distribution information; and performing purity calculation according to the component distribution information. According to the method, the content prediction precision is improved, the resolution capability on different components is improved, the influence of abnormal samples on the content prediction result is reduced, the dependence on chromatographic analysis is reduced, the time and the cost are saved, and the accuracy and the robustness of purity detection are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of purity detection, and particularly to a method for detecting the extraction purity of Dendrobium officinale based on artificial intelligence. Background Art

[0002] Dendrobium officinale has been widely studied and applied due to its rich medicinal value, especially in the extraction and purity control of active ingredients such as polysaccharides, flavonoids, and alkaloids, which has important medical and health significance. Current purity detection methods mainly rely on chromatographic analysis (such as high-performance liquid chromatography HPLC, gas chromatography GC) and traditional spectroscopic techniques (such as ultraviolet-visible spectroscopy UV-Vis, infrared spectroscopy IR).

[0003] In the prior art, there are deficiencies in predicting the content of extract components: In existing methods, the content prediction of Dendrobium officinale extract usually relies on a single technical means, and the prediction accuracy is limited. For example, HPLC or spectroscopic analysis is used alone, but these methods all have limitations, including complex sample pretreatment, long analysis time, and difficulty in accurately predicting the content of target components.

[0004] In the prior art, there are deficiencies in detecting the purity of extract: Current purity determination mainly relies on chromatographic methods such as HPLC and GC. Although these methods can provide high-precision data, the experimental operation is complex, the instrument cost is high, and they cannot meet the needs of large-scale and rapid detection. Moreover, due to the problem of spectral signal overlap between target components and impurities, it is difficult for spectroscopic analysis to independently determine the purity of the extract. Therefore, in the prior art, spectroscopic methods are often used to assist chromatographic analysis rather than independently complete purity determination. Summary of the Invention

[0005] Aiming at the deficiencies of the prior art, the present invention provides a method for detecting the extraction purity of Dendrobium officinale based on artificial intelligence to solve the problems raised in the above background art.

[0006] To achieve the above object, the present invention provides the following technical solutions: In the first aspect, an embodiment of the present invention provides a method for detecting the extraction purity of Dendrobium officinale based on artificial intelligence, including the following steps: S1. Collecting hyperspectral data of Dendrobium officinale extract and classifying components to obtain a classification result; S2. Predicting the content according to the classification result to obtain a content prediction result; S3. Verifying by chromatographic analysis according to the content prediction result to obtain a verification result; S4. Predicting the purity according to the verification result to obtain a purity prediction result; S5. Identifying the spatial distribution of components according to the purity prediction result to obtain component distribution information; S6. Calculate the purity based on the component distribution information.

[0007] To further optimize this technical solution, the content prediction in S2 includes: Based on the classification result, use the content prediction model to calculate the content of each component in the Dendrobium officinale extract.

[0008] To further optimize this technical solution, the content prediction model includes: ; Where: : The content of the k-th predicted component; : The spectral value of the i-th sample in the j-th spectral band; : Bias term; : The weight of each sample; : Model weight; : The total number of samples; : The total number of spectral bands.

[0009] To further optimize this technical solution, the purity prediction in S4 includes: Based on the verification result and hyperspectral data, use the purity prediction model to perform purity prediction.

[0010] To further optimize this technical solution, the purity prediction model includes: ; Where: : Predicted purity value; : The reflection value of the i-th sample in the spectral band corresponding to the k-th component; : Bias term; : Contribution coefficient of the k-th component; : Nonlinear influence coefficient of the k-th component; : Sample importance coefficient; : The number of component types considered in the model.

[0011] To further optimize this technical solution, the contribution coefficient includes: ; Wherein: : The total content of all components; : The normalization coefficient of component k.

[0012] To further optimize this technical solution, the non-linear influence coefficient includes: ; Wherein: : The standard deviation of the spectral signal of component k; : The spectral feature complexity factor of component k.

[0013] To further optimize this technical solution, the spectral feature complexity factor includes: ; Wherein: , : The spectral values of component k at wavelengths m and m - 1.

[0014] To further optimize this technical solution, the sample importance coefficient includes: ; Wherein: : The spectral credibility of sample i.

[0015] To further optimize this technical solution, the spectral credibility includes: ; : The spectral value of sample i at wavelength m; : The average spectral value of all samples at wavelength m.

[0016] In a second aspect, an embodiment of the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and wherein: when the computer program instructions are executed by the processor, the steps of a method for detecting the extraction purity of Dendrobium officinale based on artificial intelligence as described in the first aspect of the present invention are implemented.

[0017] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, and wherein: when the computer program instructions are executed by the processor, the steps of a method for detecting the extraction purity of Dendrobium officinale based on artificial intelligence as described in the first aspect of the present invention are implemented.

[0018] Compared with the prior art, the present invention provides a method for detecting the extraction purity of Dendrobium officinale based on artificial intelligence, which has the following beneficial effects: The method for detecting the extraction purity of Dendrobium officinale based on artificial intelligence predicts the content through a content prediction model, combines spectral signals, chromatographic data, and sample characteristic information of different wavelengths, and performs content prediction through an adaptive sample weight calculation method, improving the content prediction accuracy, enhancing the resolution ability for different components, and reducing the influence of abnormal samples on the content prediction results.

[0019] Through the purity prediction model, the introduction of spectral feature weights, adaptive component contribution calculation, and spectral signal deviation correction reduces the dependence on chromatographic analysis, saves time and cost, and improves the accuracy and robustness of purity detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0021] Figure 1 It is a schematic flow chart of a method for detecting the extraction purity of Dendrobium officinale based on artificial intelligence proposed by the present invention; Figure 2 It is a schematic flow chart of a content prediction model of a method for detecting the extraction purity of Dendrobium officinale based on artificial intelligence proposed by the present invention; Figure 3 It is a schematic flow chart of a purity prediction model of a method for detecting the extraction purity of Dendrobium officinale based on artificial intelligence proposed by the present invention; Figure 4 It is a schematic flow chart of a component space distribution recognition model of a method for detecting the extraction purity of Dendrobium officinale based on artificial intelligence proposed by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0022] In order to make the above objects, features, and advantages of the present invention more obvious and understandable, the following will provide a detailed description of the specific embodiments of the present invention with reference to the accompanying drawings of the specification.

[0023] Many specific details are set forth in the following description in order to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar generalizations without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0024] Second, the "one embodiment" or "embodiment" mentioned herein refers to specific features, structures, or characteristics that may be included in at least one implementation of the present invention. The "in one embodiment" that appears in different places in this specification does not all refer to the same embodiment, nor is it an embodiment that is separate or selectively exclusive of other embodiments.

[0025] Embodiment 1: Referring to Figures 1 to 4 , which is the first embodiment of the present invention. This embodiment provides a method for detecting the extraction purity of Dendrobium officinale based on artificial intelligence, including the following steps: S1. Collect hyperspectral data of Dendrobium officinale extract and classify its components to obtain a classification result.

[0026] In this embodiment, the hyperspectral data collection and component classification include: Using hyperspectral imaging technology HSI, select the spectral range of 400nm - 1000nm suitable for biological samples, perform spectral scanning on the samples, use a short-wave infrared hyperspectral camera to obtain the spectral images of the samples under a controlled light source, ensure the uniformity of illumination, adopt standard normal variate transformation to eliminate background noise, remove the influence of illumination, and improve the stability of spectral data, so as to obtain the spectral information of Dendrobium officinale extract in different bands, obtain the hyperspectral data of Dendrobium officinale extract, and establish a spectral feature library of Dendrobium officinale extract according to the physical information (including color, particle uniformity) and chemical information (content of the main components of Dendrobium officinale extract) in multiple bands to record the spectral features in different bands. Use the convolutional neural network CNN (prior art) to process the hyperspectral data of Dendrobium officinale extract, extract the spectral features of the image, and classify and identify the components of Dendrobium officinale extract. The key modules of the convolutional neural network CNN include a convolutional layer (used for extracting hyperspectral features to obtain a feature map), a pooling layer (used for reducing the size of the feature map and retaining the main features), and a flattening and fully connected layer (used for flattening the feature vector for classification and identification of the components of Dendrobium officinale extract).

[0027] S2. Perform content prediction according to the classification result to obtain a content prediction result.

[0028] In this embodiment, the content prediction includes: According to the classification result, use a content prediction model to calculate the content of each component in the Dendrobium officinale extract, so as to reduce the experimental time consumption and improve the detection efficiency.

[0029] Furthermore, the content prediction model includes: ; Wherein: : The content of the k-th predicted component, including polysaccharides, flavonoids, etc.; : The spectral value of the i-th sample in the j-th spectral band; : The bias term, adjusted based on the constant value after model training to ensure the accuracy of the prediction result; : The weight of each sample, representing the importance of different samples in training, adjusted according to the output of the model; : The model weight, indicating the influence of the i-th sample in the j-th band; : The total number of samples; : The total number of spectral bands.

[0030] This model describes how to predict the component content of Dendrobium officinale extracts through spectral data.

[0031] Traditional prediction of component content often relies on classical regression models such as principal component regression and partial least squares regression, or simple linear regression methods. It lacks the ability to adaptively adjust according to the sample situation, with insufficient accuracy and adaptability. It requires combining chromatographic data, increasing time and cost consumption. This model can adjust according to the actual situation of each sample, capture complex non-linear relationships through deep learning, discover the interactions between different bands, improve the prediction accuracy of spectral data, and does not require additional chemical reactions or complex sample processing steps, avoiding the time and cost consumption in chromatographic analysis.

[0032] The usage steps of this model include: Data acquisition: Obtain the spectral values of the samples through the hyperspectral data of Dendrobium officinale extracts obtained in step S1 ; Model training: Obtain spectral features through step S1, train the model, establish the non-linear relationship between samples and spectral data, and obtain the weight parameters , ; Content prediction: Input the obtained data combined with the trained weight parameters into the model for calculation to obtain the prediction results of the content of each component in Dendrobium officinale extracts , and the prediction results are verified by comparing with the verification results in step S4 to ensure the accuracy of the prediction results.

[0033] S3. Conduct chromatographic analysis verification according to the content prediction results to obtain the verification results.

[0034] In this embodiment, the chromatographic analysis verification includes: According to the content prediction results, the extract is detected using high performance liquid chromatography (HPLC), the content of the target component is calculated using the external standard method, and finally the error between the HPLC detection value and the predicted value is calculated to verify the content prediction results. If the error ≤ 5%, it is determined that the model prediction is accurate, thus ensuring that the component content predicted by the model conforms to the actual detection data.

[0035] S4. Perform purity prediction based on the verification results to obtain the purity prediction results.

[0036] In this embodiment, the purity prediction includes: Based on the verification results and the hyperspectral data, using the purity prediction model, by establishing the mapping relationship between the hyperspectral data and the component purity of the Dendrobium officinale extract, perform purity prediction and analyze the component purity of the Dendrobium officinale extract. This step combines the hyperspectral data with the results after the component content prediction verification, and performs non-linear mapping between the multi-dimensional spectral data and the content of each component, thereby directly predicting the purity of the Dendrobium officinale extract, achieving rapid and high-throughput prediction of purity values, and being used for real-time and rapid judgment of whether the extraction batch meets the standards in the industrial field.

[0037] Further, the purity prediction model includes: ; Where: : Predicted purity value, that is, the predicted purity value of the Dendrobium officinale extract; : The reflection value of the i-th sample in the spectral band corresponding to the k-th component; : Bias term, as an adjustment term to ensure the accuracy of the model prediction; : Contribution coefficient of the k-th component; : Non-linear influence coefficient of the k-th component, reflecting the influence of the component spectral characteristics on the purity calculation; : Sample importance coefficient, used to suppress the influence of abnormal spectral data on the purity calculation; : The number of types of components considered in the model, that is, the number of types of main components such as polysaccharides and flavonoids.

[0038] This model describes how to perform purity prediction based on spectral data.

[0039] Traditional purity determination methods rely on complex chromatographic techniques (such as HPLC). These methods not only require time-consuming experimental processes but also expensive equipment support, and mostly rely on a single data source (such as chromatography), resulting in insufficient accuracy. This model directly predicts the purity of Dendrobium officinale extracts through the non-linear mapping of spectral data and the fusion of multi-modal data, can capture the complex effects of different components on purity more precisely, reduces the dependence on chromatographic analysis, saves time and costs, and realizes real-time and rapid judgment of whether the purity of Dendrobium officinale extracts meets the standard, improving the accuracy and robustness of the prediction.

[0040] The use of the above model includes: Data acquisition: Obtain the content of each component to be predicted through step S2 , obtain the reflection value of the spectral band from the hyperspectral data obtained through step S1 , calculate the coefficients in the model from the hyperspectral data obtained through step S1; Purity prediction: Input the obtained data and the calculated coefficients into the model to calculate the predicted purity value, and obtain the predicted purity value of the Dendrobium officinale extract ; Purity verification: Compare and verify the obtained purity value of the Dendrobium officinale extract with a sample of known purity. If the error is within the preset allowable range, it indicates that the model is accurate. If the error is large, adjust the coefficients for optimization.

[0041] Furthermore, the contribution coefficients include: ; Where: : The total content of all components; : The normalization coefficient of component k, indicating the deviation of the spectral intensity of this component relative to the average spectral intensity of all components.

[0042] Furthermore, the non-linear influence coefficients include: ; Where: : The standard deviation of the spectral signal of component k, indicating the spectral change amplitude of this component; : The spectral feature complexity factor of component k, reflecting the smoothness of the spectral signal.

[0043] Furthermore, the spectral feature complexity factor includes: ; Where: , : The spectral values of component k at wavelengths m and m - 1.

[0044] Furthermore, the sample importance coefficient includes: ; Where: : The spectral credibility of sample i.

[0045] Furthermore, the spectral credibility includes: ; : The spectral value of sample i at wavelength m; : The average spectral value of all samples at wavelength m.

[0046] S5. Identify the component spatial distribution based on the purity prediction result to obtain the component distribution information.

[0047] In this embodiment, the component spatial distribution identification includes: According to the purity prediction result, use the component spatial distribution identification model to identify the spatial distribution of different components in the Dendrobium officinale extract, analyze the uniformity of the components in the Dendrobium officinale extract, obtain the distribution of each component in the Dendrobium officinale extract in space, including the type, content, and purity of each component at each pixel point, and determine whether the component distribution of the Dendrobium officinale extract is uniform to improve the prediction accuracy.

[0048] Furthermore, the component spatial distribution identification model includes: ; Where: : The spatial uniformity index, which reflects the degree of spatial uniformity of each component in the Dendrobium officinale extract. The smaller the value, the higher the uniformity; : The number of pixels in the space, and the spatial resolution is determined by the spectral image; : The predicted content value of the k-th component at the i-th pixel point; : The predicted purity value at the i-th pixel point; : The adjustment coefficient, which is used to balance the influence of component uniformity and spectral volatility; : The spectral value fluctuation degree of the i-th pixel point, that is, the variance, which represents the local change degree of the component distribution.

[0049] This model describes a method for evaluating the spatial distribution of each component in the extract of Dendrobium officinale Kimura et Migo.

[0050] Traditional methods for spatial uniformity analysis usually rely on simple statistical indicators (such as standard deviation, mean, etc.) or image processing techniques. By combining the spatial information of spectral data with the component content, this model can comprehensively evaluate the spatial distribution of the extract of Dendrobium officinale Kimura et Migo, accurately capture the complex relationship between spectral data and component distribution, and improve the accuracy and precision of uniformity evaluation.

[0051] The use of the described model includes: Spatial feature extraction: Obtaining the spatial features of the spectral image from the hyperspectral data obtained through step S1; Parameter calculation: Obtaining the predicted values of component content at each pixel point through steps S2 and S4 and the predicted values of purity ; Uniformity calculation: Inputting the obtained data and the calculated parameters into the model for uniformity calculation to obtain the spatial uniformity index , and analyzing the uniformity of the components of the extract of Dendrobium officinale Kimura et Migo to obtain the spatial distribution of each component in the extract of Dendrobium officinale Kimura et Migo, including the type, content, and purity of each component at each pixel point. If the spatial uniformity index is large, it indicates that the component distribution of the extract sample of Dendrobium officinale Kimura et Migo is uneven, and the spatial uniformity of the extract can be improved by optimizing the extraction process or adjusting the sample treatment method.

[0052] S6. Calculating the purity according to the component distribution information.

[0053] In this embodiment, the purity calculation includes: Although the real-time and rapid prediction of the purity of the extract of Dendrobium officinale Kimura et Migo is achieved in step S4, if there are local impurity enrichments, adulterations, or uneven component distributions in the extract of Dendrobium officinale Kimura et Migo, the prediction method in this step may mask the real quality problems. Therefore, in step S6, it is necessary to calculate the purity of the extract of Dendrobium officinale Kimura et Migo to discover quality problems that cannot be identified by the spectrum, which is suitable for final quality control and batch evaluation and cannot obtain results as real-time and rapid as in step S4.

[0054] Based on the component distribution information obtained in step S5, that is, the distribution of each component in the Dendrobium officinale extract in space, including the type, content, and purity of each component at each pixel point, calculate the comprehensive spatial purity value of the Dendrobium officinale extract. First, assign a standard purity value to each component in the Dendrobium officinale extract (set according to the importance of the component in combination with the actual component analysis report. For example, the standard purity value of the target component polysaccharide is set to 0.95, the secondary component flavonoids is set to 0.85, the impurity class is set to 0.1, the interfering component is set to 0.05, and the blank area is set to 0). Then, combine the type and purity of each component at each pixel point in the component distribution information, calculate the corresponding purity value at all pixel points in the non-blank area of the space, and perform arithmetic averaging to obtain the comprehensive spatial purity value of the Dendrobium officinale extract. Compare the comprehensive spatial purity value with the set qualified purity threshold of the Dendrobium officinale extract (the qualified purity threshold is set according to actual requirements. For example, the qualified purity threshold is set to 0.85). If the comprehensive spatial purity value is greater than the qualified purity threshold, it is determined that the purity of the Dendrobium officinale extract is qualified; otherwise, further investigate the abnormal area or remove the unqualified part.

[0055] Example Two: This example also provides a computer device, applicable to the situation of an artificial intelligence-based Dendrobium officinale extraction purity detection method, including a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement an artificial intelligence-based Dendrobium officinale extraction purity detection method as proposed in the above example.

[0056] This example also provides a storage medium, on which a computer program is stored, and when the program is executed by the processor, it implements an artificial intelligence-based Dendrobium officinale extraction purity detection method as proposed in the above example.

[0057] The computer device may be a terminal, which includes a processor, a memory, a communication interface, a display screen, and an input device connected via a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner, and the wireless manner can be implemented through WIFI, carrier networks, NFC (Near Field Communication), or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads provided on the housing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0058] If a function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes: USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs, etc., which are various media that can store program codes.

[0059] The logic and / or steps represented in the flowchart or described in other ways herein, for example, can be considered as a definite sequence list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch instructions from the instruction execution system, apparatus, or device and execute the instructions), or in combination with these instruction execution systems, apparatuses, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device.

[0060] More specific examples (a non-exhaustive list) of computer-readable media include the following: electrical connections (electronic devices) having one or more wirings, portable computer diskettes (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber devices, and portable compact disc read-only memory (CDROM). Additionally, the computer-readable media can even be paper or other suitable media on which a program can be printed, as the program can be obtained electronically, for example, by optically scanning the paper or other media, then editing, interpreting, or otherwise processing it as appropriate, and then storing it in a computer memory.

[0061] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or a combination of the following techniques well known in the art can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.

[0062] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. A method for detecting the purity of Dendrobium officinale extraction based on artificial intelligence, characterized in that: The following steps are involved: S1. Hyperspectral data collection and component classification of Dendrobium officinale extract to obtain classification results; S2. Predicting the content according to the classification results to obtain a content prediction result; S3. Perform chromatographic analysis verification based on the content prediction result to obtain a verification result; S4. Perform purity prediction according to the verification result to obtain a purity prediction result; S5. Identify the spatial distribution of components according to the purity prediction results to obtain component distribution information; S6. Calculate purity based on component distribution information.

2. The method for detecting the purity of Dendrobium officinale extraction based on artificial intelligence according to claim 1, characterized in that: The content prediction in S2 includes: According to the classification results, the content prediction model was used to calculate the content of each component in the Dendrobium officinale extract.

3. The method for detecting the purity of Dendrobium officinale extraction based on artificial intelligence according to claim 2, characterized in that: The content prediction model includes: ； in: : The predicted content of the kth component; : The spectral value of the i-th sample in the j-th spectral band; : deviation term; : The weight of each sample; : model weight; : The total number of samples; : The total number of spectral bands.

4. The method for detecting the purity of Dendrobium officinale extraction based on artificial intelligence according to claim 1, characterized in that: The purity prediction in S4 includes: Purity prediction is performed based on the verification results and hyperspectral data using the purity prediction model.

5. The method for detecting the purity of Dendrobium officinale extraction based on artificial intelligence according to claim 4, characterized in that: The purity prediction model includes: ； in: : Predicted purity value; : The reflectance value of the spectral band corresponding to the kth component of the i-th sample; : bias term; : Contribution coefficient of the kth component; : nonlinear influence coefficient of the kth component; : Sample importance coefficient; : The number of component types considered in the model.

6. The method for detecting the purity of Dendrobium officinale extraction based on artificial intelligence according to claim 5, characterized in that: The contribution coefficients include: ； in: : Total content of all ingredients; : Normalization coefficient of component k.

7. The method for detecting the purity of Dendrobium officinale extraction based on artificial intelligence according to claim 5, characterized in that: The nonlinear influence coefficients include: ； in: : Standard deviation of the spectral signal of component k; : The spectral feature complexity factor of component k.

8. The method for detecting the purity of Dendrobium officinale extraction based on artificial intelligence according to claim 7, characterized in that: The spectral feature complexity factors include: ； in: , : Spectral values ​​of component k at wavelengths m and m-1.

9. The method for detecting the purity of Dendrobium officinale extraction based on artificial intelligence according to claim 5, characterized in that: The sample importance coefficients include: ； in: : Spectral credibility of sample i.

10. The method for detecting the purity of Dendrobium officinale extraction based on artificial intelligence according to claim 9, characterized in that: The spectral credibility includes: ； : Spectral value of sample i at wavelength m; : The average value of the spectrum of all samples at wavelength m.

Images