A method for identifying adulterated tea oil based on machine learning algorithm

Abstract

The present invention discloses a method for identifying adulterated tea oil based on a machine learning algorithm, and relates to the technical field of identifying adulterated tea seeds. The method comprises the following steps: obtaining a CSA preparation deviation degree evaluation index based on CSA preparation deviation degree data and analyzing whether to perform color-sensitive sensor preparation adjustment; obtaining a pre-reaction CSA image deviation degree evaluation index based on pre-reaction CSA image deviation degree data and analyzing whether to perform pre-reaction image adjustment; obtaining a post-reaction CSA image deviation degree evaluation index based on post-reaction CSA image deviation degree data and analyzing whether to perform adulterated tea oil identification. The present invention achieves the effect of improving the accuracy of analysis data in the identification of adulterated tea oil by analyzing whether to perform post-reaction image adjustment based on the post-reaction CSA image deviation degree evaluation index and analyzing whether to perform post-reaction image adjustment, and performing adulterated tea oil identification after the post-reaction image adjustment is qualified, thereby solving the problem of inaccurate analysis data in the identification of adulterated tea oil in the prior art.

Description

Technical Field

[0001] The invention relates to the technical field of adulterated tea oil seed identification, and in particular to a method for identifying adulterated tea oil based on a machine learning algorithm. Background Art

[0002] Camellia oil (i.e. tea oil) has high nutritional value and commodity price, but the phenomenon of adulteration and sale of camellia oil in the market is relatively common. There are two common ways of adulterating camellia oil: one is to add other types of low-quality and low-priced edible oils into camellia oil, and the other is to add leached camellia oil with poor processing technology into pressed camellia oil. The categories of adulterated oils are wide and the methods of adulteration are complicated. It is difficult to accurately and quickly identify adulterated edible vegetable oils. The traditional method of adulterating camellia oil mainly relies on fatty acid composition to judge the type of oil, but this one-dimensional identification method has large errors, especially in the detection of mixed oils. There is a bottleneck. Now a method for identifying adulterated camellia oil based on machine learning algorithm has been developed. The data quality is improved by combining machine learning with outlier detection and feature selection algorithm, thereby improving the detection effect of adulterated camellia oil. However, there are still some problems in the use process that lead to inaccurate results of camellia oil identification.

[0003] The existing adulterated tea oil identification system builds an adulteration identification model based on data such as fatty acids, triglycerides and volatile components of adulterated tea seed oil using a machine learning algorithm. The adulteration identification model is used to identify whether the tea oil is adulterated, thereby realizing the identification function of adulterated tea oil.

[0004] For example, the invention patent with the announcement number CN113884594B discloses a method for identifying adulterated tea oil based on a machine learning algorithm, comprising: S1, obtaining tea seed oil by cold pressing; S2, measuring the fatty acid and triglyceride content contained in the tea seed oil obtained in S1; S3, measuring the fatty acid and triglyceride content contained in other types of adulterated oils; S4, designing an adulterated tea seed oil model; S5, data preprocessing and programming platform; S6, reading in the data of adulterated tea seed oil mixed with various types of vegetable oils, and Types are classified into categories; S7, data standardization; S8, the entire data set is divided, 20% of the data are randomly selected from the data set as the test set, and the remaining 80% of the data are used as the training set; S9, the model is trained and validated according to the 5-fold cross-validation method; S10, the decision tree classification model for visual display is trained using the entire training set, and the decision tree model is visualized using the tree.export_graphviz function of the scikit-learn library in Python.

[0005] For example, the invention patent with announcement number: CN113884594B announces a method for detecting tea oil adulteration, including: S1. Identification of the type of adulteration of tea oil: S11. Establishment of an adulteration type model; S12. Identification of the adulteration type; S2. Determination of the mixing amount of tea oil adulteration: S21. Establishment of an adulteration amount model; S22. Determination of the adulteration amount.

[0006] However, in the process of implementing the technical solution of the invention in the embodiments of the present application, the present application found that the above technology has at least the following technical problems:

[0007] In the prior art, the process of identifying adulterated tea oil involves many steps. When an error occurs in a step, the subsequent steps will be affected, which will ultimately lead to the accuracy of identifying adulterated tea oil. There is a problem of inaccurate analysis data in the identification of adulterated tea oil. Summary of the invention

[0008] The embodiment of the present application solves the problem of inaccurate analysis data in the identification of adulterated tea oil in the prior art by providing a method for identifying adulterated tea oil based on a machine learning algorithm, and realizes the identification of adulterated tea oil based on accurate data.

[0009] The embodiment of the present application provides a method for identifying adulterated tea oil based on a machine learning algorithm, comprising the following steps: obtaining CSA preparation deviation degree data, obtaining a CSA preparation deviation degree evaluation index according to the CSA preparation deviation degree data, and analyzing whether to perform color-sensitive sensor preparation adjustment; obtaining pre-reaction CSA image deviation degree data after the color-sensitive sensor preparation adjustment is qualified, obtaining a pre-reaction CSA image deviation degree evaluation index according to the pre-reaction CSA image deviation degree data, and analyzing whether to perform pre-reaction image adjustment; obtaining post-reaction CSA image deviation degree data after the pre-reaction image adjustment is qualified, obtaining a post-reaction CSA image deviation degree evaluation index according to the post-reaction CSA image deviation degree data, and analyzing whether to identify adulterated tea oil; the CSA preparation deviation degree data includes: pH indicator solution concentration, solution ultrasonic treatment time, NaOH solution usage, CSA ventilation time, pH indicator solution concentration standard value, solution ultrasonic treatment time standard value, NaOH solution usage standard value, and CSA ventilation time standard value; the CSA preparation deviation degree evaluation index is used to quantify the deviation degree of CSA preparation.

[0010] Furthermore, the specific method for obtaining the CSA preparation deviation degree evaluation index is as follows: the result of the difference operation between the pH indicator solution concentration and the pH indicator solution concentration standard value is processed with the pH indicator solution concentration standard value to obtain the pH indicator solution concentration score; the result of the ratio operation between the solution ultrasonic treatment time and the solution ultrasonic treatment time standard value is processed to obtain the treatment time score; the result of the ratio operation between the sum of the NaOH solution usage amount and the NaOH solution usage standard value and twice the NaOH solution usage standard value is processed to obtain the usage score; the result of the ratio operation between the CSA ventilation time and the CSA ventilation time standard value is processed to obtain the ventilation time score; and the CSA preparation deviation degree evaluation index is obtained according to the pH indicator solution concentration score, the treatment time score, the usage score and the ventilation time score.

[0011] Furthermore, the specific process of analyzing whether to perform color-sensitive sensor preparation adjustment is: comparing the CSA preparation deviation degree evaluation index with the CSA preparation deviation degree threshold preset in the database; when the CSA preparation deviation degree evaluation index is less than the CSA preparation deviation degree threshold preset in the database, no color-sensitive sensor preparation adjustment is performed; when the CSA preparation deviation degree evaluation index is not less than the CSA preparation deviation degree threshold preset in the database, reminding the preset personnel to perform color-sensitive sensor preparation adjustment, when the monitored CSA preparation deviation degree evaluation index is less than the preset CSA preparation deviation degree threshold, stopping the color-sensitive sensor preparation adjustment; the color-sensitive sensor preparation adjustment includes reconfiguring the color-sensitive sensor and adjusting the configuration equipment.

[0012] Furthermore, the pre-reaction CSA image deviation degree data includes: a CSA preparation deviation degree evaluation index of a qualified color-sensitive sensor preparation and adjustment, a pre-reaction CSA light intensity, a pre-reaction device lens transmittance, a pre-reaction CSA image signal-to-noise ratio, a pre-reaction CSA image chromatic aberration, a pre-reaction CSA light intensity reference value, a pre-reaction device lens transmittance reference value, a pre-reaction CSA image signal-to-noise ratio reference value, and a pre-reaction CSA image chromatic aberration reference value; the pre-reaction CSA image deviation degree evaluation index is used to quantify the deviation degree of the pre-reaction CSA image.

[0013] Furthermore, the specific method for obtaining the pre-reaction CSA image deviation degree evaluation index is as follows: the front illumination intensity score is obtained by processing the pre-reaction CSA illumination intensity and the pre-reaction CSA illumination intensity reference value; the front transmittance score is obtained by processing the pre-reaction device lens transmittance and the pre-reaction device lens transmittance reference value; the front signal-to-noise ratio score is obtained by processing the pre-reaction CSA image signal-to-noise ratio and the pre-reaction CSA image signal-to-noise ratio reference value; the front color difference score is obtained by processing the pre-reaction CSA image chromatic aberration and the pre-reaction CSA image chromatic aberration reference value; and the pre-reaction CSA image deviation degree evaluation index is obtained according to the qualified CSA preparation deviation degree evaluation index, the front illumination intensity score, the front transmittance score, the front signal-to-noise ratio score and the front color difference score of the color-sensitive sensor preparation adjustment.

[0014] Furthermore, the specific process of analyzing whether to perform pre-reaction image adjustment is as follows: comparing the pre-reaction CSA image deviation degree evaluation index with the pre-reaction image deviation threshold preset in the database; when the pre-reaction CSA image deviation degree evaluation index is less than the pre-reaction image deviation threshold preset in the database, no pre-reaction image adjustment is performed; when the pre-reaction CSA image deviation degree evaluation index is not less than the pre-reaction image deviation threshold preset in the database, pre-reaction image adjustment is performed, and the pre-reaction image adjustment includes adjusting the image acquisition illumination intensity and adjusting the lens cleanliness; when the monitored pre-reaction CSA image deviation degree evaluation index is less than the pre-reaction image deviation threshold preset in the database, the pre-reaction image adjustment is stopped.

[0015] Furthermore, the post-reaction CSA image deviation degree data includes: a CSA preparation deviation degree evaluation index for qualified color sensor preparation and adjustment, a pre-reaction CSA image deviation degree evaluation index for qualified pre-reaction image adjustment, a CSA preparation deviation degree threshold, a pre-reaction image deviation threshold, environmental data and image data; the environmental data includes post-reaction CSA light intensity, post-reaction device lens transmittance, post-reaction CSA light intensity reference value and post-reaction device lens transmittance reference value; the image data includes post-reaction CSA image clarity index, post-reaction CSA image signal-to-noise ratio, post-reaction CSA image chromatic aberration, post-reaction CSA image clarity index reference value, post-reaction CSA image signal-to-noise ratio reference value and post-reaction CSA image chromatic aberration reference value; the post-reaction CSA image deviation degree evaluation index is used to reflect the deviation degree of the post-reaction CSA image.

[0016] Furthermore, the specific method for obtaining the evaluation index of the degree of deviation of the post-reaction CSA image is as follows: the result of the difference operation between the post-reaction CSA illumination intensity and the post-reaction CSA illumination intensity reference value is processed with the post-reaction CSA illumination intensity reference value to obtain the post-illumination intensity score; the result of the ratio operation between the post-reaction device lens transmittance and the post-reaction device lens transmittance reference value is processed to obtain the post-transmittance score; the result of the ratio operation between the post-reaction CSA image clarity index and the post-reaction CSA image clarity index reference value is processed to obtain the post-clarity score; the result of the ratio operation between the sum of the post-reaction CSA image signal-to-noise ratio and the post-reaction CSA image signal-to-noise ratio reference value and the post-reaction CSA image signal-to-noise ratio reference value is processed to obtain the post-signal-to-noise ratio score; the result of the ratio operation between the post-reaction CSA image color difference and the post-reaction CSA image color difference reference value is processed to obtain the post-color difference score; the qualified CSA preparation deviation degree evaluation index adjusted according to the color sensitive sensor preparation The invention discloses a method for analyzing whether to identify adulterated tea oil by using the following methods: the method comprises the following steps: obtaining a qualified pre-reaction CSA image deviation degree evaluation index before reaction image adjustment, a CSA preparation deviation degree threshold, a pre-reaction image deviation threshold, a post-illumination intensity score, a post-transmittance score, a post-clarity score, a post-signal-to-noise ratio score and a post-color difference score to obtain a post-reaction CSA image deviation degree evaluation index after reaction; the specific process of analyzing whether to identify adulterated tea oil is as follows: comparing the post-reaction CSA image deviation degree evaluation index with the post-reaction image deviation threshold in a database; when the post-reaction CSA image deviation degree evaluation index is less than the post-reaction image deviation threshold in a database, identifying adulterated tea oil; when the post-reaction CSA image deviation degree evaluation index is not less than the post-reaction image deviation threshold in a database, identifying adulterated tea oil, and the post-reaction image adjustment comprises adjusting the image acquisition light intensity and adjusting the image acquisition settings; when the monitored post-reaction CSA image deviation degree evaluation index is less than the post-reaction image deviation threshold in a database, stopping the post-reaction image adjustment.

[0017] Furthermore, the specific limiting expression of the post-reaction CSA image deviation degree evaluation index is:

[0018]

[0019] In the formula, It is expressed as the evaluation index of the deviation degree of the CSA image after the reaction, which is qualified for the pre-reaction image adjustment. It is an evaluation index of the deviation degree of the pre-reaction CSA image after the pre-reaction image adjustment is qualified. Denoted as the pre-reaction image deviation threshold, It represents the evaluation index of the deviation degree of CSA preparation for color sensor preparation adjustment. It is expressed as the CSA preparation deviation threshold, Expressed as the fraction of post-illumination intensity, Expressed as the rear transmittance fraction, Expressed as the post-clarity score, Expressed as the posterior signal-to-noise ratio fraction, Expressed as the post-chromatic aberration score.

[0020] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0021] 1. The CSA preparation deviation degree evaluation index is obtained through the CSA preparation deviation degree data to analyze whether to adjust the color-sensitive sensor preparation. After the adjustment is qualified, the pre-reaction image adjustment is analyzed according to the pre-reaction CSA image deviation degree data. After the adjustment is qualified, the deviation degree of the post-reaction CSA image is quantified according to the post-reaction CSA image deviation degree data and it is analyzed whether to identify the adulterated tea oil, thereby improving the accuracy of the analysis data in the identification of adulterated tea oil, and effectively solving the problem of inaccurate analysis data in the identification of adulterated tea oil in the prior art.

[0022] 2. By obtaining the front illumination intensity score, front transmittance score, front signal-to-noise ratio score and front chromatic aberration score according to the pre-reaction CSA image deviation degree data, the pre-reaction CSA image deviation degree evaluation index is obtained to quantify the deviation degree of the pre-reaction CSA image, thereby realizing accurate pre-reaction image adjustment according to the pre-reaction CSA image deviation degree evaluation index.

[0023] 3. By obtaining the post-illumination intensity score, post-transmittance score, post-clarity score, post-signal-to-noise ratio score and post-color difference score according to the post-reaction CSA image deviation degree data, the post-reaction CSA image deviation degree evaluation index is obtained to quantify the post-reaction CSA image deviation degree, thereby realizing accurate post-reaction image adjustment according to the post-reaction CSA image deviation degree evaluation index. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A flow chart of a method for identifying adulterated tea oil based on a machine learning algorithm provided in an embodiment of the present application;

[0025] Figure 2 This is a graph of the CSA preparation deviation evaluation index provided in the embodiment of the present application. DETAILED DESCRIPTION

[0026] The embodiment of the present application solves the problem of inaccurate analysis data in the identification of adulterated tea oil in the prior art by providing a method for identifying adulterated tea oil based on a machine learning algorithm, by obtaining a CSA preparation deviation degree evaluation index based on the CSA preparation deviation degree data and analyzing whether to perform color-sensitive sensor preparation adjustment, and obtaining a pre-reaction CSA image deviation degree evaluation index based on the pre-reaction CSA image deviation degree data after the color-sensitive sensor preparation adjustment is qualified, and analyzing whether to perform pre-reaction image adjustment, and thus obtaining a post-reaction CSA image deviation degree evaluation index based on the post-reaction CSA image deviation degree data after the pre-reaction image adjustment is qualified, and analyzing whether to perform adulterated tea oil identification, thereby improving the accuracy of the analysis data in the identification of adulterated tea oil.

[0027] The technical solution in the embodiment of the present application is to solve the above-mentioned problem of inaccurate analysis data in the identification of adulterated tea oil. The overall idea is as follows:

[0028] By acquiring CSA preparation deviation degree data, obtaining CSA preparation deviation degree evaluation indicators according to the CSA preparation deviation degree data and analyzing whether to perform color-sensitive sensor preparation adjustment, then acquiring pre-reaction CSA image deviation degree data after the color-sensitive sensor preparation adjustment is qualified, then obtaining pre-reaction CSA image deviation degree evaluation indicators according to the pre-reaction CSA image deviation degree data and analyzing whether to perform pre-reaction image adjustment, then acquiring post-reaction CSA image deviation degree data after the pre-reaction image adjustment is qualified, then obtaining post-reaction CSA image deviation degree evaluation indicators according to the post-reaction CSA image deviation degree data and analyzing whether to perform post-reaction image adjustment, thereby performing adulterated tea oil identification after the post-reaction image adjustment is qualified, thereby achieving the effect of improving the accuracy of analysis data in the identification of adulterated tea oil.

[0029] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0030] like Figure 1As shown, it is a flow chart of a method for identifying adulterated tea oil based on a machine learning algorithm provided in an embodiment of the present application, the method comprising the following steps: obtaining CSA preparation deviation degree data, obtaining a CSA preparation deviation degree evaluation index according to the CSA preparation deviation degree data, and analyzing whether to perform color-sensitive sensor preparation adjustment; obtaining pre-reaction CSA image deviation degree data after the color-sensitive sensor preparation adjustment is qualified, obtaining a pre-reaction CSA image deviation degree evaluation index according to the pre-reaction CSA image deviation degree data, and analyzing whether to perform pre-reaction image adjustment; obtaining post-reaction CSA image deviation degree data after the pre-reaction image adjustment is qualified, obtaining a post-reaction CSA image deviation degree evaluation index according to the post-reaction CSA image deviation degree data, and analyzing whether to perform adulterated tea oil identification; CSA preparation deviation degree data include: pH indicator solution concentration, solution ultrasonic treatment time, NaOH solution usage, CSA ventilation time, pH indicator solution concentration standard value, solution ultrasonic treatment time standard value, NaOH solution usage standard value, and CSA ventilation time standard value; CSA preparation deviation degree evaluation index is used to quantify the deviation degree of CSA preparation.

[0031] In a specific embodiment, the specific process of identifying adulterated tea oil is as follows:

[0032] A total of 192 oil samples were prepared, including 144 adulterated camellia oil samples and 48 pure oil samples. The adulterated samples were rapeseed oil, soybean oil and corn oil mixed with camellia oil at different ratios of 20%, 40%, 60% and 80%, with 12 samples at each ratio. The pure oil samples included pure camellia oil, pure rapeseed oil, pure soybean oil and pure corn oil, with 12 samples each.

[0033] Four porphyrin compounds (methylphenylporphyrin cobalt, methoxyphenylporphyrin cobalt, porphyrin manganese and porphyrin iron) and five pH indicators (bromocresol green, methyl red, cresol red, bromophenol blue and neutral red) were selected as color-sensitive materials and dissolved in dichloromethane and anhydrous ethanol to form a 2 mg / mL solution. Ultrasonic treatment was performed for 15 minutes to ensure complete dissolution, and NaOH was added to the pH indicator solution to adjust the pH value. Using PVDF membrane as the substrate, the color-sensitive material was drop-coated on a 3 cm × 3 cm PVDF membrane by a capillary dropper to prepare a 3 × 3 colorimetric sensor array (CSA). The prepared CSA was placed in a fume hood for 20 minutes to allow the solvent to evaporate and then used for detection.

[0034] In a darkroom environment, an industrial camera was used to record the initial image of CSA before reaction. The prepared oil sample was placed in a petri dish and reacted at 100°C for 20 minutes. The image of CSA was taken immediately after the reaction. The images before and after the reaction were uploaded to a computer, and the image data was processed using the C programming language to generate an RGB difference image, thereby obtaining 27-dimensional color features. This step was repeated to build a complete data set.

[0035] The data set is randomly divided into 30% as the test set and 70% as the training set. The training set is used to train the prediction model, and the test set is used to evaluate the model performance.

[0036] The ECOD algorithm (Unsupervised Outlier Detection Using Empirical Cumulative Distribution Functions) is used to remove outliers in the data set, and then the PSO algorithm is used to remove redundant information to obtain an optimized data set.

[0037] The optimized data set was input into the support vector machine (SVM) prediction model for training to build a model that can accurately detect adulterated oil samples.

[0038] The specific process of obtaining the evaluation index of the degree of deviation of the CSA image after reaction and analyzing whether to identify the adulterated tea oil is as follows: analyzing whether the post-reaction image adjustment is required according to the evaluation index of the degree of deviation of the CSA image after reaction, and identifying the adulterated tea oil when the post-reaction image adjustment is qualified, uploading the post-reaction CSA image and the pre-reaction CSA image to the computer, processing the image data using the C programming language, generating an RGB difference image, and obtaining a 27-dimensional color feature. Repeat this step to construct a complete data set. And continue the remaining steps of the specific process of identifying the type of adulterated oil.

[0039] Specifically, the CSA (Color Sensitive Sensor Array) preparation deviation degree data is used to analyze the CSA preparation deviation degree evaluation index; the pH indicator solution concentration indicates the concentration of the configured pH indicator solution; the solution ultrasonic treatment time indicates the time for the pH indicator solution to be ultrasonically treated; the NaOH solution usage indicates the usage of the NaOH solution; the CSA ventilation time indicates the time for the CSA to be ventilated in the fume hood; the pH indicator solution concentration standard value indicates the concentration of the pH indicator solution configured when the CSA preparation is qualified; the solution ultrasonic treatment time standard value indicates the time for the pH indicator solution to be ultrasonically treated when the CSA preparation is qualified; the NaOH solution usage standard value indicates the usage of the NaOH solution when the CSA preparation is qualified; the CSA ventilation time standard value indicates the time for the CSA to be ventilated in the fume hood when the CSA preparation is qualified.

[0040] In this embodiment, the density of the pH indicator solution is measured by a pycnometer or a densitometer, and the concentration of the pH indicator solution is obtained by combining the density of the known solute with the relationship between density and concentration. The timer is used to record the ultrasonic treatment time of the solution subjected to ultrasonic treatment of the pH indicator solution. The amount of NaOH solution used is accurately measured or titrated using a pipette or a burette. The timer is used to record the CSA ventilation time of the CSA ventilation.

[0041] The pH indicator solution concentration, solution ultrasonic treatment time, NaOH solution usage and CSA ventilation time of historical CSA preparation qualifications were obtained from the database and averaged to obtain the standard value of pH indicator solution concentration, solution ultrasonic treatment time, NaOH solution usage and CSA ventilation time.

[0042] The CSA preparation deviation degree evaluation index was obtained by analyzing the pH indicator solution concentration, solution ultrasonic treatment time, NaOH solution usage and CSA ventilation time specified in the preparation of CSA. The deviation degree of CSA preparation was analyzed according to the CSA preparation deviation degree evaluation index, which achieved accurate and reliable analysis of the deviation degree of CSA preparation, which was conducive to more accurate preparation adjustment of color-sensitive sensors.

[0043] Furthermore, the specific method for obtaining the CSA preparation deviation evaluation index is as follows: the result of the difference operation between the pH indicator solution concentration and the pH indicator solution concentration standard value is processed with the pH indicator solution concentration standard value to obtain the pH indicator solution concentration score (i.e. ); the result of the ratio operation of the solution ultrasonic treatment time and the standard value of the solution ultrasonic treatment time is processed to obtain the treatment time fraction (i.e. ); The result of the ratio calculation of the sum of the amount of NaOH solution used and the standard value of the amount of NaOH solution used and twice the standard value of the amount of NaOH solution used is processed to obtain the usage score (i.e. ); The result of the ratio calculation of CSA ventilation time and CSA ventilation time standard value is processed to obtain the ventilation time score (i.e. ); The CSA preparation deviation degree evaluation index was obtained according to the pH indicator solution concentration score, treatment time score, usage score and ventilation time score.

[0044] Specifically, the specific limiting expression of the CSA preparation deviation evaluation index is:

[0045]

[0046] In the formula, It is an evaluation index of the CSA preparation deviation degree when preparing color-sensitive sensors. It is expressed as the concentration of pH indicator solution when preparing the color sensitive sensor. It is the ultrasonic treatment time of the solution during the preparation of the color-sensitive sensor. It represents the amount of NaOH solution used in the preparation of color sensitive sensor. It is the CSA ventilation time when the color sensitive sensor is prepared. It is the standard value of pH indicator solution concentration when preparing color sensitive sensor. It is the standard value of the ultrasonic treatment time of the solution when preparing the color-sensitive sensor. It represents the standard value of the amount of NaOH solution used in the preparation of color-sensitive sensors. It is the standard value of CSA ventilation time when preparing color sensitive sensors. Expressed as pH indicator solution concentration fraction, Expressed as a fraction of processing time, Expressed as a usage fraction, Expressed as ventilation duration fraction.

[0047] In this embodiment, the algorithm of this embodiment is combined with the CSA preparation deviation degree data for comprehensive analysis to obtain the CSA preparation deviation degree evaluation index. In the algorithm of this embodiment, the CSA preparation deviation degree data is not independent, and there is a correlation between them. If the concentration of the pH indicator solution is too high, the reaction rate may be too fast, making the reaction difficult to control, resulting in a large deviation of the pH value of the pH indicator solution, thereby increasing the deviation of the CSA preparation. On the contrary, if the concentration of the pH indicator solution is too low, although the reaction rate will slow down, it may take longer to complete the reaction, which may also cause preparation deviation, because more interference factors may be introduced during the long reaction process. Appropriate ultrasonic treatment time promotes the mixing and uniformity of the pH indicator solution, thereby reducing the preparation deviation. However, too long ultrasonic treatment may cause excessive changes in the properties of the pH indicator solution, and even destroy the structure of the color-sensitive material, thereby increasing the preparation deviation. The appropriate amount of NaOH solution used can accurately adjust the pH value of the solution. When the amount of NaOH solution used is too much or insufficient, the pH value of the solution will be far from the target pH value. Appropriate ventilation helps to keep the working environment safe and clean, thereby reducing the preparation deviation. However, if the ventilation time is too long, it may cause changes in temperature, humidity and other conditions in the working environment.

[0048] Therefore, when the difference between the pH indicator solution concentration, the solution ultrasonic treatment time, the NaOH solution usage and the CSA ventilation time and the standard value of the pH indicator solution concentration, the standard value of the solution ultrasonic treatment time, the standard value of the NaOH solution usage and the CSA ventilation time respectively is smaller, that is, the smaller the pH indicator solution concentration score, the treatment time score, the usage score and the ventilation time score are, the smaller the corresponding CSA preparation deviation degree evaluation index is, the smaller the CSA preparation deviation is, and the less need is made to adjust the color-sensitive sensor preparation; conversely, the larger the CSA preparation deviation degree evaluation index is, the larger the CSA preparation deviation is, and the more need is made to adjust the color-sensitive sensor preparation.

[0049] The algorithm of this embodiment combines the pH indicator solution concentration, the solution ultrasonic treatment time, the NaOH solution usage and the CSA ventilation time to obtain the CSA preparation deviation degree evaluation index and perform color-sensitive sensor preparation adjustment, thereby achieving more accurate color-sensitive sensor preparation adjustment.

[0050] like Figure 2As shown, it is a CSA preparation deviation evaluation index curve provided in an embodiment of the present application, with the pH indicator solution concentration fraction as the independent variable and the CSA preparation deviation evaluation index as the dependent variable. Through the increase and decrease analysis of curves 1, 2 and 3, it can be seen that when the pH indicator solution concentration fraction is larger, the CSA preparation deviation evaluation index is larger; when the pH indicator solution concentration fraction is smaller, the CSA preparation deviation evaluation index is smaller. The statistical table of relevant data of the CSA preparation deviation evaluation index is shown in Table 1:

[0051]

[0052] When the parameters of curve 1, curve 2 and curve 3 are as shown in the above table, the CSA preparation deviation evaluation indexes of curve 1, curve 2 and curve 3 are 0.207, 0.320 and 0.441 respectively. It can be seen that the greater the pH indicator solution concentration score, processing time score, usage score and ventilation time score, the greater the CSA preparation deviation evaluation index, and the more adjustment is needed for the preparation of the color-sensitive sensor to make the identification of adulterated tea oil more accurate.

[0053] Furthermore, the specific process of analyzing whether to adjust the preparation of the color-sensitive sensor is as follows: comparing the CSA preparation deviation degree evaluation index with the CSA preparation deviation degree threshold preset in the database; when the CSA preparation deviation degree evaluation index is less than the CSA preparation deviation degree threshold preset in the database, no color-sensitive sensor preparation adjustment is performed; when the CSA preparation deviation degree evaluation index is not less than the CSA preparation deviation degree threshold preset in the database, the preset personnel are reminded to adjust the preparation of the color-sensitive sensor, and when the CSA preparation deviation degree evaluation index monitored after the color-sensitive sensor preparation adjustment is performed is less than the preset CSA preparation deviation degree threshold, the color-sensitive sensor preparation adjustment is stopped; the color-sensitive sensor preparation adjustment includes reconfiguring the color-sensitive sensor and adjusting the configuration equipment.

[0054] In this embodiment, the preparation and adjustment of the color-sensitive sensor includes reconfiguring the color-sensitive sensor and adjusting the configuration equipment. The reason for the inaccurate configuration of the color-sensitive sensor may be that the configuration equipment is not accurate enough. Therefore, adjusting the configuration equipment includes using a more accurate pycnometer or densitometer to measure the density of the solution, using a more accurate timer to record the time of ultrasonic treatment, using a more accurate pipette or burette to accurately measure or titrate the volume of NaOH solution, and using a more accurate timer to record the ventilation time. The color-sensitive sensor is reconfigured because the configured CAS has a large deviation and cannot be used for subsequent identification of adulterated tea oil, so the CSA needs to be reconfigured.

[0055] Furthermore, the pre-reaction CSA image deviation degree data includes: a CSA preparation deviation degree evaluation index of a qualified color-sensitive sensor preparation and adjustment, a pre-reaction CSA illumination intensity, a pre-reaction device lens transmittance, a pre-reaction CSA image signal-to-noise ratio, a pre-reaction CSA image chromatic aberration, a pre-reaction CSA illumination intensity reference value, a pre-reaction device lens transmittance reference value, a pre-reaction CSA image signal-to-noise ratio reference value, and a pre-reaction CSA image chromatic aberration reference value; the pre-reaction CSA image deviation degree evaluation index is used to quantify the deviation degree of the pre-reaction CSA image.

[0056] In this embodiment, the pre-reaction CSA image deviation degree data is used to analyze and obtain a pre-reaction CSA image deviation degree evaluation index; the pre-reaction CSA illumination intensity represents the illumination intensity when the pre-reaction CSA image is acquired; the pre-reaction device lens transmittance represents the transmittance of the device lens for acquiring the CSA image before the reaction; the pre-reaction CSA image signal-to-noise ratio represents the signal-to-noise ratio of the CSA image acquired before the reaction; the pre-reaction CSA image chromatic aberration represents the chromatic aberration of the CSA image acquired before the reaction; the pre-reaction CSA illumination intensity reference value represents the illumination intensity of the environment corresponding to when the pre-reaction CSA image is qualified; the pre-reaction device lens transmittance reference value represents the transmittance corresponding to the device lens for acquiring the CSA image when the pre-reaction CSA image is qualified; the pre-reaction CSA image signal-to-noise ratio reference value represents the signal-to-noise ratio of the acquired CSA image when the pre-reaction CSA image is qualified; the pre-reaction CSA image chromatic aberration reference value represents the chromatic aberration of the acquired CSA image when the pre-reaction CSA image is qualified.

[0057] The light intensity of the environment is measured by a light intensity sensor or photometer when the pre-reaction CSA image is acquired, which is the CSA light intensity before the reaction. The spectral transmittance of the lens of the device that acquires the CSA image is measured by a spectral transmittance tester, which is the light transmittance of the lens of the device before the reaction. The signal-to-noise ratio of the pre-reaction CSA image is calculated by image processing software, which is the signal-to-noise ratio of the pre-reaction CSA image. The color difference of the image is obtained by color analysis of the pre-reaction CSA image by a color measuring instrument or software, which is the color difference of the pre-reaction CSA image.

[0058] The illumination intensity, equipment lens transmittance, image signal-to-noise ratio and image chromatic aberration of the historical pre-reaction CSA image when the deviation degree was qualified were obtained from the database, and the mean values ​​were processed to obtain the pre-reaction CSA illumination intensity reference value, the pre-reaction equipment lens transmittance reference value, the pre-reaction CSA image signal-to-noise ratio reference value and the pre-reaction CSA image chromatic aberration reference value.

[0059] By comprehensively analyzing the CSA preparation deviation degree evaluation index of the color-sensitive sensor preparation and adjustment that is qualified, the CSA light intensity before reaction, the equipment lens transmittance before reaction, the CSA image signal-to-noise ratio before reaction, and the CSA image chromatic aberration before reaction, the CSA image deviation degree evaluation index before reaction is obtained, making the CSA image deviation degree evaluation index before reaction more comprehensive.

[0060] Furthermore, the specific method for obtaining the pre-reaction CSA image deviation evaluation index is as follows: the pre-reaction CSA illumination intensity and the pre-reaction CSA illumination intensity reference value are processed to obtain the pre-reaction illumination intensity score (i.e. ); The front transmittance score (i.e. ); The pre-reaction CSA image signal-to-noise ratio score (i.e. ); The pre-reaction color difference score (i.e. ); According to the qualified CSA preparation deviation degree evaluation index, the front illumination intensity score, the front transmittance score, the front signal-to-noise ratio score and the front color difference score of the color-sensitive sensor preparation adjustment, the front CSA image deviation degree evaluation index is obtained.

[0061] Specifically, the specific limiting expression of the pre-reaction CSA image deviation degree evaluation index is:

[0062]

[0063] In the formula, It is expressed as the evaluation index of the degree of CSA image deviation before the color sensor reacts. It represents the evaluation index of the deviation degree of CSA preparation for color sensor preparation adjustment. It is expressed as the CSA preparation deviation threshold, It is expressed as the CSA light intensity before the color sensor reacts. It is expressed as the light transmittance of the device lens before the color sensor reacts. It is expressed as the signal-to-noise ratio of the CSA image before the reaction of the color-sensitive sensor, It is expressed as the CSA image color difference before the color sensor reacts, It is expressed as the CSA light intensity reference value before the color sensor reacts. It is expressed as the reference value of the light transmittance of the device lens before the color sensor reacts. It is expressed as the reference value of the CSA image signal-to-noise ratio before the color sensor reacts. It is expressed as the CSA image color difference reference value before the color sensor reacts. Expressed as the front light intensity fraction, Expressed as the front transmittance fraction, Expressed as the front signal-to-noise ratio fraction, Expressed as the front color difference score.

[0064] In this embodiment, the algorithm of this embodiment combines the pre-reaction CSA image deviation degree data for comprehensive analysis to obtain the pre-reaction CSA image deviation degree evaluation index. The pre-reaction CSA image deviation degree data in the algorithm of this embodiment does not exist independently, but is interrelated. When the CSA preparation deviation degree evaluation index of the color-sensitive sensor preparation and adjustment is qualified, the CSA preparation is more accurate, and the pre-reaction CSA image signal-to-noise ratio and pre-reaction CSA image color difference obtained according to the pre-reaction CSA image are more accurate; within an appropriate range, the light intensity is positively correlated with the image accuracy, that is, when the light intensity is moderate, the image accuracy is high; when the light is too strong or too weak, the image accuracy may be reduced. When the transmittance of the lens is smaller, the image accuracy is lower, which may lead to a smaller image signal-to-noise ratio and worse image quality; when the image accuracy is lower and the image quality is worse, the image color difference may be greater.

[0065] When the pre-reaction CSA illumination intensity, pre-reaction device lens transmittance, pre-reaction CSA image signal-to-noise ratio, and pre-reaction CSA image chromatic aberration deviate more from the pre-reaction CSA illumination intensity reference value, pre-reaction device lens transmittance reference value, pre-reaction CSA image signal-to-noise ratio reference value, and pre-reaction CSA image chromatic aberration reference value, respectively, the front illumination intensity score, the front transmittance score, the front signal-to-noise ratio score, and the front chromatic aberration score will be larger, and thus the pre-reaction CSA image deviation degree evaluation index will be larger. At this time, the greater the deviation degree of the pre-reaction CSA image, the more necessary it is to adjust the pre-reaction image; and vice versa.

[0066] The algorithm of this embodiment obtains a pre-reaction CSA image deviation degree evaluation index by performing nonlinear transformation on the comprehensive pre-reaction CSA illumination intensity, the pre-reaction device lens transmittance, the pre-reaction CSA image signal-to-noise ratio and the pre-reaction CSA image chromatic aberration, so that the pre-reaction CSA image deviation degree evaluation index has higher reliability, which is conducive to more accurate pre-reaction image adjustment according to the pre-reaction CSA image deviation degree evaluation index.

[0067] Furthermore, the specific process of analyzing whether to perform pre-reaction image adjustment is as follows: comparing the pre-reaction CSA image deviation degree evaluation index with the pre-reaction image deviation threshold preset in the database; when the pre-reaction CSA image deviation degree evaluation index is less than the pre-reaction image deviation threshold preset in the database, no pre-reaction image adjustment is performed; when the pre-reaction CSA image deviation degree evaluation index is not less than the pre-reaction image deviation threshold preset in the database, pre-reaction image adjustment is performed, and the pre-reaction image adjustment includes adjusting the image acquisition illumination intensity and adjusting the lens cleanliness; when the pre-reaction CSA image deviation degree evaluation index monitored after the pre-reaction image adjustment is less than the pre-reaction image deviation threshold preset in the database, the pre-reaction image adjustment is stopped.

[0068] In this embodiment, the pre-reaction image adjustment includes adjusting the illumination intensity for image acquisition and adjusting the cleanliness of the lens. Adjusting the illumination intensity for image acquisition refers to adjusting step by step (e.g., adjusting by 5% each time) the illumination intensity of the environment for acquiring the CSA image. Adjusting the cleanliness of the lens refers to wiping the lens clean. When the monitored pre-reaction CSA image deviation degree evaluation index is less than the pre-reaction image deviation threshold value preset in the database, the pre-reaction image adjustment is stopped. If the pre-reaction CSA image deviation degree evaluation index is still not less than the pre-reaction image deviation threshold value preset in the database after a preset number of times (set by the preset personnel, generally 5 times), a higher-precision device is used to adjust the illumination intensity of the environment step by step.

[0069] Furthermore, the post-reaction CSA image deviation degree data includes: a CSA preparation deviation degree evaluation index for qualified color-sensitive sensor preparation and adjustment, a pre-reaction CSA image deviation degree evaluation index for qualified pre-reaction image adjustment, a CSA preparation deviation degree threshold, a pre-reaction image deviation threshold, environmental data and image data; the environmental data includes the post-reaction CSA light intensity, the post-reaction device lens transmittance, the post-reaction CSA light intensity reference value and the post-reaction device lens transmittance reference value; the image data includes the post-reaction CSA image clarity index, the post-reaction CSA image signal-to-noise ratio, the post-reaction CSA image chromatic aberration, the post-reaction CSA image clarity index reference value, the post-reaction CSA image signal-to-noise ratio reference value and the post-reaction CSA image chromatic aberration reference value; the post-reaction CSA image deviation degree evaluation index is used to reflect the deviation degree of the post-reaction CSA image.

[0070] In this embodiment, the CSA image deviation degree data after reaction is used to analyze and obtain the CSA image deviation degree evaluation index after reaction; the CSA light intensity after reaction indicates the light intensity when the CSA image after reaction is acquired; the light transmittance of the device lens after reaction indicates the light transmittance of the lens of the device when the CSA image is acquired after reaction; the CSA image clarity index after reaction indicates the clarity index of the CSA image acquired after reaction; the CSA image signal-to-noise ratio after reaction indicates the signal-to-noise ratio of the CSA image acquired after reaction; the CSA image chromatic aberration after reaction indicates the chromatic aberration of the CSA image acquired after reaction; The illumination intensity reference value indicates the illumination intensity when the CSA image is acquired when the post-reaction CSA image is qualified; the post-reaction device lens transmittance reference value indicates the transmittance of the lens of the device when the CSA image is acquired when the post-reaction CSA image is qualified; the post-reaction CSA image clarity index reference value indicates the clarity index of the CSA image acquired when the post-reaction CSA image is qualified; the post-reaction CSA image signal-to-noise ratio reference value indicates the signal-to-noise ratio of the CSA image acquired when the post-reaction CSA image is qualified; the post-reaction CSA image chromatic aberration reference value indicates the chromatic aberration of the CSA image acquired when the post-reaction CSA image is qualified.

[0071] The ambient light when the CSA image is acquired is directly measured by a light intensity measuring instrument (such as a light intensity meter or a radiometer) to obtain the CSA light intensity after the reaction. The spectral transmittance of the lens of the device that acquires the CSA image is measured by a spectral transmittance tester to obtain the transmittance of the device lens after the reaction. The clarity of the CSA image is measured by an image clarity meter. The clarity meter measures the maximum and minimum luminous flux of the image and uses the clarity calculation formula to obtain the clarity index, which is the clarity index of the CSA image after the reaction. The signal-to-noise ratio of the CSA image is calculated by image processing software or algorithm to obtain the signal-to-noise ratio of the CSA image after the reaction. The color difference of the CSA image is measured by a color difference detector or a spectrophotometer to obtain the color difference of the CSA image after the reaction.

[0072] The post-reaction CSA light intensity, post-reaction equipment lens transmittance, post-reaction CSA image clarity index, post-reaction CSA image signal-to-noise ratio and post-reaction CSA image color difference when the deviation degree of the historical post-reaction CSA image is qualified are obtained from the database and averaged respectively to obtain the post-reaction CSA light intensity reference value, post-reaction equipment lens transmittance reference value, post-reaction CSA image clarity index reference value, post-reaction CSA image signal-to-noise ratio reference value and post-reaction CSA image color difference reference value.

[0073] This embodiment obtains a post-reaction CSA image deviation degree evaluation index by comprehensively analyzing the post-reaction CSA illumination intensity, the post-reaction device lens transmittance, the post-reaction CSA image clarity index, the post-reaction CSA image signal-to-noise ratio and the post-reaction CSA image chromatic aberration, so that the post-reaction CSA image deviation degree evaluation index is considered more detailed and specific, which is conducive to more accurate analysis of whether to perform post-reaction image adjustment.

[0074] Furthermore, the specific method for obtaining the evaluation index of the deviation degree of the post-reaction CSA image is as follows: the result of the difference operation between the post-reaction CSA illumination intensity and the post-reaction CSA illumination intensity reference value is processed with the post-reaction CSA illumination intensity reference value to obtain the post-reaction illumination intensity score (i.e. ); The result of the ratio operation of the post-reaction device lens transmittance and the post-reaction device lens transmittance reference value is processed to obtain the post-reaction transmittance score (i.e. ); The result of the ratio operation of the post-reaction CSA image clarity index and the post-reaction CSA image clarity index reference value is processed to obtain the post-reaction clarity score (i.e. ); The sum of the post-reaction CSA image signal-to-noise ratio and the post-reaction CSA image signal-to-noise ratio reference value is processed with the result of the ratio operation to obtain the post-reaction signal-to-noise ratio score (i.e. ); The result of the ratio operation of the color difference of the post-reaction CSA image and the color difference reference value of the post-reaction CSA image is processed to obtain the post-color difference score (i.e. ); The post-reaction CSA image deviation degree evaluation index is obtained according to the CSA preparation deviation degree evaluation index of the color-sensitive sensor preparation adjustment, the pre-reaction CSA image deviation degree evaluation index of the pre-reaction image adjustment, the CSA preparation deviation degree threshold, the pre-reaction image deviation threshold, the post-illumination intensity score, the post-transmittance score, the post-clarity score, the post-signal-to-noise ratio score and the post-color difference score.

[0075] Specifically, the specific limiting expression of the evaluation index of the deviation degree of the CSA image after the reaction is:

[0076]

[0077] In the formula, It is expressed as the evaluation index of the deviation degree of the CSA image after the reaction, which is qualified for the pre-reaction image adjustment. It is an evaluation index of the deviation degree of the pre-reaction CSA image after the pre-reaction image adjustment is qualified. Denoted as the pre-reaction image deviation threshold, It represents the evaluation index of the deviation degree of CSA preparation for color sensor preparation adjustment. It is expressed as the CSA preparation deviation threshold, It is expressed as the CSA light intensity after the color sensor reacts. It is expressed as the light transmittance of the device lens after the color sensor reacts. It is expressed as the CSA image clarity index after the color sensor reacts. It is expressed as the CSA image signal-to-noise ratio after the color sensor reacts, It is expressed as the CSA image color difference after the color sensor reacts, It is expressed as the CSA light intensity reference value after the color sensor reacts. It is expressed as the reference value of the light transmittance of the device lens after the color sensor reacts. It is expressed as the CSA image clarity index reference value after the color sensor reacts. It is expressed as the reference value of the CSA image signal-to-noise ratio after the color sensor reacts. It is expressed as the CSA image color difference reference value after the color sensor reacts. Expressed as the fraction of post-illumination intensity, Expressed as the rear transmittance fraction, Expressed as the post-clarity score, Expressed as the posterior signal-to-noise ratio fraction, Expressed as the posterior color difference score, Expressed as a natural constant.

[0078] In this embodiment, the algorithm of this embodiment combines the post-reaction CSA image deviation degree data for comprehensive analysis to obtain a post-reaction CSA image deviation degree evaluation index. In the algorithm of this embodiment, the post-reaction CSA image deviation degree data are not independent and are interrelated. The smaller the pre-reaction CSA image deviation degree evaluation index is when the pre-reaction image is adjusted to be qualified, the smaller the preparation deviation of the CSA is, and the smaller the impact on the pre-reaction CSA image is. In turn, the smaller the CSA preparation deviation degree evaluation index is when the color-sensitive sensor is prepared and adjusted to be qualified, the more accurate the CSA is before the reaction, the smaller the impact on the deviation of the post-reaction CSA image is. At this time, when the post-reaction CSA light intensity and the post-reaction device lens transmittance are The smaller the differences between the post-reaction CSA image clarity index, post-reaction CSA image signal-to-noise ratio and post-reaction CSA image color difference and the post-reaction CSA illumination intensity reference value, post-reaction equipment lens transmittance reference value, post-reaction CSA image clarity index reference value, post-reaction CSA image signal-to-noise ratio reference value and post-reaction CSA image color difference reference value respectively, the smaller the rear illumination intensity score, rear transmittance score, rear clarity score, rear signal-to-noise ratio score and rear color difference score, and thus the smaller the post-reaction CSA image deviation degree evaluation index. The smaller the post-reaction CSA image deviation degree evaluation index is, the smaller the deviation degree of the post-reaction CSA image is, and the less need for post-reaction image adjustment is.

[0079] The algorithm of this embodiment obtains a post-reaction CSA image deviation degree evaluation index by comprehensively analyzing the post-reaction CSA illumination intensity, the post-reaction device lens transmittance, the post-reaction CSA image clarity index, the post-reaction CSA image signal-to-noise ratio and the post-reaction CSA image chromatic aberration, so that the post-reaction CSA image deviation degree evaluation index considers a more comprehensive and specific angle, so that it is more accurate to analyze whether to perform post-reaction image adjustment based on the post-reaction CSA image deviation degree evaluation index, and to perform post-reaction image adjustment more accurately.

[0080] Furthermore, the specific process of analyzing whether to identify adulterated tea oil is as follows: comparing the post-reaction CSA image deviation degree evaluation index with the post-reaction image deviation threshold in the database; when the post-reaction CSA image deviation degree evaluation index is less than the post-reaction image deviation threshold in the database, no post-reaction image adjustment is performed, and adulterated tea oil identification is performed; when the post-reaction CSA image deviation degree evaluation index is not less than the post-reaction image deviation threshold in the database, post-reaction image adjustment is performed, and the post-reaction image adjustment includes adjusting the image acquisition light intensity and adjusting the image acquisition settings; when the post-reaction CSA image deviation degree evaluation index monitored after the post-reaction image adjustment is less than the post-reaction image deviation threshold in the database, the post-reaction image adjustment is stopped.

[0081] Post-reaction image adjustment includes adjusting the image acquisition illumination intensity and adjusting the image acquisition settings. Adjusting the image acquisition illumination intensity means adjusting the illumination intensity of the environment for acquiring the post-reaction CSA image step by step (e.g., adjusting by 3% each time). Adjusting the lens cleanliness means wiping the lens of the device for acquiring the CSA image. Adjusting the image acquisition settings means adjusting the lens focal length (e.g., adjusting by 1 mm each time), aperture (e.g., adjusting by 0.5 F value units each time), and focus (e.g., adjusting by one tenth of a circle each time) of the device for acquiring the post-reaction CSA image step by step. When the post-reaction image adjustment is completed for the preset number of times, and the monitored post-reaction CSA image deviation evaluation index is not less than the post-reaction image deviation threshold in the database, use a higher-precision device (e.g., a camera with higher adjustable precision) to adjust the above data step by step.

[0082] The technical solution in the above-mentioned embodiment of the present application obtains the CSA preparation deviation degree evaluation index through the CSA preparation deviation degree data and analyzes whether to perform color-sensitive sensor preparation adjustment; after the color-sensitive sensor preparation adjustment is qualified, the pre-reaction CSA image deviation degree evaluation index is obtained according to the pre-reaction CSA image deviation degree data and analyzes whether to perform pre-reaction image adjustment; after the pre-reaction image adjustment is qualified, the post-reaction CSA image deviation degree evaluation index is obtained according to the post-reaction CSA image deviation degree data and analyzes whether to perform adulterated tea oil identification, thereby achieving the improvement of the accuracy of the analysis data in the identification of adulterated tea oil, and effectively solving the problem of inaccurate analysis data in the identification of adulterated tea oil in the prior art.

[0083] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0084] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0085] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0086] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0087] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0088] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for identifying adulterated tea oil based on a machine learning algorithm, characterized in that: The following steps are involved: Acquire CSA preparation deviation degree data, obtain CSA preparation deviation degree evaluation index according to the CSA preparation deviation degree data, and analyze whether to make color sensitive sensor preparation adjustment; After the color-sensitive sensor is prepared and adjusted to be qualified, the pre-reaction CSA image deviation degree data is obtained, and the pre-reaction CSA image deviation degree evaluation index is obtained according to the pre-reaction CSA image deviation degree data, and whether to perform pre-reaction image adjustment is analyzed; After the pre-reaction image adjustment is qualified, the post-reaction CSA image deviation degree data is obtained, and the post-reaction CSA image deviation degree evaluation index is obtained according to the post-reaction CSA image deviation degree data, and whether to identify the adulterated tea oil is analyzed; The CSA preparation deviation degree data include: pH indicator solution concentration, solution ultrasonic treatment time, NaOH solution usage, CSA ventilation time, pH indicator solution concentration standard value, solution ultrasonic treatment time standard value, NaOH solution usage standard value and CSA ventilation time standard value; The CSA preparation deviation degree evaluation index is used to quantify the deviation degree of CSA preparation; The specific method for obtaining the CSA preparation deviation evaluation index is as follows: The result of the difference operation between the concentration of the pH indicator solution and the standard value of the concentration of the pH indicator solution is processed with the standard value of the concentration of the pH indicator solution to obtain a concentration score of the pH indicator solution; The result of the ratio calculation of the ultrasonic treatment time of the solution and the standard value of the ultrasonic treatment time of the solution is processed to obtain a treatment time score; The result of the ratio calculation of the sum of the amount of NaOH solution used and the standard value of the amount of NaOH solution used and twice the standard value of the amount of NaOH solution used is processed to obtain the usage score; The result of the ratio calculation of CSA ventilation time and CSA ventilation time standard value is processed to obtain the ventilation time score; The evaluation index of the degree of deviation in CSA preparation was obtained based on the pH indicator solution concentration score, treatment time score, usage score and ventilation time score; The specific method for obtaining the pre-reaction CSA image deviation degree evaluation index is as follows: The pre-light intensity score is obtained by processing the pre-reaction CSA light intensity and the pre-reaction CSA light intensity reference value; The front transmittance score is obtained by processing the front device lens transmittance and the front device lens transmittance reference value; The pre-signal-to-noise ratio score is obtained by processing the pre-reaction CSA image signal-to-noise ratio and the pre-reaction CSA image signal-to-noise ratio reference value; The pre-color difference score is obtained by processing the pre-reaction CSA image color difference and the pre-reaction CSA image color difference reference value; According to the qualified CSA preparation deviation evaluation index, front illumination intensity score, front transmittance score, front signal-to-noise ratio score and front color difference score of the color-sensitive sensor preparation adjustment, the front CSA image deviation evaluation index is obtained.

2. A method for identifying adulterated tea oil based on a machine learning algorithm as claimed in claim 1, characterized in that: The specific process of analyzing whether to perform color-sensitive sensor preparation adjustment is as follows: Compare the CSA preparation deviation evaluation index with the CSA preparation deviation threshold preset in the database; When the CSA preparation deviation evaluation index is less than the CSA preparation deviation threshold preset in the database, the color sensor preparation adjustment is not performed; When the CSA preparation deviation evaluation index is not less than the CSA preparation deviation threshold preset in the database, the preset personnel are reminded to adjust the color-sensitive sensor preparation; when the monitored CSA preparation deviation evaluation index is less than the preset CSA preparation deviation threshold, the color-sensitive sensor preparation adjustment is stopped; The color-sensitive sensor preparation and adjustment includes reconfiguring the color-sensitive sensor and adjusting the configuration equipment.

3. A method for identifying adulterated tea oil based on a machine learning algorithm as claimed in claim 2, characterized in that: The pre-reaction CSA image deviation degree data includes: a CSA preparation deviation degree evaluation index of a qualified color-sensitive sensor preparation and adjustment, a pre-reaction CSA illumination intensity, a pre-reaction device lens transmittance, a pre-reaction CSA image signal-to-noise ratio, a pre-reaction CSA image chromatic aberration, a pre-reaction CSA illumination intensity reference value, a pre-reaction device lens transmittance reference value, a pre-reaction CSA image signal-to-noise ratio reference value, and a pre-reaction CSA image chromatic aberration reference value; The pre-reaction CSA image deviation degree evaluation index is used to quantify the deviation degree of the pre-reaction CSA image.

4. A method for identifying adulterated tea oil based on a machine learning algorithm as claimed in claim 3, characterized in that: The specific process of analyzing whether to perform pre-reaction image adjustment is as follows: Compare the pre-reaction CSA image deviation degree evaluation index with the pre-reaction image deviation threshold preset in the database; When the pre-reaction CSA image deviation evaluation index is less than the pre-reaction image deviation threshold preset in the database, no pre-reaction image adjustment is performed; When the pre-reaction CSA image deviation evaluation index is not less than the pre-reaction image deviation threshold preset in the database, the pre-reaction image adjustment is performed, wherein the pre-reaction image adjustment includes adjusting the image acquisition illumination intensity and adjusting the lens cleanliness; When the monitored pre-reaction CSA image deviation degree evaluation index is less than the pre-reaction image deviation threshold preset in the database, the pre-reaction image adjustment is stopped.

5. A method for identifying adulterated tea oil based on a machine learning algorithm as claimed in claim 1, characterized in that: The post-reaction CSA image deviation degree data includes: a CSA preparation deviation degree evaluation index of a qualified color sensor preparation adjustment, a pre-reaction CSA image deviation degree evaluation index of a qualified pre-reaction image adjustment, a CSA preparation deviation degree threshold, a pre-reaction image deviation threshold, environmental data and image data; The environmental data includes CSA light intensity after reaction, device lens transmittance after reaction, CSA light intensity reference value after reaction, and device lens transmittance reference value after reaction; The image data includes a post-reaction CSA image clarity index, a post-reaction CSA image signal-to-noise ratio, a post-reaction CSA image color difference, a post-reaction CSA image clarity index reference value, a post-reaction CSA image signal-to-noise ratio reference value, and a post-reaction CSA image color difference reference value; The post-reaction CSA image deviation degree evaluation index is used to reflect the deviation degree of the post-reaction CSA image.

6. A method for identifying adulterated tea oil based on a machine learning algorithm as claimed in claim 5, characterized in that: The specific method for obtaining the evaluation index of the post-reaction CSA image deviation degree is as follows: The result of the difference operation between the post-reaction CSA light intensity and the post-reaction CSA light intensity reference value is processed with the post-reaction CSA light intensity reference value to obtain the post-reaction light intensity score; The result of the ratio operation of the light transmittance of the lens of the post-reaction device and the light transmittance reference value of the lens of the post-reaction device is processed to obtain the light transmittance score; The result of the ratio operation of the post-reaction CSA image clarity index and the post-reaction CSA image clarity index reference value is processed to obtain a post-reaction clarity score; The result of the ratio operation of the sum of the post-reaction CSA image signal-to-noise ratio and the post-reaction CSA image signal-to-noise ratio reference value and the post-reaction CSA image signal-to-noise ratio reference value is processed to obtain a post-reaction signal-to-noise ratio score; The result of the ratio operation of the post-reaction CSA image color difference and the post-reaction CSA image color difference reference value is processed to obtain the post-reaction color difference score; According to the CSA preparation deviation evaluation index of the color sensor preparation adjustment, the pre-reaction CSA image deviation evaluation index of the pre-reaction image adjustment, the CSA preparation deviation threshold, the pre-reaction image deviation threshold, the post-illumination intensity score, the post-transmittance score, the post-clarity score, the post-signal-to-noise ratio score and the post-color difference score, the post-reaction CSA image deviation evaluation index is obtained; The specific process of analyzing whether to identify adulterated tea oil is as follows: Compare the evaluation index of the post-reaction CSA image deviation degree with the post-reaction image deviation threshold in the database; When the evaluation index of the deviation degree of the CSA image after reaction is less than the deviation threshold of the image after reaction in the database, the adulterated tea oil is identified; When the post-reaction CSA image deviation evaluation index is not less than the post-reaction image deviation threshold in the database, performing post-reaction image adjustment, wherein the post-reaction image adjustment includes adjusting the image acquisition illumination intensity and adjusting the image acquisition settings; When the monitored post-reaction CSA image deviation degree evaluation index is less than the post-reaction image deviation threshold in the database, adulterated tea oil identification is performed.

7. A method for identifying adulterated tea oil based on a machine learning algorithm as claimed in claim 6, characterized in that: The specific limiting expression of the evaluation index of the post-reaction CSA image deviation degree is: ; In the formula, It is expressed as the evaluation index of the deviation degree of the CSA image after the reaction, which is qualified for the pre-reaction image adjustment. It is an evaluation index of the deviation degree of the pre-reaction CSA image after the pre-reaction image adjustment is qualified. Denoted as the pre-reaction image deviation threshold, It is an evaluation index of the deviation degree of CSA preparation that is qualified for color sensor preparation adjustment. It is expressed as the CSA preparation deviation threshold, Expressed as the fraction of post-illumination intensity, Expressed as the rear transmittance fraction, Expressed as the post-clarity score, Expressed as the posterior signal-to-noise ratio fraction, Expressed as the post-chromatic aberration score.

Images