Method for detecting salidroside content in rhodiola extract based on artificial intelligence

Abstract

The application discloses a method for detecting rhodioloside content in rhodiol extract based on artificial intelligence, which comprises the following steps: S1, performing spectrum scanning of the extract sample at different wave bands in multiple near-infrared wavelength ranges; S2, performing multidimensional interference suppression processing and signal purification; S3, obtaining rhodioloside characteristic intensity factors through spectrum characteristic compression calculation; S4, performing rhodioloside characteristic signal-to-noise ratio enhancement and purity index calculation; S5, predicting the rhodioloside content and outputting a content prediction score; and S6, mapping the content prediction score to the final rhodioloside content. The application utilizes spectrum signal purification, nonlinear feature extraction, dynamic noise enhancement, artificial intelligence prediction modeling and secondary review mechanism, realizes adaptive analysis of extracts from different sources, significantly improves detection accuracy and stability, and can realize rapid and traceable rhodioloside content detection without large and precise instruments.

Description

Technical Field

[0001] This invention relates to the field of intelligent detection technology, specifically to an artificial intelligence-based method for detecting the content of rhodioloside in Rhodiola rosea extract. Background Technology

[0002] The detection of rhodioloside content mainly relies on laboratory analytical methods such as high-performance liquid chromatography (HPLC). While these methods offer high accuracy, they are time-consuming and highly dependent on equipment, making them unsuitable for rapid on-site testing. Some rapid detection methods estimate content based on a linear fit between the spectral signal and a fixed proportionality coefficient, which fails to address the nonlinear response characteristics in complex sample backgrounds. Furthermore, they lack dynamic correction mechanisms for matrix differences between different batches and extraction processes, resulting in insufficient prediction accuracy and stability. In addition, existing methods generally lack real-time verification of detection results, failing to adaptively correct rapid detection results and thus resulting in untimely correction of result deviations. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides an artificial intelligence-based method for detecting rhodioloside content in Rhodiola rosea extract, thereby solving the problems mentioned in the background section.

[0004] To achieve the above objectives, the present invention provides the following technical solution:

[0005] In a first aspect, the present invention provides a method for detecting the content of rhodioloside in Rhodiola rosea extract based on artificial intelligence, comprising the following steps:

[0006] S1. A multi-band synchronous acquisition device is used to perform spectral scanning of the extract sample in different bands within multiple near-infrared wavelength ranges, and the total signal energy and spectral intensity of each band are recorded.

[0007] S2. Based on the obtained total signal energy and spectral intensity of each band, perform multi-dimensional interference suppression and signal purification to obtain the purified effective signal intensity.

[0008] S3. Using the purified effective signal intensity as input, the characteristic intensity factor of rhodioloside is obtained by spectral feature compression calculation.

[0009] S4. Based on the characteristic intensity factor of rhodioloside, calculate the characteristic signal-to-noise ratio enhancement and purity index of rhodioloside;

[0010] S5. Using the calculated purity index, the content of rhodioloside is predicted by the prediction scoring model and the content prediction score is output.

[0011] S6. Map the content prediction score to the final rhodioloside content and quickly output the detection results.

[0012] To further optimize this technical solution, in step S1, a signal energy calculation model is established to uniformly measure the energy of spectral signals in different bands.

[0013] The signal energy calculation model is shown below:

[0014]

[0015] in,

[0016] Total signal energy;

[0017] The intensity of the spectral band acquired is the i-th band.

[0018] The weight of the i-th band;

[0019] This represents the total number of bands.

[0020] This model integrates multi-band spectral information in a weighted manner to avoid the impact of occasional interference from a single band on the overall signal characteristics.

[0021] To further optimize this technical solution, in step S2, the extract contains non-target compounds, including polysaccharides and flavonoids, which interfere with the spectral signal of rhodioloside. A dynamic characteristic peak comparison mechanism is introduced, which compares the reference characteristic peak database in real time to identify non-rhodioloside characteristic peaks and perform signal subtraction to purify the effective signal intensity.

[0022] To further optimize this technical solution, the calculation method for the effective signal strength after purification is as follows:

[0023]

[0024] in,

[0025] The effective signal strength after purification;

[0026] The intensity of the j-th interference peak detected;

[0027] Let be the suppression coefficient of the j-th interference peak;

[0028] This represents the total number of interfering peaks.

[0029] This model achieves pure spectral feature extraction of rhodioloside by calculating the weighted energy of the interfering components and subtracting them from the total signal.

[0030] To further optimize this technical solution, in step S3, a characteristic intensity factor calculation model is constructed, and by considering the nonlinear differences in the response of different bands to rhodioloside, a nonlinear compression function is used for fusion.

[0031] The feature intensity factor calculation model is as follows:

[0032]

[0033] in,

[0034] The characteristic intensity factor of rhodioloside;

[0035] The characteristic intensity of the reference standard rhodioloside sample in band i;

[0036] For correction factors;

[0037] The model outputs Used to characterize the relative closeness between the purification signal and the standard characteristics, directly reflecting the potential content level of rhodioloside.

[0038] To further optimize this technical solution, in step S4, when enhancing the signal-to-noise ratio of rhodioloside features, a signal-to-noise ratio enhancement model is constructed to dynamically attenuate the residual low-frequency noise in the purified signal.

[0039] The signal-to-noise ratio enhancement model is shown below:

[0040]

[0041] in,

[0042] To enhance signal strength;

[0043] This is residual noise energy;

[0044] This is the noise attenuation coefficient.

[0045] To further optimize this technical solution, after the signal-to-noise ratio enhancement model outputs the enhanced signal strength and residual noise energy, the purity index is calculated:

[0046]

[0047] in, This is a purity index used to measure the purity of the rhodioloside signal.

[0048] To further optimize this technical solution, in step S5, the prediction score model is as follows:

[0049]

[0050] in,

[0051] For content prediction score;

[0052] This is the score amplification factor;

[0053] This is the sensitivity correction factor;

[0054] This represents the theoretical purity value of the standard rhodioloside sample.

[0055] Parameters are continuously optimized through artificial intelligence training. and This enables it to accurately output the predicted score of rhodioloside content for different production batches of Rhodiola rosea extract. .

[0056] To further optimize this technical solution, in step S6, a dynamic ratio conversion model is constructed, and a nonlinear mapping relationship is established through the reference concentration gradient to achieve rapid quantitative analysis.

[0057] The dynamic scaling conversion model is as follows:

[0058]

[0059] in,

[0060] This represents the final rhodioloside content;

[0061] This is the average concentration value of the reference standard solution;

[0062] This refers to the dynamic gain coefficient.

[0063] By predicting the content score and Combined, the content of rhodioloside in actual samples can be quickly calculated.

[0064] To further optimize this technical solution, this method constructs an intelligent verification mechanism after detecting the rhodioloside content in the actual sample:

[0065] By selecting a very small amount of sample from the same batch of extract, rapid colorimetric or electrochemical response detection is performed using pre-set standard rapid reaction test strips or micro biosensors.

[0066] Cross-compare the content results with those output in step S6;

[0067] When the deviation between the two exceeds the set threshold, the testing personnel are prompted to conduct a second test or adjust the parameter weights of the prediction score model.

[0068] In a second aspect, the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program instructions are executed by the processor, the steps of the method for detecting the content of rhodioloside in Rhodiola rosea extract based on artificial intelligence as described in the first aspect of the present invention are implemented.

[0069] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program instructions are executed by a processor, the steps of the method for detecting the content of rhodioloside in Rhodiola rosea extract based on artificial intelligence as described in the first aspect of the present invention are implemented.

[0070] Compared with existing technologies, this invention provides an artificial intelligence-based method for detecting rhodioloside content in Rhodiola rosea extract, which has the following beneficial effects:

[0071] This AI-based rapid detection method for rhodioloside content in Rhodiola rosea extract addresses the challenge of balancing detection speed and result reliability in existing technologies by constructing a multi-step, layer-by-layer optimized intelligent detection process. Utilizing spectral signal purification, nonlinear feature extraction, dynamic noise enhancement, AI predictive modeling, and a secondary verification mechanism, this method achieves adaptive analysis of extracts from different sources, significantly improving detection accuracy and stability. It enables rapid and traceable detection of rhodioloside content without the need for large, precision instruments, effectively supporting on-site batch testing and industrial continuous monitoring needs. Attached Figure Description

[0072] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0073] Figure 1 This is a flowchart illustrating the method for detecting rhodioloside content in Rhodiola rosea extract based on artificial intelligence, as proposed in this invention. Detailed Implementation

[0074] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0075] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0076] Secondly, the term "an embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places throughout this specification does not necessarily refer to the same embodiment, nor is it a single embodiment or an embodiment selectively excluded from other embodiments.

[0077] Example 1:

[0078] Reference Figure 1 This is the first embodiment of the present invention, which provides an artificial intelligence-based method for detecting the content of rhodioloside in Rhodiola rosea extract, comprising the following steps:

[0079] S1. In the initial stage of detection, it is necessary to perform spectral scanning on the Rhodiola rosea extract. A multi-band synchronous acquisition device (such as a spectrometer) is used to perform spectral scanning on the extract sample in different bands within multiple near-infrared wavelength ranges, and the total signal energy and spectral intensity of each band are recorded.

[0080] In existing technologies, the detection of rhodioloside content mainly relies on traditional methods such as high-performance liquid chromatography (HPLC) and ultraviolet spectrophotometry. Although these methods have high accuracy, they suffer from problems such as long detection cycles, strict requirements for sample pretreatment, complex operation processes, high equipment costs, and difficulty in adapting to rapid on-site detection. Furthermore, existing spectral detection methods often employ fixed detection modes with single or a few wavelengths, failing to comprehensively capture the complex spectral information in Rhodiola rosea extract. This makes the detection signal susceptible to interference from impurities, reducing the accuracy of the results. Additionally, the signal energy cannot be uniformly quantified, leading to potential biases in subsequent data analysis.

[0081] Since the characteristic absorption intensities of rhodioloside molecules differ across different wavelengths, this invention significantly improves the completeness of detection information by simultaneously acquiring data from multiple wavelengths. A signal energy calculation model is established to uniformly measure the energy of spectral signals across different wavelengths.

[0082] The signal energy calculation model is shown below:

[0083]

[0084] in,

[0085] This represents the total signal energy.

[0086] The intensity of the i-th spectral band is denoted as . The Rhodiola rosea extract is scanned and acquired within a specific wavelength range using a high-sensitivity spectrophotometer, and the intensity value of each band is obtained using integrated photometric detection. Spectral data can be directly recorded using a multi-channel photodiode array or a CCD array.

[0087] The weight of the i-th band is determined by analyzing the contribution of each band signal to the characteristic absorption peak of rhodioloside through multiple batches of experiments on standard samples. It is typically assigned weights using artificial intelligence algorithms or expert experience to highlight bands sensitive to the detection of the target component. The value ranges from 0 to 1, and the sum of all band weights is usually standardized to 1 to ensure consistency in the total energy calculation.

[0088] This represents the total number of bands.

[0089] This model integrates multi-band spectral information in a weighted manner, avoiding the impact of occasional interference from a single band on the overall signal characteristics. It unifies the dimensions and importance of signals from different bands, reducing the adverse effects of high background noise or low-response bands on the overall detection results. Through this model, the detection system can characterize the entire spectral feature in the form of quantified energy, providing standardized input for subsequent steps.

[0090] S2. Based on the obtained total signal energy and spectral intensity of each band, perform multidimensional interference suppression and signal purification to obtain the purified effective signal intensity.

[0091] In existing technologies, interference processing of spectral signals during the detection of rhodioloside content generally relies on simple filtering or threshold denoising methods. These methods often use fixed frequency thresholds, making it difficult to accurately identify and subtract dynamic interference from non-target compounds such as polysaccharides and flavonoids in complex extracts. Traditional methods cannot adjust the intensity of interference peak subtraction based on real-time detection conditions, leading to weakened target signals or residual interference when the signal-to-noise ratio is high, ultimately affecting the accuracy and stability of the detection results.

[0092] Because extracts often contain non-target compounds such as polysaccharides and flavonoids, they can interfere with the spectral signal of rhodioloside. This invention introduces a dynamic characteristic peak comparison mechanism. Unlike traditional fixed threshold filtering, it compares each detected spectral peak against a rhodioloside characteristic peak database, automatically identifying and distinguishing interfering peaks from the target peak. This step simultaneously handles two main types of interference: one is interference from the chemical matrix of the sample itself (such as characteristic peak interference from non-target compounds like polysaccharides and flavonoids), and the other is physical noise generated by the external environment or equipment (such as ambient light, scattered light, and detector dark noise). It removes spectral overlap interference from sample components and suppresses external physical noise, ensuring comprehensive signal purification. The model precisely controls the reduction ratio of each type of interference to the total signal energy by setting suppression coefficients for different interference peaks, avoiding the accidental deletion of valid signals. This not only achieves higher purity of the target component signal in complex extracts but also automatically adapts to the interference characteristics of different batches and sources of samples, making the detection process more intelligent and adaptive.

[0093] The effective signal strength after purification is calculated as follows:

[0094]

[0095] in,

[0096] The effective signal strength after purification.

[0097] The intensity of the j-th interference peak detected. The noise signal intensity value measured by a spectrometer in a region without characteristic absorption peaks or in a reference blank sample, used to characterize non-target component signals such as ambient light and scattered light.

[0098] This represents the suppression coefficient for the j-th interference peak. The average influence coefficient for each type of noise is obtained by analyzing the background signal of multiple batches of Rhodiola rosea extract samples and fitting the results using blank solvent measurements or a noise database. The value typically ranges from 0 to 1, reflecting the proportion of noise intensity to the original signal.

[0099] This represents the total number of interference peaks.

[0100] This model achieves pure spectral feature extraction of rhodioloside by calculating the weighted energy of the interfering components and subtracting them from the total signal.

[0101] In actual use, the system automatically loads the interference peak database, compares the spectral spectrum in real time and calculates the energy of each interference peak, and dynamically adjusts the subtraction ratio to achieve accurate removal of noise and non-target component signals, significantly improving the accuracy and reliability of subsequent Rhodiola rosea glycoside content detection.

[0102] In this step, the process of identifying and distinguishing interference peaks from target peaks is as follows: First, multi-band spectral data is compared one by one with the rhodioloside characteristic peak database to pinpoint the theoretically expected location and relative intensity range of the characteristic peaks. Second, if the detected actual peak significantly deviates from the characteristic peak database record in terms of position, shape, or intensity ratio, it is determined to be an interference peak. Third, the interference intensity of the signal determined to be an interference peak is calculated. And based on the inhibition coefficient obtained from the fitting Weighted subtraction is performed on it; finally, the target peak signal that passes the matching determination is retained to form the purified effective signal. .

[0103] S3. After signal purification, in order to reduce redundant information and highlight the characteristics of rhodioloside, the purified effective signal intensity is used as input, and the characteristic intensity factor of rhodioloside is obtained by spectral feature compression calculation.

[0104] In existing technologies, linear regression or simple peak area integration methods are commonly used to estimate the characteristic intensity of target components when processing spectral signals. However, the multi-band spectral response of Rhodiola rosea extract often exhibits nonlinear characteristics, especially in the high concentration range, where the signal is prone to saturation or nonlinear amplification, causing the linear model to fail to accurately reflect the true content. Furthermore, existing technologies often lack effective compression strategies for processing multi-band information, leading to problems such as excessively high input data dimensionality, noise accumulation, and low computational efficiency, thus limiting the application effectiveness of rapid detection.

[0105] In this invention, a characteristic intensity factor calculation model is constructed. By considering the nonlinear differences in the response of rhodioloside to different bands, a nonlinear compression function is used for fusion to reduce the saturation distortion effect of high-concentration signals.

[0106] The feature intensity factor calculation model is as follows:

[0107]

[0108] in,

[0109] This is the characteristic intensity factor of rhodioloside.

[0110] The characteristic intensity of the reference standard rhodioloside sample in band i is used. The spectral response signal value under each calibration condition is recorded in the standard sample database and obtained through historical experimental data or online real-time reference detection. In this step, the detection object is a high-purity, known-concentration rhodioloside standard solution, representing the spectral intensity of the standard sample in the same band. This band represents the characteristic absorption or emission position of rhodioloside, hence the term "characteristic intensity." To easily distinguish between "unknown detection data" and "standard reference data," the terms "spectral intensity" and "characteristic intensity" are used in the naming.

[0111] This is a correction factor. By performing multiple spectral analyses on a standard rhodioloside solution, the influence of the spectral response on concentration prediction under each reference condition (such as different temperatures, pH levels, or solvent conditions) is calculated, and an artificial intelligence algorithm quantifies this influence. The value typically ranges from 0.1 to 5; a higher value indicates a stronger influence of the reference condition on the final content prediction.

[0112] The model outputs This factor is used to characterize the relative closeness of the purified signal to the standard characteristics, directly reflecting the potential content level of rhodioloside. A logarithmic function is used to reduce the error caused by excessive signal amplification in high-concentration samples. The system utilizes this factor as an input parameter for the next step of the artificial intelligence prediction model, significantly reducing redundant data dimensions, improving computational efficiency, and ensuring stable and reliable rhodioloside content detection results even in complex sample environments.

[0113] S4. Based on the characteristic intensity factor of rhodioloside, the signal-to-noise ratio enhancement and purity index of rhodioloside are calculated.

[0114] For spectral signal analysis of complex plant extracts, most existing methods rely solely on fixed filtering parameters or simple noise subtraction techniques to improve the signal-to-noise ratio. These methods typically ignore the variations in noise distribution caused by differences in the sample matrix, making it impossible to dynamically adjust the noise suppression process. When the target is Rhodiola rosea extract, non-target components (such as polysaccharides, amino acids, and pigments) introduce random noise at different frequencies. Traditional processing methods cannot effectively distinguish and dynamically adapt to these residual interference signals, easily leading to insufficient purity of the target rhodioloside signal, which in turn affects the accuracy of subsequent concentration prediction.

[0115] Step S3 utilizes some "noise patterns" in the original signal as auxiliary features for fitting, while step S4 further removes random noise that is detrimental to result presentation and subsequent storage after mapping, thus achieving a balance between retaining useful feature information and improving signal purity. This sequence is not simply noise reduction, but a strategy of feature extraction followed by signal purification, ensuring that the final output is both accurate and easy for subsequent analysis and application.

[0116] When performing signal-to-noise ratio enhancement for rhodioloside features, a signal-to-noise ratio enhancement model is constructed to dynamically attenuate residual low-frequency noise in the purified signal.

[0117] The signal-to-noise ratio enhancement model is shown below:

[0118]

[0119] in,

[0120] To enhance signal strength.

[0121] This refers to residual noise energy. The residual background signal intensity in non-characteristic peak regions is statistically analyzed using spectral analysis, or residual noise energy is measured in real-time using noise detection techniques.

[0122] This represents the noise attenuation coefficient. Through multiple batches of experiments comparing the impact of residual noise on the prediction results, a fitting method or an artificial intelligence adaptive algorithm is used to determine the optimal attenuation coefficient. The value typically ranges from 0.01 to 5; a larger value indicates stronger noise suppression of the signal.

[0123] After the signal-to-noise ratio enhancement model outputs the enhanced signal strength and residual noise energy, the purity index is calculated:

[0124]

[0125] in, This is a purity index used to measure the purity of the rhodioloside signal. Values ​​range from 0 to 1; values ​​close to 1 indicate a clean spectral signal with extremely low noise interference.

[0126] This step can adaptively adjust the signal enhancement intensity according to the actual noise level of each test, making the signal purification result more consistent with the actual sample characteristics, thereby improving the distinction between the rhodioloside signal and background noise, and significantly improving the input data quality for downstream artificial intelligence content prediction.

[0127] In actual use, the detection system automatically adjusts according to the real-time noise level. The enhancement effect is more pronounced in samples with high noise levels, while avoiding over-amplification errors in samples with low noise levels, thus achieving more accurate signal processing. This method overcomes the problem of insufficient noise suppression in complex extracts by traditional techniques, providing high-purity feature input signals for subsequent AI-based content prediction.

[0128] S5. Using the calculated purity index, the content of rhodioloside is predicted by the prediction scoring model and the content prediction score is output.

[0129] Rhodiola rosea glycoside content prediction often employs traditional modeling methods based on linear fitting or multiple regression, estimating content by simply comparing the difference between the input signal purity parameter and the standard reference concentration curve. In practical applications, these methods are easily affected by factors such as sample batch variations, noise fluctuations, and spectral nonlinear responses, resulting in poor model generalization ability. Frequent parameter recalibration is required for different extract samples, leading to low detection efficiency and difficulty in guaranteeing the accuracy and stability of prediction scores, thus failing to meet the needs of rapid detection and industrial-scale batch testing.

[0130] The prediction score model constructed in this invention is as follows:

[0131]

[0132] in,

[0133] This is the content prediction score. A higher score indicates that the test result is closer to the standard reference conditions, and the prediction accuracy is higher.

[0134] This is the score amplification factor. It is obtained through training on historical detection data and represents the weight of signal purity's contribution to the prediction result. The optimal value is automatically determined by the algorithm. The value typically ranges from 0.5 to 5; a larger value indicates a stronger influence of signal purity on the prediction score.

[0135] This is the sensitivity correction coefficient. Obtained through a machine learning optimization process, it adjusts the sensitivity of the predicted score to deviations from the standard purity signal, ensuring the results better conform to the standard reference conditions. It is generally between 0.1 and 10; a larger value results in a stronger penalty for deviation.

[0136] This represents the theoretical purity value of the standard rhodioloside sample. A reference purity value was obtained through multiple experiments on the standard rhodioloside solution and serves as the baseline input parameter for the predictive scoring model. The value typically ranges from 0.8 to 1, set according to standard laboratory conditions.

[0137] Parameters are continuously optimized through artificial intelligence training. and This enables it to accurately output the predicted score of rhodioloside content for different production batches of Rhodiola rosea extract. .

[0138] This step not only utilizes spectral purity indicators but also incorporates dynamic correction predictions based on deviations from the theoretical purity values ​​of standard rhodioloside, enabling rapid adaptation to extracts from different sources and under varying processing conditions. Compared to traditional fixed-parameter linear models, this method possesses self-learning and adaptive capabilities, allowing for parameter optimization through artificial intelligence algorithms. This significantly improves the accuracy and robustness of content prediction, reduces manual parameter tuning, and enhances the overall automation level of the detection process.

[0139] S6. Map the content prediction score to the final rhodioloside content and quickly output the detection results.

[0140] In existing technologies, quantitative analysis of rhodioloside content largely relies on laboratory detection methods such as high-performance liquid chromatography (HPLC). While these methods offer high accuracy, they are time-consuming and require complex pretreatment steps and expensive instruments. Even the few rapid detection methods available often directly convert spectral signal intensity or predicted scores into content using a fixed proportionality coefficient, lacking the ability to dynamically correct for batch-to-batch differences and fluctuations in extraction processes. As a result, quantitative results show significant errors for extracts from different sources and with varying concentration gradients, indicating insufficient reliability and universality of rapid detection methods.

[0141] This invention constructs a dynamic proportional conversion model and establishes a nonlinear mapping relationship through a reference concentration gradient to achieve rapid quantitative analysis.

[0142] The dynamic scaling conversion model is as follows:

[0143]

[0144] in,

[0145] This represents the final content of rhodioloside.

[0146] This serves as a reference for the average concentration of standard solutions. The concentration of rhodioloside in standard solutions is established based on known values, typically determined by high-purity standard solutions prepared in a laboratory and stored in a database.

[0147] This is the dynamic gain coefficient. It is obtained by fitting a large amount of sample detection data with the actual measured concentration values, and is used to reflect the sensitivity of the prediction score to concentration correction. It is usually between 0.01 and 1, and the larger the value, the more significant the influence of the prediction score on the final concentration result.

[0148] By predicting the content score and Combined, the content of rhodioloside in actual samples can be quickly calculated.

[0149] Compared to traditional methods, this step can adapt to various uncertainties such as fluctuations in raw material concentration and differences in extraction processes, enabling more flexible and adaptive quantitative calculations and significantly improving the accuracy and versatility of rapid detection methods in different production environments.

[0150] In this embodiment, although rapid detection results of rhodioloside content have been given, slight deviations in detection results may occur due to matrix effects caused by differences in raw material origin, extraction solvent ratio, temperature fluctuations, etc. between different batches of extract.

[0151] Therefore, this method constructs an intelligent verification mechanism after detecting the rhodioloside content in the actual sample:

[0152] By selecting a very small amount of sample from the same batch of extracts, and using a pre-set standard rapid reaction test strip or a micro biosensor, rapid colorimetric or electrochemical response detection is performed, and the verification signal serves as a "soft verification" reference value.

[0153] The reference value is cross-compared with the content result output in step S6;

[0154] When the deviation between the two exceeds the set threshold, the inspector is prompted to conduct a second inspection, or the parameter weights of the prediction score model are adjusted to adapt to the new matrix characteristics.

[0155] Compared to existing technologies, this step significantly improves the stability and reliability of rapid detection methods. At the same time, it eliminates the need for large instruments or extended detection time, relying solely on low-cost, rapid biochemical sensing to assist in judgment. This significantly enhances the robustness of the detection system, making it particularly suitable for continuous testing on production lines and real-time verification of large batches of samples, ensuring more reliable and traceable results.

[0156] Example 2:

[0157] This embodiment also provides a computer device applicable to the detection method of rhodioloside content in Rhodiola rosea extract based on artificial intelligence, 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 realize the detection method of rhodioloside content in Rhodiola rosea extract based on artificial intelligence as proposed in the above embodiment.

[0158] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the artificial intelligence-based method for detecting the content of rhodioloside in Rhodiola rosea extract as proposed in the above embodiments.

[0159] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0160] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the 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 to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0161] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0162] More specific examples (a non-exhaustive list) of computer-readable media include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0163] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0164] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for detecting rhodioloside content in Rhodiola rosea extract based on artificial intelligence, characterized in that, Includes the following steps: S1. A multi-band synchronous acquisition device is used to perform spectral scanning of the extract sample in different bands within multiple near-infrared wavelength ranges, and the total signal energy and spectral intensity of each band are recorded. S2. Based on the obtained total signal energy and spectral intensity of each band, perform multi-dimensional interference suppression and signal purification to obtain the purified effective signal intensity. S3. Using the purified effective signal intensity as input, the characteristic intensity factor of rhodioloside is obtained by spectral feature compression calculation. S4. Based on the characteristic intensity factor of rhodioloside, calculate the characteristic signal-to-noise ratio enhancement and purity index of rhodioloside; S5. Using the calculated purity index, the content of rhodioloside is predicted by the prediction scoring model and the content prediction score is output. S6. Map the content prediction score to the final rhodioloside content and quickly output the detection results.

2. The method for detecting rhodioloside content in Rhodiola rosea extract based on artificial intelligence according to claim 1, characterized in that, In step S1, a signal energy calculation model is established to perform a unified energy measurement of spectral signals in different bands. The signal energy calculation model is shown below: ； in, Total signal energy; The intensity of the spectral band acquired is the i-th band. The weight of the i-th band; This represents the total number of bands. This model integrates multi-band spectral information in a weighted manner to avoid the impact of occasional interference from a single band on the overall signal characteristics.

3. The method for detecting rhodioloside content in Rhodiola rosea extract based on artificial intelligence according to claim 1, characterized in that, In step S2, a dynamic feature peak comparison mechanism is introduced. By comparing the reference feature peak database in real time, non-rhodioloside feature peaks are identified and signal subtraction is performed to purify the effective signal intensity.

4. The method for detecting rhodioloside content in Rhodiola rosea extract based on artificial intelligence according to claim 3, characterized in that, The effective signal strength after purification is calculated as follows: ； in, The effective signal strength after purification; The intensity of the j-th interference peak detected; Let be the suppression coefficient of the j-th interference peak; This represents the total number of interfering peaks. This model achieves pure spectral feature extraction of rhodioloside by calculating the weighted energy of the interfering components and subtracting them from the total signal.

5. The method for detecting rhodioloside content in Rhodiola rosea extract based on artificial intelligence according to claim 1, characterized in that, In step S3, a characteristic intensity factor calculation model is constructed, and the nonlinear differences in the response of different bands to rhodioloside are considered, and the nonlinear compression function is used for fusion. The feature intensity factor calculation model is as follows: ； in, The characteristic intensity factor of rhodioloside; The characteristic intensity of the reference standard rhodioloside sample in band i; For correction factors; The model outputs Used to characterize the relative closeness between the purification signal and the standard characteristics, directly reflecting the potential content level of rhodioloside.

6. The method for detecting rhodioloside content in Rhodiola rosea extract based on artificial intelligence according to claim 1, characterized in that, In step S4, when performing signal-to-noise ratio enhancement of rhodioloside features, a signal-to-noise ratio enhancement model is constructed to dynamically attenuate residual low-frequency noise in the purified signal. The signal-to-noise ratio enhancement model is shown below: ； in, To enhance signal strength; This is residual noise energy; This is the noise attenuation coefficient.

7. The method for detecting rhodioloside content in Rhodiola rosea extract based on artificial intelligence according to claim 6, characterized in that, After the signal-to-noise ratio enhancement model outputs the enhanced signal strength and residual noise energy, the purity index is calculated: ； in, This is a purity index used to measure the purity of the rhodioloside signal.

8. The method for detecting rhodioloside content in Rhodiola rosea extract based on artificial intelligence according to claim 1, characterized in that, In step S5, the prediction score model is as follows: ； in, The content prediction score; This is the score amplification factor; This is the sensitivity correction factor; This represents the theoretical purity value of the standard rhodioloside sample. Parameters are continuously optimized through artificial intelligence training. and This enables it to accurately output the predicted score of rhodioloside content for different production batches of Rhodiola rosea extract. .

9. The method for detecting rhodioloside content in Rhodiola rosea extract based on artificial intelligence according to claim 1, characterized in that, In step S6, a dynamic proportional conversion model is constructed, and a nonlinear mapping relationship is established through the reference concentration gradient to achieve rapid quantitative analysis. The dynamic scaling conversion model is as follows: ； in, This represents the final rhodioloside content; This is the average concentration value of the reference standard solution; This refers to the dynamic gain coefficient. By predicting the content score and Combined, the content of rhodioloside in actual samples can be quickly calculated.

10. The method for detecting rhodioloside content in Rhodiola rosea extract based on artificial intelligence according to claim 9, characterized in that, This method constructs an intelligent verification mechanism after detecting the rhodioloside content in the actual sample: By selecting a very small amount of sample from the same batch of extract, rapid colorimetric or electrochemical response detection is performed using pre-set standard rapid reaction test strips or micro biosensors. Cross-compare the content results with those output in step S6; When the deviation between the two exceeds the set threshold, the testing personnel are prompted to conduct a second test or adjust the parameter weights of the prediction score model.

Images