An intelligent monitoring method and system for a radix ginseng and astragalus membranaceus beverage production workshop

Abstract

The present application belongs to the technical field of image processing, and particularly relates to an intelligent monitoring method and system for a party ginseng and astragalus drink production workshop, which comprises the following steps: obtaining a liquid region sequence in a liquid image sequence; recording any liquid region as a target region, extracting a profile straight line, and calculating a straight line information weight; calculating a target adhesion coefficient according to the difference between the average gray value of all profile straight lines of the target region after being affected by the corresponding weight and the total sum of all profile straight line information weights of the target region; calculating a target liquid film flow lag index according to the difference between the average displacement module length of all target displacement components and the maximum average displacement module length; calculating an end point decision index according to the relationship between the absolute value average of all pixel points of the target region and the target liquid film flow lag index, positioning the inflection point of the end point decision index curve, and performing control. The present application solves the technical problem that it is difficult for a density sensor to determine a wall-hanging state.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of image processing. More particularly, the present application relates to an intelligent monitoring method and system for a radix notoginseng and astragalus beverage production workshop. BACKGROUND

[0002] In the production process of radix notoginseng and astragalus beverage, the concentration step of the extract is crucial, as it directly determines the effective component concentration and taste of the beverage. Only by controlling the concentration process to the appropriate endpoint can the quality of the beverage be guaranteed to meet the standards, so it is necessary to accurately monitor the concentration endpoint.

[0003] The traditional method of monitoring the concentration endpoint relies on manual visual observation of the wall-hanging phenomenon, i.e., experienced pharmacists judge the endpoint by observing the thickness, color and flow rate of the liquid film on the inner wall of the tank. However, this method relies entirely on personal experience and is highly subjective, with differences in judgment criteria among different pharmacists and instability in the judgment of the same pharmacist at different times, leading to large fluctuations in the viscosity, taste and effective component content of different batches of products, making it difficult to achieve standardized production.

[0004] The prior art uses a density sensor to monitor the density of the extract in real time, and judges the concentration endpoint based on the change in density. This method achieves efficient monitoring of the concentration endpoint.

[0005] However, this method has limitations for radix notoginseng and astragalus liquor, i.e., the specific polysaccharides and saponins dissolved during the co-boiling of radix notoginseng and astragalus will cause the liquor to exhibit wall-hanging characteristics. The density sensor can only reflect the physical properties of the overall solution and cannot accurately calculate the wall-hanging state and fluid dynamics changes caused by the polysaccharides such as saponins, which are crucial for determining the optimal endpoint. Therefore, the monitoring results still have deviations and cannot fully meet the needs of high-quality production. SUMMARY

[0006] To solve the technical problem that the density sensor cannot calculate the wall-hanging state, the present application provides solutions in the following aspects.

[0007] In a first aspect, the present application provides an intelligent monitoring method for a radix notoginseng and astragalus beverage production workshop, comprising:

[0008] After the images of the observation window are continuously collected by the industrial camera at a preset time, preprocessed and grayscaled, the liquid image sequence is obtained by arranging in time sequence; the liquid region sequence in the liquid image sequence is obtained according to the gray value of the liquid surface region; any liquid region is recorded as a target region, the profile straight line is extracted, the straight line information weight is calculated according to the gray value of the profile straight line and the stability of the gray value; the target adhesion coefficient is calculated according to the difference between the average gray value of all profile straight lines of the target region after being affected by the corresponding weight and the total weight of all profile straight lines of the target region; the target liquid film flow stagnation index is calculated according to the target adhesion coefficient; the target displacement component is calculated according to the movement state of the pixel points in the target region corresponding liquid image and the last liquid image; the target liquid film flow stagnation index is calculated according to the difference between the average displacement module length of all target displacement components and the maximum average displacement module length; the end decision index is calculated according to the relationship between the absolute value mean of all pixel points in the target region and the target liquid film flow stagnation index, the inflection point of the end decision index curve is located, and control is performed.

[0009] The present application can automatically collect images, calculate wall sticking degree, flow state and decision index, and does not need manual subjective judgment throughout the process, thereby reducing the judgment difference of different personnel and different time periods; the present application can accurately identify the concentration characteristics of the medicine liquid by analyzing the wall sticking degree, flow change and saponin streak of the liquid film, thereby avoiding the monitoring deviation caused by only relying on density data. Meanwhile, the present application can reduce the problems of inaccurate data at the initial stage of concentration, difference between different batches of equipment, and single factor misjudgment by determining effective liquid film conditions, normalizing data, determining the end point by the inflection point, and the like, thereby ensuring that the monitoring result is more accurate and avoiding excessive or insufficient concentration, solving the main defects of the traditional method and the prior art in the concentration monitoring of the medicine liquid, and ensuring the stable product quality.

[0010] Preferably, the liquid region sequence in the liquid image sequence is obtained, including:

[0011] The image sequence of the observation window is continuously collected at a preset time interval of 1 second within the concentration time , and the noise is removed; the grayscaling is performed on the result image after the noise removal, and the liquid image sequence is obtained by arranging according to the shooting time; the threshold segmentation algorithm is used, the region with a gray value lower than the optimal segmentation threshold in the liquid image sequence is regarded as a liquid region, and all the liquid regions in the liquid image sequence are recorded as the liquid region sequence.

[0012] Preferably, the straight line information weight satisfies the following expression:

[0013] ;

[0014] In the formula, The line information weight of the m-th profile line; Indicates the maximum grayscale value; This represents the average gray value of the m-th profile line; The gray standard deviation of the m-th profile line; Represents the normalization function; It represents a very small positive number, and guarantees that the denominator is not 0.

[0015] Preferably, the target adhesion coefficient satisfies the following expression:

[0016] ;

[0017] In the formula, Indicates the target adhesion coefficient; The line information weight of the m-th profile line; Indicates the baseline grayscale value; This represents the average gray value of the m-th profile line; This represents the normalization function.

[0018] This invention accurately measures the degree to which the liquid film adheres to the tank wall. By using the grayscale value of the liquid before concentration as a benchmark, it eliminates the need for visual estimation of film thickness or adhesion, reducing the variability in human judgment. Furthermore, weighted calculation using linear information weights highlights the influence of reliable regions and weakens the role of interfering regions. Simultaneously, by normalizing the weighted sum, it eliminates the differences caused by different numbers or weights of profile lines, providing a clear visual indication of the changing trend of liquid film adhesion during concentration and offering a definitive basis for judging the concentration state.

[0019] Preferably, the calculation of the target liquid film hysteresis index is initiated, including:

[0020] Through three preliminary experiments, the concentration of the Codonopsis and Astragalus decoction was recorded three times consecutively in each experiment. Value, taken 3 times The average value as When the target adhesion coefficient does not exceed Continue monitoring changes in the target adhesion coefficient; when the target adhesion coefficient exceeds When it is determined that an effective wall liquid film has been formed, the calculation of the target liquid film hysteresis index is initiated.

[0021] Preferably, calculating the target displacement components includes:

[0022] The liquid image corresponding to the target region is positioned with the last frame liquid image, the Shi-Tomasi corner detection algorithm is used to select the feature points in the target region, and the Lucas-Kanade optical flow method is used to calculate the displacement components in the horizontal direction and the displacement components in the vertical direction of the feature points in a 3*3 neighborhood around each feature, which are denoted as target displacement components.

[0023] The present application can accurately calculate the fine movement of the liquid film, without relying on the human eye to observe the speed of the liquid film flow, and avoids the judgment deviation caused by the inability of the eye to observe small amplitude movement. By extracting feature points on the liquid film and then calculating the movement of these points in the horizontal and vertical directions, even small movements can be accurately calculated, allowing the flow state of the liquid film to be clearly represented by data. The displacement components obtained in this way can truly reflect the flow of the liquid film, providing an accurate basis for subsequent measurement of the degree of slowing down of the liquid film flow, and ensuring more accurate monitoring of the flow state of the liquid film.

[0024] Preferably, the target liquid film flow lag index satisfies the following expression:

[0025] ;

[0026] In the formula, The target liquid film flow lag index is denoted by, and the value range is 0 to 1; 、 The feature point The optical flow displacement components in the horizontal direction and the vertical direction; The number of feature points in the target region is denoted by; The maximum modulus length is denoted by; The minimum positive number is denoted by, to ensure that the denominator is not 0; The exponential function with the natural constant as the base number is denoted by.

[0027] The present application can clearly reflect the degree of slowing down of the liquid film flow, with a value between 0 and 1, making it easy to understand the flow state of the liquid film. The larger the value, the more difficult it is for the liquid film to flow. The present application can also eliminate the differences in absolute values of the liquid film flow speed under different batches and different devices. Without relying on the human eye to observe the speed difference of the liquid film flow, the present application reduces the deviation of subjective judgment, and directly observes the change of the liquid film from easy flow to difficult flow during the concentration process, helping the staff to accurately grasp the concentration progress.

[0028] Preferably, the end point decision index satisfies the following expression:

[0029] ;

[0030] In the formula, The end point decision index is denoted by; The target liquid film flow lag index is denoted by; The ginsenoside banding index is denoted by; represents a logarithmic function.

[0031] The application determines the concentration end point in combination with the flowing state of the liquid film and the streaks formed by saponins. When the liquid film is still flowing or has no streaks, the decision index will not increase, avoiding early determination of the end point. Only when the liquid film is basically stationary and the streaks are obvious, the decision index will increase significantly.

[0032] Preferably, the inflection point of the end point decision index curve is located, and the control includes:

[0033] All end point decision indexes in the liquid image sequence are continuously calculated, and a time evolution curve of the end point decision index is drawn with time as the horizontal axis and the end point decision index as the vertical axis. When a turning point is detected in which the end point decision index value changes from continuous increase to decrease, it is determined that the concentration process has reached the optimal end point, and a corresponding control instruction is triggered immediately.

[0034] In a second aspect, the application provides an intelligent monitoring system for a radix ginseng and astragalus drink production workshop, comprising a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned intelligent monitoring method for the radix ginseng and astragalus drink production workshop is realized.

[0035] By adopting the above technical solution, the above-mentioned intelligent monitoring method for the radix ginseng and astragalus drink production workshop is generated into a computer program and stored in the memory to be loaded and executed by the processor, so that a terminal device is made according to the memory and the processor, and use is facilitated.

[0036] The application has the advantages that the application provides an efficient and reliable concentration monitoring scheme for radix ginseng and astragalus drink production, realizes automation of concentration monitoring from the production level, reduces manual operation links, reduces the probability of human error, improves production efficiency, avoids waste of raw materials caused by delay or error in manual judgment, ensures the content of effective ingredients, taste and viscosity of each batch of products to be consistent from the quality level, improves product quality stability, enhances consumer trust and product market competitiveness, and provides a reference for concentration monitoring of other drinks containing special ingredients from the industry level, promotes the standardization and intelligent development of the drink production industry, reduces quality fluctuations caused by improper monitoring methods in the industry, helps the industry to improve the overall production level, and realizes a high-quality and sustainable production mode. BRIEF DESCRIPTION OF DRAWINGS

[0037] Figure 1 is a flowchart schematically showing an intelligent monitoring method for a radix ginseng and astragalus drink production workshop in the application;

[0038] Figure 2 is a schematic view of a concentration tank observation window and a drug solution concentration gray scale diagram in the application. Detailed Implementation

[0039] This invention discloses an intelligent monitoring method for use in a Codonopsis and Astragalus beverage production workshop, referring to... Figure 1 This includes steps S1-S4:

[0040] S1: Use an industrial camera to continuously acquire images of the observation window at preset times, and after preprocessing and grayscale conversion, arrange them in chronological order to obtain a liquid image sequence; based on the grayscale value of the liquid surface area, obtain the liquid region sequence in the liquid image sequence.

[0041] It should be noted that in production scenarios requiring precise control of the concentration endpoint, such as for Codonopsis and Astragalus beverages, traditional manual monitoring methods rely on experience to judge the adhesion status, which suffers from strong subjectivity and inconsistent batch quality. To achieve automated and precise monitoring, this invention constructs a monitoring system based on industrial vision technology. By placing an industrial camera parallel to the observation window of the concentration tank, a vertically stable shooting angle is ensured. Images from the observation window are continuously acquired at fixed time intervals. After image denoising and grayscale conversion, a time-series image sequence is formed. Then, a threshold segmentation algorithm is used to process the images, automatically distinguishing the liquid area from the tank wall background based on the difference in grayscale values. The liquid areas at different times are then arranged into a liquid area sequence in chronological order.

[0042] Specifically, a temporal pixel variance profile detection algorithm is used to locate the observation window where the liquid surface appears in the concentration tank. The industrial camera is placed parallel to the observation window at the same height to ensure that the camera's shooting angle can vertically and stably cover the observation window area. During the concentration time, the camera operates at preset 1-second time intervals. The system continuously acquires image sequences from the observation window, performs denoising, and then converts the denoised image to grayscale. Figure 2 The right image is shown in the observation window of the concentration tank and the grayscale image of the concentrated liquid. The images are arranged according to the shooting time to obtain a liquid image sequence. A threshold segmentation algorithm is used to treat the regions in the liquid image sequence whose grayscale values ​​are lower than the optimal segmentation threshold as liquid regions. All the corresponding liquid regions in the liquid image sequence are recorded as a liquid region sequence.

[0043] Thus, the liquid region sequence was obtained.

[0044] S2: Denote any liquid region as the target region, extract the profile lines, and calculate the line information weight based on the gray value of the profile lines and the stability of the gray value. Calculate the target adhesion coefficient based on the difference between the average gray value of all profile lines in the target region after being affected by their corresponding weights and the sum of the information weights of all profile lines in the target region.

[0045] It should be noted that in the concentration process, the pharmaceutical workers tend to find the most ideal observation area in the liquid area, that is, the area can both reflect that the liquid is thick enough and ensure that the observation information is pure and reliable. The difference between the average gray value and the maximum gray value is calculated to represent the thickness or viscosity of the liquid film, and the gray standard deviation is used to evaluate the stability of the observation information, effectively identifying and suppressing visual noise caused by boiling, bubbles or reflection. And the present application calculates the comprehensive score of each profile straight line in the liquid area, gives different straight line information weights to different profile straight lines in the liquid area, and the profile straight line with the highest score represents the thicker liquid film area and the purer observation signal, so as to ensure that the subsequent analysis is always based on the most reliable and most representative concentration state of the liquid film state.

[0046] It should be noted that when dividing the profile straight line from the liquid area, one profile straight line is extracted every 10 pixel points, and 192 profile straight lines are extracted in each target area. Corresponding to the width of 1920 pixels, 1920 is the resolution setting value of the industrial camera 1920x1080. This method can ensure that the obtained profile straight line covers the entire target area, so here a is 10 and b is 192.

[0047] Specifically, in the liquid area sequence, any liquid area is extracted as a target area; a vertical straight line perpendicular to the horizontal coordinate in the target area is located, which is called a profile straight line; every a pixels are extracted along the horizontal direction of the image to extract a profile straight line, and b profile straight lines are extracted in total; the average gray value and the gray value standard deviation of all pixel points on a single profile straight line are calculated; according to the difference between the average gray value and the gray standard deviation, the internal information quality weight of the vertical profile is calculated, which is called the straight line information weight, including:

[0048] The straight line information weight satisfies the following expression:

[0049] ;

[0050] In the formula, represents the straight line information weight of the mth profile straight line; represents the maximum gray value; represents the average gray value of the mth profile straight line; represents the gray standard deviation of the mth profile straight line; represents a normalization function; represents a very small positive number, which ensures that the denominator is not 0.

[0051] In the formula, represents the difference between the average gray value and the maximum gray value of the mth profile straight line. If the liquid film in the concentration tank is thicker, the average gray value of the profile straight line is lower, smaller, The greater, the larger the molecule is; The greater, the more unstable the liquid film gray value changes, and the stronger the noise on the profile straight line is; In the formula, the numerator represents the viscosity of the liquid film in the region where the mth profile straight line is located, and the denominator represents the degree of disorder of the information when judging the viscosity of the liquid film, The ratio between the viscosity of the liquid film in the region where the mth profile straight line is located and the degree of disorder of the information when judging the viscosity of the liquid film is directly calculated, and in order to obtain a profile straight line with a high weight, the following conditions must be met The greater, and The smaller; for the profile straight line with a lower viscosity of the adhered liquid film, Close to , Smaller; for the profile straight line with a higher viscosity of the adhered liquid film, the numerator is maximized, and the denominator is minimized, Maximum.

[0052] It should be noted that in the concentration process of the radix ginseng and astragalus drink, in order to solve the problem of strong subjectivity and difficulty in standardization in the traditional manual visual inspection of the adhesion state of the liquid film, and realize accurate monitoring of the adhesion of the liquid film, the present application first obtains a reference gray value based on the liquid image before concentration, and then for each profile straight line in the target region, the absolute difference between the average gray value of the straight line and the reference gray value is weighted by using the straight line information weight, the weighted results of all straight lines are summed to represent the adhesion degree of the liquid film in the whole target region, and finally the sum of the straight line information weights is normalized to eliminate the influence of the number of profile straight lines or the total weight difference, so that the target adhesion coefficient corresponding to the liquid image at different times has comparability, thereby intuitively presenting the change rule of the adhesion characteristics of the liquid film with time in the concentration process.

[0053] Preferably, the liquid image before concentration is obtained from historical data and is denoted as a liquid image before concentration; the position of the liquid region in the liquid image is mapped to the liquid image before concentration to obtain a liquid region before concentration; the average gray value of the liquid region before concentration is calculated and is denoted as a reference gray value; the average gray value of the mth profile straight line corresponding to the target region is obtained, and the liquid film adhesion coefficient of the target region is calculated according to the reference gray value and the straight line information weight of the mth profile straight line, and is denoted as a target adhesion coefficient, including:

[0054] The target adhesion coefficient satisfies the following expression:

[0055] ;

[0056] In the formula, The target adhesion coefficient is denoted as; The straight line information weight of the mth profile straight line is denoted as; The reference gray value is denoted as; represents the average gray value of the mth profile straight line; represents a normalization function.

[0057] In the formula, represents the absolute difference value between the average gray value of the mth profile straight line and the reference gray value, which is used to represent the liquid film adhesion degree of the mth profile, and the greater the absolute difference value, the greater the liquid film adhesion degree of the mth profile; represents the weighted liquid film adhesion degree of the mth profile straight line; represents the sum of the weighted liquid film adhesion degrees of all profile straight lines in the target region, which is used to represent the liquid film adhesion degree of the target region; and the denominator represents the total weight of all profile straight line information in the target region, which is recorded as the target region information weight, and is used to normalize the liquid film adhesion degree of the target region, eliminate the difference in the number of profile straight lines or the total weight of the straight line information of the profile straight lines, make the target adhesion coefficients corresponding to the liquid images of different frames comparable, and more intuitively reflect the change of the liquid film adhesion characteristics with the concentration time.

[0058] S3: According to the size of the target adhesion coefficient, the target liquid film flow stagnation index is calculated; according to the movement state of the pixel points in the target region corresponding to the liquid image and the last frame liquid image, the target displacement component is calculated; according to the difference between the average displacement module length of all target displacement components and the maximum average displacement module length, the target liquid film flow stagnation index is calculated.

[0059] It should be noted that during the concentration of the radix ginseng and astragalus liquid, due to the specific polysaccharide saponin contained therein, a unique wall-hanging state is formed, and the traditional manual judgment of whether this state reaches the effective degree required for concentration has the problems of standard ambiguity and large batch difference. The viscosity of the liquid after concentration will increase sharply, and the downward flow speed of the wall-hanging liquid film will slow down significantly, that is, the flow stagnation state, which is an important signal that the concentration is close to the end. However, if the liquid film has not formed an effective thickness, such as the liquid film is thin and unstable at the initial stage of concentration, the flow stagnation index obtained at this time is not accurate enough, and the flow stagnation index exceeding the threshold value is required as the standard for calculating the flow stagnation index.

[0060] ​It should be noted that traditional manual observation is difficult to calculate dynamic details such as sliding and stagnation of liquid films at the millimeter or even micrometer level. Optical flow method can solve this problem precisely. By tracking the motion trajectory of feature points of the liquid film in consecutive frames of images, it accurately calculates the horizontal and vertical displacement components, transforming the liquid film flow that is difficult to detect with the naked eye into specific values. Especially for scenarios such as Codonopsis and Astragalus decoction, where there may be local unevenness in liquid film thickness and differences in flow velocity, the motion estimation capability of Lucas-Kanade optical flow method in small neighborhoods can effectively analyze subtle local flow differences and avoid ignoring local stagnation features due to the smooth overall flow of the liquid film.

[0061] Specifically, through three preliminary experiments, the concentration of the Codonopsis and Astragalus decoction was recorded three times consecutively in each experiment. Value, taken 3 times The average value as When the target adhesion coefficient does not exceed Continue monitoring changes in the target adhesion coefficient; when the target adhesion coefficient exceeds When it is determined that an effective wall liquid film has been formed, the calculation of the target liquid film hysteresis index is initiated.

[0062] It should be noted that, during the concentration process of Codonopsis pilosula and Astragalus membranaceus decoction, this invention focuses on analyzing the dynamic changes of the liquid film from flow to stagnation. This change is directly related to the degree of concentration of the Codonopsis pilosula and Astragalus membranaceus decoction. That is, as concentration proceeds, the concentration of components such as polysaccharides and saponins in the decoction increases, and the liquid film will gradually change from being easy to slide and flowing quickly in the initial stage to being viscous and adhering and flowing slowly, until it reaches a stable stagnation state at the end of the concentration. In this invention, the sum of the displacement modulus of all feature points is calculated and then averaged to eliminate the influence of the difference in the number of feature points at different locations on the liquid film. For example, there are fewer feature points in places where the liquid film is thin, so that the calculation result can reflect the overall flow level of the liquid film. In this invention, the average flow displacement of the liquid film is compared with the maximum average displacement modulus to normalize the difference in the absolute value of the flow rate under different batches and different equipment. For example, the initial fluidity of different batches of raw materials may be different, so that the flow state at different stages can be directly compared.

[0063] Preferably, the liquid image corresponding to the target area is compared with the previous frame liquid image. The Shi-Tomasi corner detection algorithm is used to select feature points within the target area. The Lucas-Kanade optical flow method is used to calculate the horizontal and vertical displacement components of the feature points in a 3×3 neighborhood around each feature, which are denoted as the target displacement components.

[0064] Preferably, all target displacement components in the target region are extracted, the average displacement module length is calculated according to all target displacement components, and the maximum average displacement module length in the initial stage of single batch concentration in the target region is obtained, which is denoted as target maximum module length; the liquid film flow stagnation index of the target region is calculated according to the target displacement component, the target maximum module length, and the number of feature points in the target region, which is denoted as target liquid film flow stagnation index, including:

[0065] The target liquid film flow stagnation index satisfies the following expression:

[0066]

[0067] In the formula, The target liquid film flow stagnation index is denoted, and the value range is 0 to 1; The feature point The optical flow displacement component in the horizontal direction and the vertical direction; The number of feature points in the target region is denoted; The target maximum module length is denoted; The minimum positive number is denoted to ensure that the denominator is not 0; The exponential function with the natural constant as the base number is denoted.

[0068] In the formula, The displacement module length sum of all feature points in the target region is denoted, reflecting the displacement of the liquid film as a whole; The average displacement module length of the target region is obtained by dividing the displacement module length sum of all feature points in the target region by the number of feature points in the target region, that is, the average flow displacement of the liquid film; The average flow displacement of the liquid film is divided by the target maximum module length to convert the maximum average displacement module length into a normalized average flow rate, eliminating the difference in absolute value of the flow rate in different scenarios; The average flow displacement of the liquid film is smaller when the flow rate is slower, the flow stagnation degree is higher, and the target liquid film flow stagnation index is larger.

[0069] S4: According to the mutual relationship between the absolute value mean of all pixel points in the target region and the target liquid film flow stagnation index, the end point decision index is calculated, the inflection point of the end point decision index curve is located, and control is performed.

[0070] It should be noted that the slowly growing saponin in the radix codonopsis and astragalus liquid will form a unique fine spot texture in the late concentration stage, which is an important feature that distinguishes it from other herbal liquids, and is also a key signal for the end of concentration. The target liquid film flow stagnation index can only be used to reflect the liquid film flow, while the saponin spot is an important feature of the radix codonopsis and astragalus liquid in the late concentration stage, and the concentration end of the radix codonopsis and astragalus liquid can be judged from the dynamic and micro texture.

[0071] ​​Specifically, the target region is filtered with Laplacian operator, and the mean value of the absolute value of all pixel points is calculated, denoted as saponin streak index; according to the saponin streak index and the target liquid film flow stagnation index, the radix of sophora flavescens and astragalus membranaceus concentration end point decision index is calculated, denoted as end point decision index, including:

[0072] The end point decision index satisfies the following expression:

[0073] ;

[0074] In the formula, denotes the end point decision index; denotes the target liquid film flow stagnation index; denotes the saponin streak index; denotes the logarithmic function.

[0075] In the formula, denotes the logarithmic enhancement of the saponin streak index, when there is no saponin streak, T close to 0, also close to 0, the whole end point decision index is suppressed to 0, when the saponin streak appears, T the value increases, will slowly increase, and play an enhancing role; using the logarithmic function can smooth T the sharp fluctuations of the value, making the model more stable; denotes that only when the flow stagnation index has reached a relatively high level, that is, the liquid film has basically stagnated, the appearance of saponin streak, T the value increases, can effectively increase the final decision index, if the liquid film is still flowing fast, R the value is low, even if there is noise similar to the streak, it will be suppressed by a very small R value.

[0076] It should be noted that during the concentration process, the end point decision index increases with the increase of concentration, until it reaches the balance point of the peak concentration of active ingredients and the best physical state, at which point further concentration will break the balance and cause the index to decrease. The inflection point corresponds to this balance state, ensuring that the concentration meets the dual goals of achieving the standard concentration and suitable physical state at the same time. The decision index at a single time point cannot reflect this dynamic trend, while the time series curve can fully present the whole process of exponential growth, slow growth, and inflection point decline. Through the precise positioning of the critical point by the first order derivative in mathematics, the situation of false judgment due to the high absolute value of the index is avoided.

[0077] Preferably, all end-point decision indexes in the liquid image sequence are continuously calculated, and a time evolution curve of the end-point decision index is drawn with time as the horizontal axis and the end-point decision index as the vertical axis, when a turning point is detected in which the end-point decision index value changes from continuously rising to falling, it is determined that the concentration process has reached the optimal end point, and corresponding control instructions are triggered immediately.

[0078] The embodiment of the present application further discloses an intelligent monitoring system for a radix ginseng and astragalus drink production workshop, comprising a processor and a memory, and the memory stores computer program instructions, which realize the intelligent monitoring method for the radix ginseng and astragalus drink production workshop according to the present application when executed by the processor.

[0079] The system further comprises other components such as a communication bus and a communication interface, which are well known to those skilled in the art, and their settings and functions are known in the art, so they will not be described here.

[0080] Although the present specification has shown and described several embodiments of the present application, it is obvious to those skilled in the art that such embodiments are provided only in an exemplary manner. Those skilled in the art will think of many changes, changes and alternatives without departing from the idea and spirit of the present application. It should be understood that various alternatives to the embodiments of the present application described herein can be employed in practicing the present application.

Claims

1. An intelligent monitoring method for a production workshop of a radix codonopsis astragalus drink, characterized in that, The method comprises the following steps: After the images of the observation window are continuously collected by an industrial camera at a preset time, preprocessed and grayscaled, a liquid image sequence is obtained by arranging the images in time sequence; According to the gray value of the liquid surface region, a liquid region sequence in the liquid image sequence is obtained; Any liquid region is recorded as a target region, a profile straight line is extracted, the gray value of the profile straight line and the stability of the gray value are calculated to obtain a straight line information weight, the difference between the average gray value of all profile straight lines of the target region after being affected by the corresponding weight and the total weight of all profile straight lines of the target region is calculated to obtain a target adhesion coefficient; According to the size of the target adhesion coefficient, a target liquid film flow stagnation index is calculated, and a target displacement component is calculated according to the movement of the pixel points in the target region corresponding liquid image and the previous liquid image; According to the difference between the average displacement module length of all target displacement components and the maximum average displacement module length, a target liquid film flow stagnation index is calculated; According to the relationship between the absolute value mean of all pixel points of the target region and the target liquid film flow stagnation index, an end decision index is calculated, and a turning point of the end decision index curve is located to control. 2.The intelligent monitoring method for a party ginseng and astragalus drink production workshop according to claim 1, characterized in that, The liquid region sequence in the liquid image sequence is obtained by: At the preset 1-second time interval within the concentration time , the image sequence of the observation window is continuously collected, denoising is performed, the denoised result image is grayed, and the liquid image sequence is obtained according to the shooting time; the threshold segmentation algorithm is used, the area with a gray value lower than the optimal segmentation threshold in the liquid image sequence is regarded as the liquid area, and all the liquid areas corresponding in the liquid image sequence are recorded as the liquid area sequence. 3.The intelligent monitoring method for a party ginseng astragalus drink production workshop according to claim 1, characterized in that, The straight line information weight satisfies the following expression: ； wherein represents the line information weight of the mth profile line; represents the maximum gray value; represents the average gray value of the mth profile line; represents the gray standard deviation of the mth profile line; represents the normalization function; represents a minimum positive number, which guarantees that the denominator is not 0. 4.The intelligent monitoring method for a party ginseng astragalus drink production workshop according to claim 1, characterized in that, The target adhesion coefficient satisfies the following expression: ； In the formula, represents a target adhesion coefficient; represents a linear information weight of the mth profile straight line; represents a reference gray value; represents an average gray value of the mth profile straight line; represents a normalization function. 5.The intelligent monitoring method for a party ginseng astragalus drink production workshop according to claim 1, characterized in that, The target liquid film flow stagnation index is calculated by: Through 3 pre-experiments, the average value of the values of the prescribed concentration of the Dangshen Huangqi liquid medicine recorded for 3 times continuously in each experiment is taken as the target adhesion coefficient; when the target adhesion coefficient does not exceed the target adhesion coefficient, the change of the target adhesion coefficient is continuously monitored; when the target adhesion coefficient exceeds the target adhesion coefficient, it is determined that an effective wall liquid film has been formed, and the calculation of the target liquid film flow lag index is started. ; when the target adhesion coefficient does not exceed the target adhesion coefficient, the change of the target adhesion coefficient is continuously monitored; when the target adhesion coefficient exceeds the target adhesion coefficient, it is determined that an effective wall liquid film has been formed, and the calculation of the target liquid film flow lag index is started.​​​​ 6.The intelligent monitoring method for a party ginseng astragalus drink production workshop according to claim 1, characterized in that, The target displacement component is calculated by: The liquid image corresponding to the target region and the previous liquid image are located, the Shi-Tomasi corner detection algorithm is used to select feature points in the target region, and the Lucas-Kanade optical flow method is used to calculate the displacement component in the horizontal direction and the displacement component in the vertical direction of the feature points in a 3*3 neighborhood around each feature, which is recorded as the target displacement component. 7.The intelligent monitoring method for a party ginseng astragalus drink production workshop according to claim 1, characterized in that, The target liquid film flow stagnation index satisfies the following expression: ； In the formula, represents the target liquid film flow lag index, and the value range is 0 to 1; , is a feature point The horizontal and vertical optical flow displacement components are shown in the following table: represents the number of feature points in the target area; represents the maximum modulus length of the target; represents a very small positive number, which ensures that the denominator is not zero; represents the exponential function with the natural constant as the base. 8.The intelligent monitoring method for a party ginseng astragalus drink production workshop according to claim 1, characterized in that, The end decision index satisfies the following expression: ； wherein represents an end-point decision index; represents a target liquid film flow lag index; represents a saponin banding index; represents a logarithmic function. 9.The intelligent monitoring method for a party ginseng astragalus drink production workshop according to claim 1, characterized in that, The turning point of the end decision index curve is located to control, which comprises: All end decision indexes in the liquid image sequence are continuously calculated, and a time evolution curve of the end decision index is drawn with time as the horizontal axis and the end decision index as the vertical axis. When the turning point of the end decision index value from continuous rise to decline is detected, it is determined that the concentration process has reached the best end point, and the corresponding control instruction is triggered immediately.

10. An intelligent monitoring system for a production workshop of a radix codonopsis astragalus drink, characterized in that, The method comprises the following steps: A processor and a memory, the memory stores computer program instructions, when the computer program instructions are executed by the processor, a method for intelligent monitoring of a radix rubra and astragali beverage production workshop according to any one of claims 1-9 is realized.

Images