Acid value determination method and system based on AI image recognition

Abstract

The invention provides an acid value determination method and system based on AI image recognition, and belongs to the field of AI image recognizing.The method comprises the steps that data are collected in real time in the acid value titration process, and the data collected in real time are preprocessed; constructing a color gradient dynamic extraction model, and generating a comprehensive feature matrix; calculating the feature importance weight of the feature mutation point, quantifying the contribution degree of the feature mutation point in a key time period, weighting the comprehensive feature matrix, generating a weighted feature matrix, extracting a titration end point to obtain a titration end point prediction result, and calculating the volume of a titrant according to the titration end point; constructing a confidence scoring model to perform confidence scoring on the titration end point prediction time, and calculating the final titrant volume according to the final end point time; and calculating the acid value according to the final end point time and generating an experimental report. According to the method, the defects in the prior art are effectively overcome, a comprehensive, efficient and intelligent solution is provided for the field of acid value detection, and the method has extremely high application value.

Description

Technical Field

[0001] The present invention belongs to the field of AI image recognition, and in particular relates to an acid value determination method and system based on AI image recognition. Background Art

[0002] In the field of acid value detection, titration is widely used in chemical, food, pharmaceutical and environmental testing to evaluate the acidic components in liquid or solid samples. Traditional titration methods mainly rely on manual judgment of the titration endpoint or monitoring color changes through simple optical detection equipment. However, these methods have obvious limitations: Large errors in manual judgment: Manual judgment of the endpoint is based on the operator's subjective perception and is easily affected by personal experience, reaction speed, and changes in ambient light, resulting in inaccurate results and poor experimental repeatability.

[0003] Simple optical detection has low robustness: Some automated equipment uses a single light source or color sensor to detect color changes, but it is not adaptable enough to complex experimental conditions (such as uneven lighting and changes in sample transparency), and the detection results are prone to deviate from the actual titration endpoint.

[0004] Lack of modeling of dynamic characteristics of chemical reactions: Existing titration monitoring systems are mainly based on simple algorithms of static color or optical changes, which cannot capture dynamic changes during the titration process (such as color gradient trends or spectral feature changes), and show significant limitations in titration experiments with complex reactions.

[0005] Single application of spectral data: As an important means of characterizing reactions, spectral data is often used in a single application in titration detection and has not been fully explored. The endpoint is determined only by the absorbance threshold, and it is impossible to comprehensively analyze the changes in color and spectrum for accurate analysis.

[0006] To address the above issues, there is an urgent need for a method that can fully utilize multimodal data under complex experimental conditions to capture the dynamic changes in the titration process, and combine it with an intelligent analysis algorithm to achieve accurate identification of the titration endpoint and rapid calculation of the acid value. Summary of the invention

[0007] The purpose of the present invention is to propose an acid value determination method and system based on AI image recognition, which effectively overcomes the shortcomings of the prior art and provides a comprehensive, efficient and intelligent solution for the field of acid value detection, with extremely high application value.

[0008] In order to achieve the above object, a first aspect of the present invention provides an acid value determination method based on AI image recognition, the method comprising: S1. During the acid value titration process, data is collected in real time, and the real-time collected data is preprocessed based on time series consistency to obtain preprocessed data; wherein the data includes image data and spectral data; S2, constructing a color gradient dynamic extraction model to extract image dynamic features from the preprocessed image data, constructing an absorbance change dynamic extraction model to extract spectral dynamic features from the preprocessed spectral data, and combining the image dynamic features and spectral dynamic features to generate a comprehensive feature matrix; S3, calculating the feature importance weight of the feature mutation point according to the comprehensive feature matrix, which is used to quantify the contribution of the feature mutation point in the key time period, and then weighting the comprehensive feature matrix based on the feature importance weight to generate a weighted feature matrix, and within the key time window, extracting the titration endpoint through a nonlinear judgment model based on the weighted feature matrix to obtain a titration endpoint prediction result, and at the same time calculating the titrant volume according to the titration endpoint to obtain the titration endpoint prediction time and titrant volume; S4. Construct a confidence scoring model to score the confidence of the titration endpoint prediction time, determine whether the titration endpoint prediction time needs to be corrected, and recalculate the final titrant volume according to the final endpoint time: If the confidence score is less than a preset threshold, it is considered that there is a potential error in the titration endpoint prediction time, and correction is performed, and the corrected endpoint time is output as the final endpoint time. The correction is achieved through a nonlinear error correction model, and the nonlinear error correction model is achieved in combination with the time series smoothness of the weighted feature matrix; If the confidence score is greater than or equal to the preset threshold, it is considered that there is no potential error in the predicted titration endpoint time and it is directly output as the final endpoint time; S5. Calculate the acid value according to the final endpoint time and generate a test report.

[0009] Furthermore, the preprocessing also includes introducing a dynamic illumination normalization model , for image data To perform adaptive light intensity correction: ; in, is the rectified image; is the image width and height; and are the mean and standard deviation of the light intensity of the image, respectively; is a regularization parameter to prevent over-normalization, is the pixel value of the i-th row and j-th column in the image, i is the row index of the image, and j is the column index of the image; Using Gaussian filtering based on frequency domain To perform noise suppression: ; in, is the preprocessed image data, is the frequency domain component of the image, is the noise control parameter, which is dynamically optimized through experiments; The preprocessing also includes using an adaptive baseline correction model Baseline correction and dynamic feature enhancement of spectral data: ; in, is the corrected spectral data, i.e. the preprocessed spectral data; is the second-order dynamic fitting baseline; It is a dynamic fitting parameter, which is optimized in real time to adapt to different experimental conditions; After correction, the spectral data Extract the dynamic change rate of absorbance : ; in, is the wavelength of the spectral data, For time, is the spectral data after correction.

[0010] Furthermore, the color gradient dynamic extraction model is expressed as: ; in, is the image feature matrix, each column represents the time The eigenvector of is the preprocessed image data; and are convolution kernels and biases, specifically designed to capture color variations; It is a color gradient dynamic extraction model; is the color gradient regularization term, which is used to emphasize the color mutation point; is the regularization coefficient, which controls the weight of the mutation point in the model; is an activation function that retains the salient features of color changes; Dynamically extract models via color gradients Process the image frames at each time point , extract the color change sequence matrix : ; Among them, each column is the color feature of a time point, and T is the total number of time points; The absorbance change dynamic extraction model , expressed as: ; in, is the spectral feature matrix, each column represents the time The absorbance characteristics of is the preprocessed spectral data; It represents the rate of change of absorbance over time and is used to capture dynamic changes; It is the second-order characteristic of absorbance changing with wavelength, which is used to identify bands with gradual or sudden changes; is the regularization coefficient, which is used to balance the variation information of time and wavelength; By Model For each time point, the spectral data Processing to generate spectral change feature matrix : ; Each column represents the spectral characteristics at a certain time point.

[0011] Furthermore, the image dynamic features and spectrum dynamic features are integrated to generate a comprehensive feature matrix, which is expressed as: Introducing modal weighted fusion model ,Will and Fusion into comprehensive feature matrix : ; in, and is a weight matrix, which is dynamically optimized according to the importance of the modal data; is the modality alignment regularization term, which is used to reduce the modality difference between image and spectral features; is a regularization parameter that controls the degree of mode alignment.

[0012] Furthermore, the feature importance weight of the feature mutation point is calculated according to the comprehensive feature matrix, specifically: ; in, A specific time point The feature importance weights of Controlling feature strength The impact of Indicates a specific point in time , represents the specific time point of the current processing; is the loop variable index at the time point; Controlling temporal proximity changes The impact of is the eigenvector of the tth column in the comprehensive feature matrix, representing the multimodal features at a specific time point t. is the eigenvector of the tth column in the comprehensive feature matrix, representing the multimodal features at a specific time point t-1; the denominator is normalized to ensure ; represents the two-norm; Based on a specific point in time The feature importance weights The feature importance weight vector The comprehensive feature matrix Perform weighting to generate a weighted feature matrix , where the comprehensive feature matrix is a matrix containing features at all time points: ; in, is the feature importance weight vector, in the weighted feature matrix Extract the time point with the largest feature intensity , and nearby time windows ,in is the window length, reflecting the key change stage of the titration reaction.

[0013] Furthermore, in the critical time window Internally, based on the weighted feature matrix Through nonlinear decision model Extraction titration endpoint prediction time , and predict the time based on the titration endpoint Calculating titrant volume : ; in, is the predicted time of the titration endpoint, given by Calculated based on the changing trend of multimodal features; It is a titrant volume calculation function, combining titration rate and time calculation.

[0014] Furthermore, the confidence scoring model is expressed as: ; in, Predicting the time for the titration endpoint The confidence score of ; Indicates the magnitude of multimodal feature changes near the endpoint, which is used to evaluate the significance of the endpoint change. is the weighted multimodal feature vector of the titration endpoint, is the weighted multimodal feature vector at the moment before the titration endpoint, where yes Predicting the time at the titration endpoint The specific value of Indicates the rate of change of titrant volume, used to capture abnormal titration reactions; and are adjustment parameters that control the impact of feature changes and volume changes on confidence respectively; The nonlinear error correction model , expressed as: ; in, is the corrected titration endpoint time; is the correction term for the prediction error, given by The time series features are weighted averaged to obtain: is the weighted multimodal feature vector at time point t; is the error correction factor used to balance the magnitude of the correction.

[0015] Furthermore, the S5 specifically includes: Obtain experimental parameters, including titrant concentration, sample weight, and conversion factor; Calculate the acid value according to the experimental parameters: ; in, For acid value results, part 1 , is the basic acid value calculation formula, is the titrant concentration, is the final titrant volume, F is the conversion factor, is the weight of the sample; correction term Dynamically adjust the acid value results according to the confidence level. When the confidence level is low, increase the conservative deviation to reduce the impact of experimental error. is the adjustment coefficient, which controls the weight of the confidence correction term on the result. () is the confidence score of the titration endpoint prediction; Generate a test report based on the acid value, including: Acid value results ; Titration process records, including endpoint correction details and confidence scores; Graphical display; Complete data for the report, ensuring that the report contains all experimental input and output data.

[0016] In another aspect of the present invention, an acid value determination system based on AI image recognition is provided, the system comprising: An acid value titration acquisition unit, used for real-time data acquisition during the acid value titration process, and preprocessing the real-time acquired data based on time sequence consistency to obtain preprocessed data; wherein the data includes image data and spectral data; A feature extraction unit is used to construct a color gradient dynamic extraction model to extract image dynamic features from the preprocessed image data, to construct an absorbance change dynamic extraction model to extract spectral dynamic features from the preprocessed spectral data, and to generate a comprehensive feature matrix by combining the image dynamic features and spectral dynamic features; A titration analysis unit, used for calculating the feature importance weight according to the comprehensive feature matrix to quantify the contribution of the feature mutation point in the key time period, and then weighting the comprehensive feature matrix based on the feature importance weight to generate a weighted feature matrix, and extracting the titration endpoint through a nonlinear judgment model based on the weighted feature matrix within the key time window to obtain a titration endpoint prediction result, and at the same time calculating the titrant volume according to the titration endpoint to obtain the titration endpoint prediction time and titrant volume; The analysis and correction unit is used to construct a confidence scoring model to perform confidence scoring on the titration endpoint prediction time, determine whether the titration endpoint prediction time needs to be corrected, and recalculate the final titrant volume based on the final endpoint time: If the confidence score is less than a preset threshold, it is considered that there is a potential error in the titration endpoint prediction time, and correction is performed, and the corrected endpoint time is output as the final endpoint time. The correction is achieved through a nonlinear error correction model, and the nonlinear error correction model is achieved in combination with the time series smoothness of the weighted feature matrix; If the confidence score is greater than or equal to the preset threshold, it is considered that there is no potential error in the predicted titration endpoint time and it is directly output as the final endpoint time; The report generation unit is used to calculate the acid value according to the final endpoint time and generate an experimental report.

[0017] The beneficial technical effects of the present invention are at least as follows: (1) The present invention captures the dynamic change trend of liquid color in real time during the titration process by introducing an improved adaptive multimodal analysis model (a nonlinear judgment model based on a support vector machine). This technology can not only identify the color mutation point, but also analyze the speed and direction of the color gradient change, thereby accurately locating the titration endpoint. Compared with the traditional static color analysis method, this technology has a stronger adaptability to the interference of ambient light, solving the problems of large manual judgment errors and low robustness of simple optical detection.

[0018] (2) The present invention uses a multimodal Transformer model to deeply fuse the image data and spectral absorbance data collected during the titration process. This technology makes full use of the cross-characteristics of spectral features (such as the derivative of absorbance change and peak position) and image color change to comprehensively determine the titration endpoint. Compared with a single spectral or image analysis method, this method can significantly improve the accuracy of titration endpoint detection under complex conditions and solve the problem of single application of spectral data.

[0019] (3) Based on endpoint detection, the present invention designs an automatic acid value calculation module, which directly uses the data of the endpoint detection (titrant volume, sample mass) in the acid value calculation formula to generate and automatically record the experimental results. This module can significantly improve the efficiency and accuracy of the experiment, adapt to high-throughput detection scenarios, and solve the problem of poor repeatability of traditional experiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The present invention is further described using the accompanying drawings, but the embodiments in the accompanying drawings do not constitute any limitation to the present invention. A person skilled in the art can obtain other drawings based on the following drawings without creative work.

[0021] Figure 1 This is a flow chart of a method for determining an acid value based on AI image recognition according to an embodiment of the present invention.

[0022] Figure 2 This is a framework diagram of an acid value determination system based on AI image recognition according to an embodiment of the present invention. DETAILED DESCRIPTION

[0023] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0024] like Figure 1 As shown, the acid value determination method based on AI image recognition provided by an embodiment of the present invention comprises the following steps S1-S5: S1. During the acid value titration process, data is collected in real time, and preprocessing is performed on the real-time collected data based on time series consistency to obtain preprocessed data; wherein the data includes image data and spectral data.

[0025] Specifically, during the titration process, the following data are collected in real time: Image data :Using high-definition cameras, the frame rate per second is RGB image sequence to record the color change process of the titrant.

[0026] Spectral data :Through a multi-channel spectral sensor, the wavelength Record absorbance in units The dynamic change of .

[0027] Synchronization of image and spectral data: ; in, is the acquisition timestamp, Innovative analysis: To overcome the challenge of fast titration reaction speed and instantaneous changes, a dynamic time synchronization method is used to ensure the temporal consistency of image and spectral data, thereby avoiding asynchronous errors affecting feature analysis.

[0028] Furthermore, in order to cope with the problems of uneven lighting and complex background in the experimental environment, it is necessary to ensure the accurate capture of the color change of the titrant.

[0029] Introducing a dynamic lighting normalization model , for image data To perform adaptive light intensity correction: ; in, is the rectified image; is the image width and height; and are the mean and standard deviation of the light intensity of the image, respectively; is a regularization parameter (dynamic adjustment) to prevent over-normalization. Different from traditional normalization, The dynamic adjustment is achieved based on the real-time detection results of background light changes during the titration process, thus adapting to changing experimental conditions.

[0030] Noise suppression: using frequency-domain based Gaussian filtering : ; in, is the frequency domain component of the image, It is a noise control parameter, which is dynamically optimized through experiments. It removes high-frequency noise while retaining the color gradient characteristics of the titration solution.

[0031] Furthermore, the baseline drift problem of spectral data caused by experimental equipment offset and external interference during the titration process is solved, and the key dynamic absorbance change characteristics are extracted.

[0032] Model design: Adopting adaptive baseline correction model : ; in, is the corrected spectral data; is the second-order dynamic fitting baseline; It is a dynamic fitting parameter, which is optimized in real time to adapt to different experimental conditions.

[0033] After correction, the spectral data , extract the dynamic change rate of absorbance: ; Capture the instantaneous characteristics of absorbance changes and provide dynamic feature input for subsequent endpoint identification.

[0034] Furthermore, the preprocessed data is output and ,in: It is a sequence of images after dynamic normalization and noise suppression, containing accurate color change features; It is the corrected and enhanced spectral data, which retains the dynamic characteristics of absorbance during the titration process.

[0035] The output serves as input for subsequent steps, ensuring uniform data format and high-quality signals.

[0036] S2. Construct a color gradient dynamic extraction model to extract image dynamic features from the preprocessed image data, construct an absorbance change dynamic extraction model to extract spectral dynamic features from the preprocessed spectral data, and integrate the image dynamic features and spectral dynamic features to generate a comprehensive feature matrix.

[0037] Specifically, receive the preprocessed data output from step 1 and : It is an image sequence that has been normalized for illumination and suppressed for noise, and contains the time series variation characteristics of the titration solution color; It is the baseline-corrected spectral data, which contains the information of the absorbance of the titrant changing with time and wavelength.

[0038] Time synchronization conditions: Ensure and Align by timestamp to ensure dynamic consistency of subsequent feature extraction.

[0039] Furthermore, the time series characteristics of the color change of the titration solution, especially the dynamic changes of the color gradient, are extracted to capture the color gradual and sudden change information during the titration process.

[0040] Model design: Defining a color gradient dynamic extraction model : ; in, is the image feature matrix, each column represents the time The eigenvector of and are convolution kernels and biases, specifically designed to capture color variations; is the color gradient regularization term, which is used to emphasize the color mutation point; is the regularization coefficient, which controls the weight of the mutation point in the model; is an activation function that retains the salient features of color changes.

[0041] By Model Process the image frames at each time point , extract the color change sequence matrix : ; Each column is the color feature of a time point, and T is the total number of time points.

[0042] Furthermore, the goal is to extract key features of absorbance changes in spectral data, including the dynamic rate of change over time and the gradual change pattern over the wavelength range.

[0043] Model design: Building a dynamic extraction model for absorbance changes : ; in, is the spectral feature matrix, each column represents the time Absorbance characteristics of It represents the rate of change of absorbance over time and is used to capture dynamic changes; It is the second-order characteristic of absorbance changing with wavelength, which is used to identify bands with gradual or sudden changes; is the regularization coefficient, which is used to balance the variation information of time and wavelength.

[0044] By Model For each time point, the spectral data Processing to generate spectral change feature matrix : ; Each column represents the spectral characteristics at a certain time point.

[0045] Furthermore, the image and spectral features are integrated to generate a unified feature matrix that captures the dynamic changing patterns during the titration process.

[0046] Fusion method: Introducing modal weighted fusion model ,Will and Fusion into comprehensive feature matrix : ; in, and is a weight matrix, which is dynamically optimized according to the importance of the modal data; is the modality alignment regularization term, which is used to reduce the modality difference between image and spectral features; is a regularization parameter that controls the degree of mode alignment.

[0047] Finally, output: comprehensive feature matrix , each column represents the multimodal features at a certain moment, which is used for accurate identification of the subsequent titration endpoint. By dynamically extracting and fusing image and spectral features, it is ensured that the feature matrix fully and accurately reflects the dynamic process of the titration reaction, providing high-quality input for the core tasks of the patent.

[0048] S3. Calculate the feature importance weight of the feature mutation point according to the comprehensive feature matrix to quantify the contribution of the feature mutation point in the critical time period, then weight the comprehensive feature matrix based on the feature importance weight to generate a weighted feature matrix, and within the critical time window, extract the titration endpoint through a nonlinear judgment model based on the weighted feature matrix to obtain a titration endpoint prediction result, and calculate the titrant volume according to the titration endpoint to obtain the titration endpoint prediction time and titrant volume.

[0049] Specifically, receive the output comprehensive feature matrix of step 2 , whose time series are synchronized: The image features and spectral characteristics The matrix generated by modal weighting and regularization fusion, each column represents the multimodal features at a certain moment, reflecting the dynamic changes of the color and absorbance of the titration solution.

[0050] Furthermore, for the comprehensive feature matrix Assign importance weights to each column (i.e., features at a time point) to capture the contribution of feature mutation points and key time periods. Define the feature importance weight assignment formula: ; in, A specific time point The feature importance weights of Controlling feature strength (i.e., the combined intensity of color and absorbance changes); Indicates a specific point in time , represents the specific time point of the current processing; is the loop variable index at the time point; Controlling temporal proximity changes (i.e., characteristic mutations), emphasizing key time points, is the eigenvector of the tth column in the matrix, representing the multimodal features at a specific time point t. is the eigenvector of the tth column in the matrix, representing the multimodal features at a specific time point t-1; the denominator is normalized to ensure This weight formula not only considers the feature strength, but also introduces the feature mutation term , specifically optimized for the mutation phenomenon near the titration endpoint; represents the two-norm; Furthermore, based on the time point The feature importance weights For the feature matrix Perform weighting to generate a weighted feature matrix , where the feature matrix is a matrix containing the features of all time points; ; In the weighted feature matrix Extract the time point with the largest feature intensity , and nearby time windows ,in It is the window length, which reflects the key change stage of the titration reaction. Through feature weighting and window extraction, we can focus on the time period when the features change significantly during the titration process, thereby avoiding misjudgment and noise interference.

[0051] Furthermore, in the critical time window In the nonlinear decision model Extract titration endpoint , and according to Calculating titrant volume .

[0052] ; in, is the predicted time of the titration endpoint, given by Calculated based on the changing trend of multimodal features; It is a titrant volume calculation function, combining titration rate and time calculation.

[0053] Among them, the nonlinear judgment model uses support vector machine as a nonlinear judgment model. This nonlinear judgment model uses support vector machine (SVM) to predict the titration endpoint time based on the changing trend of multimodal features. By converting the image and spectral features in the titration process into a high-dimensional feature space, the support vector machine can effectively capture the complex nonlinear relationship between features.

[0054] Specifically, the support vector machine (SVM) combines the multimodal features ( ) is mapped to a high-dimensional feature space to capture the nonlinear feature change pattern during the titration process. By introducing the nonlinear kernel function, SVM can transform the linear inseparable problem in the original feature space into a linear separable situation in the high-dimensional space. In this high-dimensional space, SVM distinguishes the different stages of the titration process by finding an optimal hyperplane, maximizes the distance from each category of data points to the decision boundary, and classifies them through support vectors. SVM performs well in dealing with complex nonlinear relationships, so it can identify the time point when the feature changes most dramatically during the titration process and accurately predict the time of the titration endpoint. In this way, SVM can not only identify the feature differences in each stage of the titration process, but also effectively capture the complex nonlinear relationship between image and spectral features, thereby improving the accuracy of titration endpoint prediction, especially when dealing with multiple factors such as illumination, sample composition and reaction rate in the reaction process, more accurate predictions can be made.

[0055] In the training phase, it is first necessary to collect training data containing multimodal features. Each data point corresponds to a label information (such as the time of the titration endpoint). SVM maps these features to a high-dimensional space and searches for the optimal hyperplane in the space that can distinguish different titration stages or key time points. Through the optimization algorithm, SVM adjusts its kernel function parameters (such as the parameters of the radial basis kernel function) and regularization parameters to ensure that the interval between categories is maximized while avoiding overfitting, thereby improving the generalization ability of the model.

[0056] During the use phase, the trained SVM model will receive new multimodal feature data. These data will be mapped to the same high-dimensional feature space as in the training phase and classified by the hyperplane determined in the model. Based on the decision boundary obtained in the training phase, the SVM will predict the key time points in the titration process, especially the time of the titration endpoint. In this way, the SVM is able to handle complex nonlinear relationships, capture subtle changes between features, and give accurate predictions of the titration endpoint.

[0057] Finally, output the predicted time of the titration endpoint and titrant volume , which is used by the subsequent acid value calculation module. This step accurately extracts the titration endpoint through multimodal feature analysis and dynamic judgment model, solving the complex judgment problem of dynamic changes in the titration reaction.

[0058] S4. Construct a confidence scoring model to perform confidence scoring on the titration endpoint prediction time, determine whether the titration endpoint prediction time needs to be corrected, and recalculate the final titrant volume based on the final endpoint time.

[0059] Specifically, receive the output of step 3: the preliminary titration endpoint prediction time and the corresponding titrant volume , and the comprehensive feature matrix : is the predicted value of the titration endpoint time, obtained by multimodal feature analysis; is based on the titration rate and Calculated titrant volume; is a matrix containing time series features describing the dynamic changes in color and absorbance during the titration process.

[0060] This data was used to further evaluate the reliability of endpoint predictions and make necessary corrections.

[0061] Furthermore, we define a confidence scoring model , used to quantify The core formula is: ; in, for The confidence score of ; Represents the magnitude of multimodal feature changes near the endpoint, which is used to evaluate the significance of the endpoint change; yes Predicting the time at the titration endpoint The specific value of Indicates the rate of change of titrant volume (the calculation formula is shown in step 3), which is used to capture abnormal titration reactions; and are adjustment parameters that control the impact of feature changes and volume changes on confidence.

[0062] The characteristic change term captures the significant features of the titration endpoint, while the volume change term further verifies the credibility of the prediction value through the stationarity of the reaction.

[0063] Furthermore, when Less than the preset threshold When There is a potential error and the correction phase is entered. The correction is done through a nonlinear error correction model To achieve this, the core process combines the time series smoothness of the comprehensive feature matrix: ; in, is the corrected titration endpoint time; is the correction term for the prediction error, given by The time series characteristics are weighted averaged; It is an error correction factor used to balance the magnitude of the correction. Through the smoothing and weighting mechanism, combined with the time distribution of significant features, the error of the initial prediction is corrected to avoid the deviation caused by single-point features.

[0064] Furthermore, according to the corrected endpoint time , recalculate the final titrant volume : ; in It is a flow velocity model and is calculated based on the actual parameters of the experimental equipment.

[0065] Finally, the corrected end time is output and titrant volume , with a confidence score , for use by the subsequent acid value calculation module.

[0066] S5. Calculate the acid value according to the final endpoint time and generate a test report.

[0067] Specifically, receive the output of step 4: End time : Final titration endpoint time after confidence analysis and correction; Titrant volume :correspond The amount of titrant used; Confidence score : To quantify the reliability of endpoint prediction for adjustment of subsequent results.

[0068] Combined with experimentally known parameters: Titrant concentration (mol / L); Sample weight (g); Conversion Factor (56.1, convert the amount of substance into acid number units).

[0069] These data form the basic input for the acid value calculation, ensuring the accuracy of the calculation process.

[0070] Furthermore, the acid value calculation formula is defined, including the basic part and the innovative correction term: ; Part I: , is the basic acid value calculation formula; Correction :Dynamically adjust the acid value results according to the confidence level, and increase the conservative deviation when the confidence level is low to reduce the impact of experimental errors; is the adjustment coefficient, which controls the weight of the confidence correction term on the result.

[0071] Furthermore, the confidence in the titration endpoint prediction directly reflects the endpoint time In the case of low confidence, Enlarging the adjustment range can effectively reduce the error of the measurement results. Different from direct addition, the overall acid value is dynamically adjusted by multiplication here, which can better reflect the global impact of confidence on the experimental results.

[0072] All parameters in the formula maintain the corresponding relationship with experimental data and physical processes, ensuring that the calculation results have practical significance.

[0073] Further, acid value calculation and verification: Calculate the acid value item by item according to the formula: Amount of titrant substance: ; Core acid value calculation: ; Confidence correction: multiply by , dynamically adjust the final result.

[0074] Verification steps: Test low-confidence samples and high-confidence samples separately to check the rationality and accuracy of the correction items on the result distribution.

[0075] Furthermore, the report content is designed as follows: Acid value results , with complete calculation process and input parameters ( wait); Titration process records, including endpoint correction details and confidence scores; Graphical display: titrant volume change curve and acid value calculation curve, intuitively reflecting the titration process and results; Complete data for the report, ensuring that the report contains all experimental input and output data.

[0076] Finally, the output is: Generate exportable PDF reports; Stored in the experiment management system to support subsequent comparison and review; Provides a visual interface where users can view charts and numerical details.

[0077] This step ensures the reliability and adaptability of the measurement results through innovative design and dynamic adjustment of the acid value calculation formula; the experimental report module provides detailed data records and visualization support, providing comprehensive guarantees for experimental analysis and optimization.

[0078] like Figure 2 As shown, in another embodiment of the present invention, an acid value determination system based on AI image recognition is provided, and the system includes: The acid value titration acquisition unit 501 is used to collect data in real time during the acid value titration process, and preprocess the real-time collected data based on time sequence consistency to obtain preprocessed data; wherein the data includes image data and spectral data; Feature extraction unit 502, used to construct a color gradient dynamic extraction model to extract image dynamic features from the preprocessed image data, construct an absorbance change dynamic extraction model to extract spectral dynamic features from the preprocessed spectral data, and synthesize the image dynamic features and spectral dynamic features to generate a comprehensive feature matrix; The titration analysis unit 503 is used to calculate the feature importance weight according to the comprehensive feature matrix to quantify the contribution of the feature mutation point in the key time period, and then weight the comprehensive feature matrix based on the feature importance weight to generate a weighted feature matrix. In the key time window, the titration endpoint is extracted through a nonlinear judgment model based on the weighted feature matrix to obtain a titration endpoint prediction result, and the titrant volume is calculated according to the titration endpoint to obtain the titration endpoint prediction time and titrant volume; The analysis and correction unit 504 is used to construct a confidence scoring model to perform confidence scoring on the titration endpoint prediction time, determine whether the titration endpoint prediction time needs to be corrected, and recalculate the final titrant volume according to the final endpoint time: If the confidence score is less than a preset threshold, it is considered that there is a potential error in the titration endpoint prediction time, and correction is performed, and the corrected endpoint time is output as the final endpoint time. The correction is achieved through a nonlinear error correction model, and the nonlinear error correction model is achieved in combination with the time series smoothness of the weighted feature matrix; If the confidence score is greater than or equal to the preset threshold, it is considered that there is no potential error in the predicted titration endpoint time and it is directly output as the final endpoint time; The report generating unit 505 is used to calculate the acid value according to the final endpoint time and generate an experimental report.

[0079] It should be noted that the workflow described above is merely illustrative and does not limit the scope of protection of the present invention. In practical applications, technicians in this field can select part or all of them according to actual needs to achieve the purpose of the present embodiment, and no limitation is made here.

[0080] In addition, for technical details that are not described in detail in this embodiment, reference can be made to the parameter operation method provided in any embodiment of the present invention, and will not be repeated here.

[0081] It should be noted that the workflow described above is merely illustrative and does not limit the scope of protection of the present invention. In practical applications, technicians in this field can select part or all of them according to actual needs to achieve the purpose of the present embodiment, and no limitation is made here.

[0082] In addition, for technical details that are not described in detail in this embodiment, reference can be made to the parameter operation method provided in any embodiment of the present invention, and will not be repeated here.

[0083] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0084] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0085] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as a read-only memory / random access memory, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0086] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for determining acid value based on AI image recognition, characterized in that: The method comprises: S1. During the acid value titration process, data is collected in real time, and the real-time collected data is preprocessed based on time series consistency to obtain preprocessed data; wherein the data includes image data and spectral data; S2, constructing a color gradient dynamic extraction model to extract image dynamic features from the preprocessed image data, constructing an absorbance change dynamic extraction model to extract spectral dynamic features from the preprocessed spectral data, and combining the image dynamic features and spectral dynamic features to generate a comprehensive feature matrix; S3, calculating the feature importance weight of the feature mutation point according to the comprehensive feature matrix, which is used to quantify the contribution of the feature mutation point in the key time period, and then weighting the comprehensive feature matrix based on the feature importance weight to generate a weighted feature matrix, and within the key time window, extracting the titration endpoint through a nonlinear judgment model based on the weighted feature matrix to obtain a titration endpoint prediction result, and at the same time calculating the titrant volume according to the titration endpoint to obtain the titration endpoint prediction time and titrant volume; S4. Construct a confidence scoring model to score the confidence of the titration endpoint prediction time, determine whether the titration endpoint prediction time needs to be corrected, and recalculate the final titrant volume according to the final endpoint time: If the confidence score is less than a preset threshold, it is considered that there is a potential error in the titration endpoint prediction time, and correction is performed, and the corrected endpoint time is output as the final endpoint time. The correction is achieved through a nonlinear error correction model, and the nonlinear error correction model is achieved in combination with the time series smoothness of the weighted feature matrix; If the confidence score is greater than or equal to the preset threshold, it is considered that there is no potential error in the predicted titration endpoint time and it is directly output as the final endpoint time; S5. Calculate the acid value according to the final endpoint time and generate a test report.

2. The method for determining acid value based on AI image recognition according to claim 1, characterized in that: The preprocessing also includes introducing a dynamic illumination normalization model , for image data To perform adaptive light intensity correction: ； in, is the rectified image; is the image width and height; and are the mean and standard deviation of the light intensity of the image, respectively; is a regularization parameter to prevent over-normalization, is the pixel value of the i-th row and j-th column in the image, i is the row index of the image, and j is the column index of the image; Using Gaussian filtering based on frequency domain To perform noise suppression: ； in, is the preprocessed image data, is the frequency domain component of the image, is the noise control parameter, which is dynamically optimized through experiments; The preprocessing also includes using an adaptive baseline correction model Baseline correction and dynamic feature enhancement of spectral data: ； in, is the corrected spectral data, i.e. the preprocessed spectral data; is the second-order dynamic fitting baseline; It is a dynamic fitting parameter, which is optimized in real time to adapt to different experimental conditions; After correction, the spectral data Extract the dynamic change rate of absorbance : ； in, is the wavelength of the spectral data, For time, is the spectral data after correction.

3. The method for determining acid value based on AI image recognition according to claim 2, characterized in that: The color gradient dynamic extraction model is expressed as: ； in, is the image feature matrix, each column represents the time The eigenvector of is the preprocessed image data; and are convolution kernels and biases, specifically designed to capture color variations; It is a color gradient dynamic extraction model; is the color gradient regularization term, which is used to emphasize the color mutation point; is the regularization coefficient, which controls the weight of the mutation point in the model; is an activation function that retains the salient features of color changes; Dynamic extraction of models through color gradients Process the image frames at each time point , extract the color change sequence matrix : ； Among them, each column is the color feature of a time point, and T is the total number of time points; The absorbance change dynamic extraction model , expressed as: ； in, is the spectral feature matrix, each column represents the time The absorbance characteristics of is the preprocessed spectral data; It represents the rate of change of absorbance over time and is used to capture dynamic changes; It is the second-order characteristic of absorbance changing with wavelength, which is used to identify bands with gradual or sudden changes; is the regularization coefficient, which is used to balance the variation information of time and wavelength; By Model For each time point, the spectral data Processing to generate spectral change feature matrix : ； Each column represents the spectral characteristics at a certain time point.

4. The method for determining acid value based on AI image recognition according to claim 3, characterized in that: The image dynamic features and the spectrum dynamic features are integrated to generate a comprehensive feature matrix, which is expressed as: Introducing modal weighted fusion model ,Will and Fusion into comprehensive feature matrix : ； in, and is a weight matrix, which is dynamically optimized according to the importance of the modal data; is the modality alignment regularization term, which is used to reduce the modality difference between image and spectral features; is a regularization parameter that controls the degree of mode alignment.

5. The method for determining acid value based on AI image recognition according to claim 4, characterized in that: The feature importance weight of the feature mutation point is calculated according to the comprehensive feature matrix, specifically: ； in, A specific time point The feature importance weights of Controlling feature strength The impact of Indicates a specific point in time , represents the specific time point of the current processing; is the loop variable index at the time point; Controlling temporal proximity changes The impact of is the eigenvector of the tth column in the comprehensive feature matrix, representing the multimodal features at a specific time point t. is the eigenvector of the tth column in the comprehensive feature matrix, representing the multimodal features at a specific time point t-1; the denominator is normalized to ensure ; represents the two-norm; Based on a specific point in time The feature importance weights The feature importance weight vector The comprehensive feature matrix Perform weighting to generate a weighted feature matrix , where the comprehensive feature matrix is a matrix containing features at all time points: ； in, is the feature importance weight vector, in the weighted feature matrix Extract the time point with the largest feature intensity , and nearby time windows ,in is the window length, reflecting the key change stage of the titration reaction.

6. The method for determining acid value based on AI image recognition according to claim 5, characterized in that: In the critical time window Internally, based on the weighted feature matrix Through nonlinear judgment model Extraction titration endpoint prediction time , and predict the time based on the titration endpoint Calculating titrant volume : ； in, is the predicted time of the titration endpoint, given by Calculated based on the changing trend of multimodal features; It is a titrant volume calculation function, combining titration rate and time calculation.

7. The method for determining acid value based on AI image recognition according to claim 6, characterized in that: The confidence scoring model is expressed as: ； in, Predicting the time for the titration endpoint The confidence score of ; Indicates the magnitude of multimodal feature changes near the endpoint, which is used to evaluate the significance of the endpoint change. is the weighted multimodal feature vector of the titration endpoint, is the weighted multimodal feature vector at the moment before the titration endpoint, where yes Predicting the time at the titration endpoint The specific value of Indicates the rate of change of titrant volume, used to capture abnormal titration reactions; and are adjustment parameters that control the impact of feature changes and volume changes on confidence respectively; The nonlinear error correction model , expressed as: ； in, is the corrected titration endpoint time; is the correction term for the prediction error, given by The time series features are weighted averaged to obtain: is the weighted multimodal feature vector at time point t; is the error correction factor used to balance the magnitude of the correction.

8. The method for determining acid value based on AI image recognition according to claim 7, characterized in that: The S5 specifically includes: Obtain experimental parameters, including titrant concentration, sample weight, and conversion factor; Calculate the acid value according to the experimental parameters: ； in, For acid value results, part 1 , is the basic acid value calculation formula, is the titrant concentration, is the final titrant volume, F is the conversion factor, is the weight of the sample; correction term Dynamically adjust the acid value results according to the confidence level. When the confidence level is low, increase the conservative deviation to reduce the impact of experimental error. is the adjustment coefficient, which controls the weight of the confidence correction term on the result. () is the confidence score of the titration endpoint prediction; Generate a test report based on the acid value, including: Acid value results ; Titration process records, including endpoint correction details and confidence scores; Graphical display; Complete data for the report, ensuring that the report contains all experimental input and output data.

9. The acid value determination system based on AI image recognition is characterized in that: The system comprises: An acid value titration acquisition unit, used for real-time data acquisition during the acid value titration process, and preprocessing the real-time acquired data based on time sequence consistency to obtain preprocessed data; wherein the data includes image data and spectral data; A feature extraction unit is used to construct a color gradient dynamic extraction model to extract image dynamic features from the preprocessed image data, to construct an absorbance change dynamic extraction model to extract spectral dynamic features from the preprocessed spectral data, and to generate a comprehensive feature matrix by combining the image dynamic features and spectral dynamic features; A titration analysis unit, used for calculating the feature importance weight according to the comprehensive feature matrix to quantify the contribution of the feature mutation point in the key time period, and then weighting the comprehensive feature matrix based on the feature importance weight to generate a weighted feature matrix, and extracting the titration endpoint through a nonlinear judgment model based on the weighted feature matrix within the key time window to obtain a titration endpoint prediction result, and at the same time calculating the titrant volume according to the titration endpoint to obtain the titration endpoint prediction time and titrant volume; The analysis and correction unit is used to construct a confidence scoring model to perform confidence scoring on the titration endpoint prediction time, determine whether the titration endpoint prediction time needs to be corrected, and recalculate the final titrant volume based on the final endpoint time: If the confidence score is less than a preset threshold, it is considered that there is a potential error in the titration endpoint prediction time, and correction is performed, and the corrected endpoint time is output as the final endpoint time. The correction is achieved through a nonlinear error correction model, and the nonlinear error correction model is achieved in combination with the time series smoothness of the weighted feature matrix; If the confidence score is greater than or equal to the preset threshold, it is considered that there is no potential error in the predicted titration endpoint time and it is directly output as the final endpoint time; The report generation unit is used to calculate the acid value according to the final endpoint time and generate an experimental report.

Images