Intelligent atomization quality inspection method, system and device for electronic atomized liquid production

Abstract

The invention discloses an electronic atomized liquid production-oriented intelligent atomization quality inspection method, system and device, and belongs to the technical field of new material detection, and the method comprises the following steps: S1, analyzing to obtain an atomization execution complex value; s2, a picture display quality judgment instruction is obtained, if the picture display quality judgment instruction is unqualified, background baseline graph updating frequency adjustment is carried out, an atomization quality inspection optimized image is shot again after adjustment, an equipment adjustment instruction is obtained through analysis, and otherwise, background baseline graph updating frequency adjustment is not executed; s3, if the equipment adjustment instruction is to execute camera equipment adjustment, performing corresponding camera equipment adjustment, otherwise, not executing camera equipment adjustment; s4, obtaining an atomization quality inspection judgment instruction of the electronic atomized liquid; and S5, a comprehensive quality inspection judgment instruction is obtained. The problem that in the prior art, the atomization quality inspection result of the electronic atomized liquid is not accurate is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of new material detection, and particularly relates to an intelligent atomization quality inspection method, system and device for electronic atomization liquid production. BACKGROUND

[0002] The existing intelligent atomization quality inspection system sets a light source and a spectrum sensor at a power interface, performs spectrum detection on the liquid to be atomized, compares the detection spectrum information with calibration spectrum information, determines whether the liquid contains harmful substances by a microcontroller, controls the equipment to shut down or standby, and issues a prompt or arranges an atomization system, a heating system and a real-time weighing device in an airflow channel to stably atomize the sample and record the weight change to measure the atomization rate of the liquid.

[0003] For example, the Chinese invention patent with the announcement number CN111650133B discloses a detection and control method of harmful substances in an electronic atomization solution and an electronic atomization device, which comprises: performing spectrum detection on a sample solution containing harmful substances to obtain calibration spectrum information, setting a light source assembly and a spectrum sensor assembly in the interface of a power device, detecting the solution to be atomized by the spectrum sensor assembly to obtain detection spectrum information, and comparing the detection spectrum information with the calibration spectrum information. If the detection spectrum information can match the calibration spectrum information, the microcontroller determines that the atomizer contains harmful substances and enters a shutdown unusable state, and a prompt unit simultaneously issues a warning prompt. If the detection spectrum information cannot match the calibration spectrum information, the microcontroller determines that the atomizer does not contain harmful substances and enters a standby usable state.

[0004] For example, the Chinese invention patent with the announcement number CN101975713B discloses a device and method for measuring the atomization rate of tobacco material, which comprises: a airflow channel and a compressed air source, a wind heating system, an atomization system and a real-time weighing system installed on the airflow channel. The measuring method is to place the tobacco material sample into the measuring device basket, and after the atomization rate of the liquid in the device, the air inlet speed, the liquid temperature and the air inlet temperature are stable, the basket is transferred to the electronic balance in the test channel of the device, and the computer connected thereto records the weight change of the tobacco material with time in real time.

[0005] However, in the process of implementing the technical scheme of the present application, the present application finds that the above-mentioned technology at least has the following technical problems:

[0006] In the prior art, due to the influence of the initial state of the electronic atomization liquid and the difference between the internal and external environments when the atomizer performs atomization, the atomizer cannot meet the energy balance required by the liquid heating and evaporation when performing a fixed liquid supply speed, so that the atomization effect deviates from the theoretical atomization state, and therefore there is a problem that the atomization quality inspection result of the electronic atomization liquid is not accurate. SUMMARY

[0007] To address the technical problem of inaccurate atomization quality inspection results for electronic atomizing fluids in existing technologies, this invention provides an intelligent atomization quality inspection method for electronic atomizing fluid production. The technical solution is as follows:

[0008] On the one hand, an intelligent atomization quality inspection method for electronic e-liquid production is provided. This method is implemented by an intelligent atomization quality inspection device for electronic e-liquid production. The method includes: S1, after the atomizer receives an atomization execution command, it acquires the atomization execution complexity parameters of the electronic e-liquid, analyzes and obtains the atomization execution complexity value, and adjusts the liquid supply speed of the atomizer accordingly. The atomization execution complexity value is used to represent the difficulty of atomization adjustment; S2, after initializing the camera device, it captures an atomization quality inspection image and analyzes it to obtain the image display quality value, thereby obtaining an image display quality judgment command. If the image display quality judgment command is unqualified, the background baseline map update frequency is adjusted, and after adjustment, an optimized atomization quality inspection image is captured again, and the device adjustment command is obtained. Otherwise, the background baseline map update is not executed. S3. Based on the analysis of the device adjustment command, if the device adjustment command is to adjust the camera device, then the corresponding camera device adjustment is performed; otherwise, the camera device adjustment is not performed. S4. The atomization quality inspection video is acquired through the camera device and input into the preset quality inspection model for initial quality inspection processing to obtain the fogging duration of the electronic atomizing liquid, thereby obtaining the atomization quality inspection judgment command of the electronic atomizing liquid. S5. Based on the analysis of the atomization quality inspection judgment command, if the atomization quality inspection judgment command is qualified, then the next round of multi-dimensional quality inspection is carried out to obtain the comprehensive quality inspection judgment command; otherwise, the liquid supply speed is adjusted a second time to obtain the atomization quality inspection secondary judgment command of the electronic atomizing liquid, thereby obtaining the comprehensive quality inspection judgment command.

[0009] On the other hand, an intelligent atomization quality inspection system for electronic e-liquid production is provided. This system includes: a liquid supply speed initial adjustment module, a camera configuration initial adjustment module, a camera equipment adjustment module, a multi-dimensional quality inspection analysis module, and a liquid supply speed secondary adjustment module. The liquid supply speed initial adjustment module is used to acquire the complex parameters of the electronic e-liquid's atomization execution after the atomizer receives the atomization execution command, analyze the atomization execution complexity value, and adjust the atomizer's liquid supply speed accordingly. The camera configuration initial adjustment module is used to initialize the camera equipment, capture atomization quality inspection images, analyze the image display quality value, and obtain an image display quality judgment command. If the image display quality judgment command is unqualified, the background baseline map update frequency is adjusted, and an optimized atomization quality inspection image is captured again after adjustment, and the equipment adjustment is analyzed. The system analyzes the device adjustment instructions. If the instruction is to adjust the camera equipment, the corresponding camera equipment adjustment is performed; otherwise, the camera equipment adjustment is not performed. The multi-dimensional quality inspection analysis module acquires atomization quality inspection video through the camera equipment and inputs it into a preset quality inspection model for initial quality inspection processing to obtain the atomization duration of the e-liquid, thereby obtaining the atomization quality inspection judgment instruction for the e-liquid. The secondary adjustment module for the liquid supply speed analyzes the atomization quality inspection judgment instruction. If the atomization quality inspection judgment instruction is qualified, the next round of multi-dimensional quality inspection is performed to obtain the comprehensive quality inspection judgment instruction; otherwise, the secondary adjustment of the liquid supply speed is performed to obtain the secondary atomization quality inspection judgment instruction for the e-liquid, thereby obtaining the comprehensive quality inspection judgment instruction.

[0010] On the other hand, an intelligent atomization quality inspection device for the production of electronic atomizing liquids is provided. The intelligent atomization quality inspection device for the production of electronic atomizing liquids includes: a processor; a memory, on which computer-readable instructions are stored. When the computer-readable instructions are executed by the processor, the intelligent atomization quality inspection method for the production of electronic atomizing liquids as described above is implemented.

[0011] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0012] 1. The intelligent atomization quality inspection method for electronic atomizing fluid production provided by this invention obtains the atomization execution complexity parameters of the electronic atomizing fluid and analyzes and calculates the atomization execution complexity value in combination with the historical sample matrix, thereby accurately analyzing the liquid supply speed adjustment requirements and realizing the dynamic optimization and adjustment of the liquid supply speed of the atomizer. This effectively solves the problem of inaccurate atomization quality inspection results of electronic atomizing fluid in the prior art.

[0013] 2. This invention obtains the display quality parameters of the fogging quality inspection images and extracts features to generate image feature vectors. It then matches and analyzes the real-time images with the theoretical set labeled in the database, thereby optimizing and adjusting the autofocus accuracy, gamma value, and brightness fusion ratio of the camera equipment. This effectively improves image clarity, color reproduction, and brightness balance, ensuring the accuracy of subsequent fogging quality inspection analysis while reducing interference from manual adjustments.

[0014] 3. By analyzing the duration of fogging in the atomization quality inspection video and processing the difference between the internal temperature of the atomizer and the ambient temperature, the atomization quality inspection status of the electronic atomizing fluid can be determined. Then, the liquid supply speed can be adjusted to increase or decrease the liquid supply speed to match the ideal atomization effect, effectively ensuring the stability of the atomization process and the continuity of fogging, and facilitating stable quality inspection. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 A flowchart illustrating the overall process of the intelligent atomization quality inspection method for electronic atomizing fluid production provided in this application embodiment;

[0017] Figure 2 A flowchart illustrating the thought process of the intelligent atomization quality inspection method for electronic atomizing fluid production provided in this application embodiment;

[0018] Figure 3 A flowchart illustrating the secondary adjustment of the liquid supply speed in an intelligent atomization quality inspection method for electronic atomizing liquid production, provided in an embodiment of this application.

[0019] Figure 4 This is a schematic diagram of the structure of an intelligent atomization quality inspection system for electronic atomizing fluid production provided in an embodiment of this application;

[0020] Figure 5 The first part of the real-time detection interface diagram of the intelligent atomization quality inspection system for electronic atomizing fluid production provided in the embodiments of this application;

[0021] Figure 6 The second part of the diagram shows the real-time detection interface of the intelligent atomization quality inspection system for electronic atomizing liquid production provided in this application embodiment. Detailed Implementation

[0022] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0023] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0024] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning.

[0025] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0026] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0027] like Figure 1The diagram shows a flowchart of an intelligent atomization quality inspection method for electronic atomizing fluid production provided in this application embodiment. The method includes the following steps: S1. After the atomizer receives an atomization execution command, it acquires the atomization execution complexity parameters of the electronic atomizing fluid, analyzes and obtains the atomization execution complexity value, and adjusts the atomizer's fluid supply speed accordingly. The atomization execution complexity value represents the difficulty of atomization adjustment. S2. After initializing the camera device, it captures an atomization quality inspection image and analyzes the image display quality value, thereby obtaining an image display quality judgment command. If the image display quality judgment command is unqualified, the background baseline map update frequency is adjusted, and an optimized atomization quality inspection image is captured again after adjustment. The device adjustment command is then analyzed; otherwise, the background baseline map update frequency adjustment is not performed. S3. Based on the analysis of the device adjustment command, if the device adjustment command is to adjust the camera device, then the corresponding camera device adjustment is performed; otherwise, the camera device adjustment is not performed. S4. The atomization quality inspection video is acquired through the camera device and input into the preset quality inspection model for initial quality inspection processing to obtain the fogging duration of the electronic atomizing liquid, thereby obtaining the atomization quality inspection ruling command for the electronic atomizing liquid. S5. Based on the analysis of the atomization quality inspection ruling command, if the atomization quality inspection ruling command is qualified, then the next round of multi-dimensional quality inspection is carried out to obtain the comprehensive quality inspection ruling command; otherwise, the liquid supply speed is adjusted a second time to obtain the secondary atomization quality inspection ruling command for the electronic atomizing liquid, thereby obtaining the comprehensive quality inspection ruling command.

[0028] In this embodiment, the image display quality adjudication instruction is obtained by: obtaining a preset image display quality threshold in the database and comparing it with the image display quality value; if the image display quality value is above the image display quality threshold, the image display quality adjudication instruction is qualified; otherwise, the image display quality adjudication instruction is unqualified, thereby adjusting the background baseline map update frequency.

[0029] The background baseline map update frequency is adjusted as follows: The image display quality threshold is subtracted from the image display quality value to obtain the image display quality difference. This difference is compared with historical image display quality differences stored in the database. The historical image display quality difference closest to this difference is used as the historical reference image display quality difference, and the corresponding historical background baseline map update frequency adjustment value is obtained as the first background baseline map update frequency adjustment value. A historical image display quality value closest to the current image display quality value is obtained and used as the second historical reference image display quality value. The corresponding historical background baseline map update frequency adjustment value is obtained as the second background baseline map update frequency adjustment value. The first and second background baseline map update frequency adjustment values ​​are averaged to obtain the final background baseline map update frequency adjustment value. Based on this adjustment value, the current background baseline map update frequency is increased (i.e., the current background baseline map update frequency is added to the adjusted value).

[0030] Video footage of the e-liquid's atomization process is acquired using camera equipment and input into a pre-set quality inspection model for initial quality control. This model determines the duration of fogging in the e-liquid. The model performs image processing and feature analysis on the input video data to quantitatively evaluate the atomization process. Specifically, after capturing video of the e-liquid's spraying process, the model first enhances the images and locates the target region in each frame, extracting features such as pixel brightness, droplet density, droplet size, and atomization coverage. Then, based on pre-set atomization performance indicators and rules, the model calculates and statistically analyzes these features to determine the duration of fogging in the e-liquid. This model set automatically converts video information into quantifiable quality inspection results, ensuring fast, accurate, and repeatable atomization performance evaluation.

[0031] It should be noted that the quality inspection model is constructed by processing image sequences of historical fogging videos and the corresponding fogging durations for each video. First, the video frames of each historical fogging video are preprocessed, including grayscale conversion, filtering and denoising, and contrast enhancement, to improve the recognizability of fogging details. Then, edge detection operators (such as Canny) are used to extract droplet boundaries and spray contours, and optical flow operators are used to analyze droplet movement trajectories between consecutive frames. Simultaneously, features such as fogging intensity distribution, droplet density, and duration for each frame are calculated, and the time-series features are normalized. Multidimensional quantitative indicators are constructed using these features, and regression analysis or supervised learning algorithms (such as weighted linear regression and support vector regression) are used to fit the training data, thus obtaining a quality inspection model capable of quantitatively evaluating the fogging process.

[0032] It should be noted that, based on the analysis of the atomization quality inspection ruling instructions for electronic atomizing fluid, if the atomization quality inspection ruling instruction is "quality inspection qualified" or "quality inspection unqualified", then the secondary adjustment of the fluid supply speed will not be executed; otherwise, the secondary adjustment of the fluid supply speed will be executed.

[0033] By initially adjusting the liquid supply rate, acquiring and analyzing atomization quality inspection images, adaptively adjusting the camera equipment, processing atomization quality inspection videos, and then further optimizing the liquid supply rate, multi-dimensional monitoring and precise control of the atomization process are achieved. Through feature extraction and database matching of image quality parameters, the camera equipment can be adaptively optimized to achieve high-precision quality inspection data acquisition. Analysis of atomization quality inspection videos using a preset quality inspection model accurately obtains the duration of fogging in the electronic atomizing liquid and compares it with a standard range for real-time quality inspection judgment. Secondary adjustments to the liquid supply rate are made through database matching of temperature effect differences and fogging duration differences, achieving dynamic optimization of the atomization flow rate.

[0034] like Figure 2 As shown, Figure 2 This is a flowchart illustrating the thought process of an intelligent atomization quality inspection method for electronic atomizing liquid production provided in this application embodiment. After the atomizer receives the atomization execution command, it obtains the atomization execution complexity value and adjusts the liquid supply speed of the atomizer accordingly. Atomization quality inspection images are captured by a camera device, and the image display quality value is calculated. Based on this quality value, an image display quality judgment command is generated. If the image display quality judgment command is "unqualified," the background baseline map update frequency is adjusted, and the atomization quality inspection image is re-captured, re-entering the quality judgment stage. If the image display quality judgment command is "qualified," an equipment adjustment command is generated to determine whether camera equipment adjustment is necessary. If the result is "yes," camera equipment adjustment is performed; if the result is "no," no adjustment is performed. The atomization quality inspection video is acquired by the camera device, and the fogging duration is calculated, generating an atomization quality inspection judgment command. If the atomization quality inspection ruling is "qualified", the process enters the multi-dimensional quality inspection process and obtains the comprehensive quality inspection ruling. If the atomization quality inspection ruling is "unqualified", the liquid supply speed is adjusted a second time and a secondary atomization quality inspection ruling is generated, ultimately obtaining the comprehensive quality inspection ruling, thus completing the process.

[0035] Furthermore, the atomization execution complexity value is obtained through the following method: The atomization execution complexity parameters of the e-liquid are acquired, including the initial temperature of the e-liquid, the near-core saturation of the e-liquid, and the specific heat capacity of the e-liquid; historical sample matrices and corresponding historical liquid supply adjustments are collected according to a preset time window, and sparse constraint terms and time difference smoothing terms are superimposed to construct an objective function based on the sum of squares of the deviations between the historical liquid supply adjustments and the historical sample matrices, and this objective function is labeled as the first objective function; the first objective function is solved using time-series regression and sparse smoothing methods to obtain an ideal time-series coefficient sequence, which characterizes the sensitivity of each atomization execution complexity parameter to the liquid supply adjustment amount under the current time window; the atomization execution complexity parameters at the current moment are acquired and substituted into the ideal time-series coefficient sequence for analysis to obtain the atomization execution complexity value.

[0036] In this embodiment, the atomization execution complexity value is obtained by the following method:

[0037] ;

[0038] ;

[0039] ;

[0040] In the formula, Denotes the first objective function. Let n represent the ideal time series coefficient sequence, t represent the time window index, t=1,2,...,T, T represent the total number of time windows, and n t This represents the number of historical samples collected within the t-th time window, where k represents the index of the historical sample within the time window, k=1,2,...,n t ,z t,k u represents the historical fluid supply adjustment amount corresponding to the k-th historical sample in time window t. t,k,r Let r represent the r-th feature component of the k-th historical sample within time window t, where r=1,2,3, where r=1 represents the initial temperature of the e-liquid, r=2 represents the near-core saturation temperature of the e-liquid, and r=3 represents the specific heat capacity of the e-liquid. The coefficient vector at time t, the r-th component Let u1 represent the linear sensitivity of the r-th feature to the output z at time t, u2 represent the time-series TV regularization hyperparameter, and a represent the time-series TV regularization hyperparameter. t * Represents the ideal time series coefficient sequence. , represents the coefficient vector at the current time, u new,r C represents the r-th standardized feature component of the newly acquired sample in real time. exec This indicates the complexity value of the atomization process.

[0041] By acquiring complex atomization parameters of electronic e-liquid, including initial temperature, near-core saturation, and specific heat capacity, and combining them with historical sample matrices and historical supply adjustment amounts, an objective function is constructed by superimposing sparse constraint terms and time difference smoothing terms. This objective function is then solved using time-series regression and sparse smoothing methods to obtain an ideal time-series coefficient sequence. Furthermore, the current complex atomization parameters are analyzed to obtain the atomization complexity value, thereby achieving a quantitative assessment of the difficulty of atomization adjustment. This accurately reflects the sensitivity of different atomization parameters to supply adjustment, providing a scientific basis for initial supply speed adjustments, avoiding atomization instability or uneven spraying caused by parameter changes, and improving the level of automation control in the atomization process and the stability of product atomization quality.

[0042] Furthermore, the liquid supply speed of the atomizer is adjusted by: obtaining the temperature difference between the inside and outside of the atomizer and matching it with the database to obtain the liquid supply speed influence coefficient; obtaining the liquid supply speed adjustment value based on the complexity value analysis of atomization execution; and performing multiplicative coupling processing (multiplication) on the liquid supply speed adjustment value and the liquid supply speed influence coefficient to obtain the comprehensive liquid supply speed adjustment value, thereby adjusting the liquid supply speed of the atomizer.

[0043] In this embodiment, the liquid supply rate influence coefficient is obtained by: acquiring the preset internal and external temperature difference of the atomizer in the database, and the internal and external temperature difference of each historical atomizer stored in the database, and comparing them with the internal and external temperature difference of the atomizer. The historical internal and external temperature difference of the atomizer that is closest to the internal and external temperature difference of the atomizer is taken as the historical reference internal and external temperature difference of the atomizer, and the historical liquid supply rate influence coefficient corresponding to the historical reference internal and external temperature difference of the atomizer is obtained. Thus, the historical liquid supply rate influence coefficient is taken as the liquid supply rate influence coefficient.

[0044] The liquid supply speed adjustment value is obtained based on the analysis of atomization execution complexity. Specifically, the atomization execution complexity value of the current atomizer is obtained, which represents the difficulty of adjusting the electronic atomizing fluid. Then, according to a preset time window, historical atomization execution complexity values ​​and corresponding actual liquid supply speed adjustment records are extracted from the database. The current complexity value is compared and analyzed with historical data to find the closest historical sample and its corresponding adjustment amount, thus obtaining the liquid supply speed adjustment value.

[0045] By acquiring the temperature difference between the inside and outside of the atomizer and matching it with a database, the liquid supply speed influence coefficient is obtained. Combined with the liquid supply speed adjustment value obtained based on the complexity value analysis of atomization execution, a comprehensive adjustment value for the liquid supply speed is generated through multiplicative coupling processing. This enables precise adjustment of the liquid supply speed of the atomizer. The liquid supply speed can be dynamically optimized according to the real-time temperature difference and the difficulty of atomization adjustment, so that the liquid supply volume is highly matched with the physical characteristics and environmental conditions of the electronic atomizing fluid, ensuring the stability and uniformity of the atomization process. At the same time, by precisely adjusting the liquid supply speed, insufficient or excessive atomization can be prevented, improving atomization quality and product consistency, ensuring the stable performance of the electronic atomizing fluid in different production batches, and thus improving production efficiency and product qualification rate.

[0046] Furthermore, the image display quality value is obtained through the following method: Image display quality parameters of the fogged quality inspection images are acquired and features are extracted to obtain image display feature vectors. These parameters include smoke edge sharpness, smoke color saturation, and brightness range. Pre-defined image quality annotation values ​​are obtained from the database. A second objective function is constructed by combining these image display feature vectors. By superimposing sparse and smooth constraints, the function is solved on the training sample set using elastic network regression to obtain an ideal weight vector. This ideal weight vector characterizes the contribution of various features to the image display quality. Finally, the image display quality parameters of the real-time acquired fogged quality inspection images are standardized to obtain the image display quality value.

[0047] In this embodiment, the image display quality value is obtained by the following method:

[0048] ;

[0049] ;

[0050] ;

[0051] ;

[0052] In the formula, J(w) represents the second objective function, and w * Let y represent the ideal weight vector, where i represents the frame number of the historical sample image, i=1,2,...,N, and N represents the total number of historical sample images. i This represents the image quality annotation value of the i-th sample image frame. Let f represent the weight vector to be solved, T represent the total number of time windows, and f i w represents the feature vector of the i-th historical sample image frame. j The elements of the weight vector are j=1, 2, 3, where j=1 represents the sharpness of the smoke edge, j=2 represents the color saturation of the smoke, and j=3 represents the brightness difference. This represents the mean square error term. Represents L1 regularization terms. Indicates L2 regularization term, This indicates that the minimum value is being sought during the training phase. Let s represent the ideal weight vector. raw Q represents the original linear score after weighting. img The values ​​represent the image display quality, α and β represent the calibration parameters of the mapping function, and μ... j Let σ represent the mean of the j-th feature in the training set. j f represents the standard deviation of the j-th feature in the training set. raw,new,j This represents the j-th feature value of a new image frame acquired in real-time online.

[0053] It should be noted that the training set data specifically consists of historical atomization quality inspection images.

[0054] The image display quality parameters of the fogging quality inspection images are obtained, and features are extracted to obtain image display feature vectors. Specifically, the raw image data of the fogging quality inspection images is acquired, and the images are preprocessed, including grayscale conversion, denoising, and image region localization. Then, various image display quality parameters are extracted from the preprocessed images. The smoke edge sharpness is calculated, the smoke color saturation is calculated by analyzing the image color channels, and the brightness range is calculated using the pixel brightness extreme value difference. These independent parameters are normalized and combined in a preset order to form image display feature vectors, creating quantifiable image feature representations that can be used for database comparison and subsequent analysis to evaluate the image quality of the fogging process.

[0055] By acquiring image display quality parameters from atomization quality inspection images and extracting features such as smoke edge sharpness, smoke color saturation, and brightness difference, an image display feature vector is formed. Further, combined with preset image quality annotation values ​​in the database, the feature vector is trained and solved using an elastic network regression method to obtain an ideal weight vector, which is used to quantify the contribution of various features to image display quality. Simultaneously, it enables scientific quantitative analysis of atomization quality inspection images, transforming subjective image quality judgments into quantifiable and reproducible indicators, improving the accuracy and stability of quality inspection data. Through feature analysis, key quality influencing factors in the image can be identified, providing a basis for adjusting camera equipment and controlling liquid supply, thereby optimizing the visualization of the atomization process and the product atomization quality, ensuring the reliability and consistency of the production process.

[0056] Furthermore, corresponding camera equipment adjustments are performed. Specifically, the image display quality parameters are compared with a preset theoretical set in the database using a difference ratio process to obtain various difference ratios. Based on these difference ratios, the camera equipment adjustment execution parameters and the camera equipment adjustment execution order are obtained, thus completing the camera equipment adjustment. The theoretical set includes theoretical values ​​for smoke edge sharpness, smoke color saturation, and brightness range. The difference ratios include the smoke edge sharpness difference ratio, smoke color saturation difference ratio, and brightness range difference ratio. The camera equipment adjustment execution parameters are obtained by analyzing these difference ratios. If the absolute value of the difference ratio of a certain image display quality parameter in a fogged quality inspection image is above the corresponding preset difference ratio threshold... If the absolute value of the difference ratio corresponds to an abnormal image display quality parameter, then the image display quality parameter corresponding to the abnormal image display quality parameter is marked as such. If the abnormal image display quality parameter is smoke edge sharpness, then based on the difference ratio analysis of smoke edge sharpness, if the difference ratio is positive, then the autofocus accuracy of the adjusted camera is reduced; otherwise, the autofocus accuracy of the adjusted camera is increased. If the abnormal image display quality parameter is smoke color saturation, then based on the difference ratio analysis of smoke color saturation, if the difference ratio is positive, then the gamma value of the adjusted camera is reduced; otherwise, the gamma value of the adjusted camera is increased. If the abnormal image display quality parameter is brightness difference, then the low-brightness blending ratio of the adjusted camera is increased, and the high-brightness blending ratio of the adjusted camera is also increased.

[0057] In this embodiment, the difference ratio is obtained by performing difference processing on the image display quality parameters and the preset theoretical set in the database. The specific method is as follows: the image display quality parameters are processed to obtain the difference value with the theoretical value corresponding to the image display quality parameters. The difference value is then divided by the theoretical value corresponding to the display quality parameters to obtain the difference ratio of the parameters corresponding to the display quality parameters. Thus, the difference ratios are obtained (which may be positive or negative. If it is positive, it means that the parameter is larger than the ideal value; otherwise, it means that the parameter is smaller. The absolute value of the difference ratio represents the degree of difference). For example, the smoke edge sharpness is processed to obtain the smoke edge sharpness difference value with the ideal value of smoke edge sharpness. The smoke edge sharpness difference ratio is then divided by the ideal value of smoke edge sharpness to obtain the smoke edge sharpness difference ratio.

[0058] If the difference ratio is positive, the autofocus accuracy of the adjusted camera is reduced; otherwise, the autofocus accuracy is increased. Specifically, the method involves matching the difference ratio of smoke edge sharpness with historical smoke edge sharpness difference ratios stored in the database. If a historical smoke edge sharpness difference ratio is closest to the current smoke edge sharpness difference ratio, this ratio is used as the historical reference smoke edge sharpness difference ratio. The corresponding historical autofocus accuracy adjustment amount is then obtained and used as the first autofocus accuracy adjustment. The system adjusts the autofocus accuracy by obtaining historical smoke edge sharpness stored in the database and comparing it with the current smoke edge sharpness. The historical smoke edge sharpness closest to the current smoke edge sharpness is taken as the historical reference smoke edge sharpness. The historical autofocus accuracy adjustment amount corresponding to the historical reference smoke edge sharpness is obtained as the second autofocus accuracy adjustment amount. The first and second autofocus accuracy adjustment amounts are then averaged to obtain the average value of the autofocus accuracy adjustment amount, which is used as the autofocus accuracy adjustment amount. This allows the system to either reduce or increase the autofocus accuracy of the camera.

[0059] If the difference ratio is positive, the gamma value of the adjusted camera is decreased; otherwise, the gamma value is increased. The specific method is as follows: The difference ratio of smoke color saturation is matched with historical smoke color saturation difference ratios stored in the database. If a historical smoke color saturation difference ratio is closest to the historical control smoke color saturation difference ratio, this ratio is used as the historical control smoke color saturation difference ratio. The corresponding historical gamma value adjustment amount is then obtained and applied. The first gamma value adjustment amount is obtained by acquiring the historical smoke color saturation stored in the database and comparing it with the current smoke color saturation. The historical smoke color saturation that is closest to the current smoke color saturation is taken as the historical reference smoke color saturation. The historical gamma value adjustment amount corresponding to the historical reference smoke color saturation is obtained as the second gamma value adjustment amount. The first gamma value adjustment amount and the second gamma value adjustment amount are then averaged to obtain the average gamma value adjustment amount, which is used as the gamma value adjustment amount. This allows for either reducing or increasing the gamma value of the camera device.

[0060] If the abnormal image display quality parameter is brightness difference, then the low-brightness blending ratio of the adjusted camera equipment and the high-brightness blending ratio of the adjusted camera equipment are increased. Specifically, the method is as follows: The difference ratio of the brightness difference is matched with the historical brightness difference difference ratios stored in the database. If a historical brightness difference difference ratio is closest to the current brightness difference ratio, this ratio is used as the historical reference brightness difference difference ratio. The corresponding historical low-brightness blending ratio adjustment and historical high-brightness blending ratio adjustment are obtained, and this historical low-brightness blending ratio adjustment is used as the first low-brightness blending ratio adjustment, and the historical high-brightness blending ratio adjustment is used as the first high-brightness blending ratio adjustment. The historical brightness differences stored in the database are then compared with the brightness differences. By comparison, the historical brightness range closest to the current brightness range is taken as the historical reference brightness range. The historical low-brightness blending ratio adjustment amount corresponding to this historical reference brightness range is obtained as the second low-brightness blending ratio adjustment amount. The historical high-brightness blending ratio adjustment amount corresponding to this historical reference brightness range is obtained as the second high-brightness blending ratio adjustment amount. The first low-brightness blending ratio adjustment amount and the second low-brightness blending ratio adjustment amount are then averaged to obtain the average low-brightness blending ratio adjustment amount, which is then used as the low-brightness blending ratio adjustment amount. The first high-brightness blending ratio adjustment amount and the second high-brightness blending ratio adjustment amount are then averaged to obtain the average high-brightness blending ratio adjustment amount, which is then used as the high-brightness blending ratio adjustment amount. This process is used to improve the low-brightness blending ratio and the high-brightness blending ratio of the camera device.

[0061] After obtaining the difference between the fogging execution complexity value and the reference fogging execution complexity value, the deviation direction of the current fogging state is first determined based on the sign of the difference. If the difference is positive, it indicates that the current fogging state is too strong, and the image detail capture intensity needs to be reduced to avoid over-focusing and causing image instability. At this time, the camera device is controlled to reduce the autofocus accuracy. If the difference is negative, it indicates that the current fogging state is too weak, and the image detail capture capability needs to be improved to ensure that the fogging state can be clearly identified. At this time, the camera device is controlled to increase the autofocus accuracy. The specific adjustment value for lowering or raising the autofocus accuracy of the camera equipment can be obtained by analyzing historical data in a database. For example, the database can be searched to find the record closest to the difference ratio, and the recommended focus accuracy value that made the image quality acceptable can be read from that record. If a perfectly matching record exists, the recommended value is directly used as the new focus accuracy. If no perfectly matching record exists, the optimal one-time target accuracy is obtained by weighting the recommended values ​​of two adjacent records (the specific method is as follows: extract the corresponding recommended focus accuracy values ​​from the two records, and calculate the weighting coefficient based on the distance between the current difference ratio and the difference ratio of the two records. The closer the record is, the greater the weight, and the farther the record is, the smaller the weight. Then, the recommended focus accuracy values ​​of the two records are linearly weighted and summed according to the weighting coefficient to obtain a unique target focus accuracy value as the optimal one-time target accuracy). The target accuracy (recommended focus accuracy value or optimal one-time target accuracy) is then directly sent to the camera device. The camera device immediately adjusts its internal focus control unit to the target accuracy and captures a frame for real-time verification. If the image quality meets the standard, the setting is confirmed to be effective and recorded in the database for subsequent optimization. If it does not meet the standard, the backup recommended value is called and sent again. If it still does not meet the standard, it reverts to the last stable accuracy and triggers a manual intervention prompt.

[0062] By performing difference ratio analysis between the image display quality parameters of atomized quality inspection images and a preset theoretical set in the database, various abnormal image display quality parameters are identified. For different anomaly types, key control parameters of the camera equipment, including autofocus accuracy, gamma value, and brightness fusion ratio, are adjusted to achieve adaptive optimization of the camera equipment. This allows for targeted adjustments to address different image quality anomalies such as smoke edge sharpness, smoke color saturation, and brightness differences, making image acquisition by the camera equipment more accurate and stable. This ensures that the clarity, color, and brightness of atomized quality inspection images meet expected standards, providing reliable data for subsequent atomized quality inspection analysis and improving the accuracy and automation level of image quality inspection in the electronic atomizing fluid production process.

[0063] Further, the camera device adjusts the execution order. The specific acquisition steps are as follows: perform an absolute value sorting based on each difference ratio to obtain the descending order of the absolute values of each difference ratio, and analyze the camera device adjustment execution order based on the descending order of the absolute values of each difference ratio and the abnormal image display quality parameters corresponding to each difference ratio.

[0064] In this embodiment, by arranging the absolute values of each difference ratio in descending order, the priority adjustment order of the abnormal image display quality parameters is determined. The purpose is to ensure that the parameters with the greatest impact on the image quality are adjusted first, so as to significantly improve the overall quality of the atomization quality inspection pictures in the shortest time. Through this order analysis, it is possible to avoid the mutual interference caused by adjusting multiple parameters simultaneously, improve the efficiency and accuracy of the camera device adjustment, quickly restore the image quality to the expected standard, and ensure the accurate and reliable visual monitoring of the atomization process.

[0065] Further, to obtain the atomization quality inspection ruling instruction for the e-liquid, the specific method is as follows: obtain the preset standard duration range of fogging in the database and compare it with the fogging duration of the e-liquid. If the fogging duration of the e-liquid is within the standard duration range of fogging, the atomization quality inspection ruling instruction for the e-liquid is qualified for inspection; obtain the internal temperature of the atomizer and the ambient temperature; if the fogging duration of the e-liquid is greater than the maximum value of the standard duration range of fogging and the internal temperature of the atomizer is greater than the ambient temperature, the atomization quality inspection ruling instruction for the e-liquid is abnormal for inspection and perform a secondary adjustment of the liquid supply speed; if the fogging duration of the e-liquid is greater than the maximum value of the standard duration range of fogging and the internal temperature of the atomizer is below the ambient temperature, the atomization quality inspection ruling instruction for the e-liquid is unqualified for inspection and do not perform a secondary adjustment of the liquid supply speed; if the fogging duration of the e-liquid is less than the minimum value of the standard duration range of fogging and the internal temperature of the atomizer is above the ambient temperature, the atomization quality inspection ruling instruction for the e-liquid is unqualified for inspection and do not perform a secondary adjustment of the liquid supply speed; if the fogging duration of the e-liquid is less than the minimum value of the standard duration range of fogging and the internal temperature of the atomizer is less than the ambient temperature, the atomization quality inspection ruling instruction for the e-liquid is abnormal for inspection and perform a secondary adjustment of the liquid supply speed.

[0066] In this embodiment, Figure 3This is the flowchart for the secondary adjustment of the liquid supply speed of the intelligent atomization quality inspection method for e-liquid production provided by the embodiments of this application. Obtain the internal temperature of the atomizer and the external ambient temperature, and calculate the temperature influence difference. Based on the temperature influence difference, match it with the database to obtain the secondary adjustment coefficient of the liquid supply speed; calculate the minimum fogging duration difference during the atomization process, and further match it with the database to obtain the secondary adjustment amount of the liquid supply speed. Compare the current fogging duration with the standard interval of the fogging duration: If the fogging duration is greater than the maximum value of the standard interval, reduce the liquid supply speed through the action of the secondary adjustment amount of the liquid supply speed and the adjustment coefficient; if the fogging duration is less than the minimum value of the standard interval, increase the liquid supply speed accordingly; if the fogging duration is within the standard interval range, do not perform any adjustment, thus completing the secondary adjustment of the liquid supply speed.

[0067] By obtaining the fogging duration of the e-liquid and comparing it with the preset standard interval of the fogging duration in the database, and at the same time obtaining the internal temperature of the atomizer and the ambient temperature, a scientific determination of the atomization quality inspection result of the e-liquid is achieved. When the fogging duration is within the standard interval, it is determined that the quality inspection is qualified, indicating that the atomization process is normal and the liquid supply speed does not need to be adjusted; when the fogging duration exceeds the standard interval and the internal temperature of the atomizer is higher than the ambient temperature or the deviation from the standard is lower than the ambient temperature, it is determined that the quality inspection is abnormal, and the secondary adjustment of the liquid supply speed is performed to compensate for the atomization abnormality caused by insufficient or excessive liquid supply; and when the fogging duration exceeds the standard interval but the temperature conditions of the atomizer and the ambient temperature are not suitable for compensation, it is determined that the quality inspection is unqualified and no adjustment is performed to avoid incorrect intervention. It can accurately identify the abnormality of the atomization process by combining the fogging duration and the environmental conditions, ensure that the liquid supply speed adjustment is only executed under reasonable conditions, thereby optimizing the atomization effect, ensuring the stability and consistency of the atomization quality of the e-liquid, and at the same time reducing the equipment load and production deviation caused by blind adjustment.

[0068] Further, the method for performing the secondary adjustment of the liquid supply speed is as follows: Perform a difference operation on the internal temperature of the atomizer and the ambient temperature to obtain the temperature influence difference; based on the temperature influence difference, match it with the database to obtain the secondary adjustment coefficient of the liquid supply speed; based on the difference operation between the fogging duration of the e-liquid and the standard interval of the fogging duration, obtain the minimum fogging duration difference, and based on the minimum fogging duration difference, match it with the database to obtain the secondary adjustment amount of the liquid supply speed; based on the comparison between the fogging duration of the e-liquid and the standard interval of the fogging duration, if the fogging duration of the e-liquid is greater than the maximum value of the standard interval of the fogging duration, reduce and adjust the liquid supply speed based on the secondary adjustment amount of the liquid supply speed and the secondary adjustment coefficient of the liquid supply speed. If the fogging duration of the e-liquid is less than the minimum value of the standard interval of the fogging duration, increase and adjust the liquid supply speed based on the secondary adjustment amount of the liquid supply speed and the secondary adjustment coefficient of the liquid supply speed, otherwise do not perform the liquid supply speed adjustment.

[0069] In this embodiment, it should be noted that when the quality inspection ruling for the e-liquid is an abnormality, a second adjustment to the liquid supply speed is performed. Although the industrial speed of the e-liquid after the second adjustment is the most reasonable (eliminating the influence of the external environment, etc.), problems may still occur because the e-liquid has already been burning for a period of time before the adjustment. Therefore, the quality inspection execution parameters (camera equipment parameters and liquid supply speed, etc.) in the next round of e-liquid quality inspection should be operated according to the adjusted parameters, rather than resetting the parameters each time.

[0070] The minimum fogging duration difference is obtained by processing the difference between the fogging duration of the e-liquid and the standard range of fogging duration. Specifically, if the fogging duration of the e-liquid is greater than the maximum value of the standard range of fogging duration, the minimum fogging duration difference is obtained by subtracting the fogging duration of the e-liquid from the maximum value of the standard range of fogging duration. If the fogging duration of the e-liquid is less than the minimum value of the standard range of fogging duration, the minimum fogging duration difference is obtained by subtracting the minimum value of the standard range of fogging duration from the fogging duration of the e-liquid.

[0071] The secondary adjustment coefficient of the liquid supply rate is obtained by matching the temperature difference with the database. The specific method is as follows: the real-time temperature of the current liquid supply environment is collected by a temperature sensor and compared with the historical temperature records stored in the database. The difference between the current temperature and the historical temperature is calculated, that is, the temperature difference. The temperature difference is then precisely matched with the database: the database stores liquid supply rate adjustment information corresponding to different temperature differences in advance, including the adjustment range. The matching process finds the historical record that is closest to the current temperature difference and determines the corresponding liquid supply rate adjustment scheme based on the matching result.

[0072] The secondary adjustment amount of the liquid supply rate is obtained by matching the minimum fogging duration difference with the database. Specifically, the method is as follows: Atomization sample data recorded in previous production processes is retrieved from the database, and a batch of historical samples with the highest similarity to the current atomization execution complex parameters and ambient temperature conditions is selected. Then, the fogging duration difference of these samples is curve-fitted with the corresponding liquid supply rate adjustment result to generate a liquid supply rate adjustment response curve. The current minimum fogging duration difference is then mapped onto this response curve, and the corresponding adjustment amount is directly read as the current secondary adjustment amount of the liquid supply rate. If the fitted curve does not have an exact matching point at the current difference, a transition value is calculated between adjacent sample points using local interpolation. This transition value is then used as the secondary adjustment amount of the liquid supply rate, and curve fitting is performed to generate a liquid supply rate adjustment response curve, specifically: In the formula, v adjIndicates the liquid supply rate, x new This represents the normalized difference in fogging duration, β=[β0,β1,β2]. ⊤ : The polynomial coefficient vector obtained from the fitting.

[0073] The liquid supply speed is reduced based on the secondary adjustment amount and the secondary adjustment coefficient of the liquid supply speed. The specific method is as follows: the reduced liquid supply speed is obtained by subtracting the product of the secondary adjustment amount and the secondary adjustment coefficient of the liquid supply speed from the current liquid supply speed.

[0074] The liquid supply speed is adjusted based on the secondary adjustment amount and the secondary adjustment coefficient of the liquid supply speed. The specific method is as follows: the liquid supply speed after adjustment is obtained by adding the product of the secondary adjustment amount and the secondary adjustment coefficient of the liquid supply speed to the current liquid supply speed.

[0075] By processing the temperature difference between the atomizer's internal temperature and the ambient temperature to obtain the temperature impact difference, and combining this with historical matching data from the database to obtain a secondary adjustment coefficient for the liquid supply rate, and simultaneously determining the secondary adjustment amount for the liquid supply rate based on the difference between the atomization duration of the e-liquid and the standard range, dynamic optimization of the atomizer's liquid supply rate is achieved. When the atomization duration of the e-liquid exceeds the upper limit of the standard range and the atomizer temperature is too high, the liquid supply rate is reduced to prevent over-atomization; when the atomization duration is below the lower limit of the standard range and the atomizer temperature is too low, the liquid supply rate is increased to enhance the atomization effect. Through this targeted secondary adjustment, the actual atomization state of the e-liquid can be accurately matched with the ideal standard, avoiding uneven atomization or quality deviations caused by ambient temperature differences or fluctuations in liquid properties, thereby improving atomization stability and product consistency, ensuring the entire production process is efficient and reliable, and reducing resource waste and the operational burden of repeated adjustments.

[0076] like Figure 4 As shown, Figure 4This is a schematic diagram of the intelligent atomization quality inspection system for electronic atomizing fluid production provided in this application embodiment. The system includes: a liquid supply speed initial adjustment module, a camera configuration initial adjustment module, a camera equipment adjustment module, a multi-dimensional quality inspection analysis module, and a liquid supply speed secondary adjustment module. The liquid supply speed initial adjustment module is used to obtain the complex parameters of the electronic atomizing fluid's atomization execution after the atomizer receives an atomization execution command, analyze the atomization execution complexity value, and thereby adjust the liquid supply speed of the atomizer. The camera configuration initial adjustment module is used to initialize the camera equipment, capture atomization quality inspection images, analyze the image display quality value, and obtain an image display quality judgment command. If the image display quality judgment command is unqualified, the background baseline map update frequency is adjusted, and a new atomization quality inspection image is captured after adjustment. The system optimizes the image for quality inspection and analyzes it to obtain equipment adjustment instructions; otherwise, the background baseline map update frequency adjustment will not be executed. The camera equipment adjustment module analyzes the equipment adjustment instructions; if the instruction is to execute camera equipment adjustment, the corresponding camera equipment adjustment is performed; otherwise, the camera equipment adjustment is not executed. The multi-dimensional quality inspection analysis module acquires atomization quality inspection video through the camera equipment and inputs it into a preset quality inspection model for initial quality inspection processing to obtain the fogging duration of the e-liquid, thereby obtaining the e-liquid atomization quality inspection judgment instruction. The secondary liquid supply speed adjustment module analyzes the atomization quality inspection judgment instruction; if the atomization quality inspection judgment instruction is qualified, the next round of multi-dimensional quality inspection is performed to obtain a comprehensive quality inspection judgment instruction; otherwise, the secondary liquid supply speed adjustment is executed to obtain a secondary atomization quality inspection judgment instruction for the e-liquid, thus obtaining a comprehensive quality inspection judgment instruction.

[0077] like Figure 5 and Figure 6 As shown, Figure 5 This is the first part of the real-time detection interface diagram of the intelligent atomization quality inspection system for electronic atomizing fluid production provided in the embodiments of this application. Figure 6 The second part of the real-time detection interface diagram of the intelligent atomization quality inspection system for electronic atomizing liquid production provided in this application embodiment shows the batch and number of the electronic atomizing liquid currently undergoing atomization quality inspection, as well as the operator. The system can adjust the atomization processing speed and atomizer execution parameters of the electronic atomizing liquid in real time according to the atomization status, thereby adjusting the atomization status of the electronic atomizing liquid and obtaining the actual continuous combustion time of the electronic atomizing liquid.

[0078] The intelligent atomization quality inspection device for electronic atomizing fluid production includes: a processor; and a memory containing computer-readable instructions.

[0079] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied 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.

[0080] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0081] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0082] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0083] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0084] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. An intelligent atomization quality inspection method for electronic atomization liquid production, characterized in that, The method comprises the following steps: S1, when the atomizer receives an atomization execution instruction, obtain an atomization execution complexity parameter of the electronic atomization liquid, analyze to obtain an atomization execution complexity value, and adjust the liquid supply speed of the atomizer according to the atomization execution complexity value, wherein the atomization execution complexity value is used to represent the difficulty of atomization adjustment; S2, after the camera equipment is initialized and configured, an atomization quality inspection picture is obtained by shooting, and a picture display quality value is obtained by analysis, and then a picture display quality ruling instruction is obtained, if the picture display quality ruling instruction is unqualified, the background baseline image update frequency is adjusted, and after the adjustment, an atomization quality inspection optimized image is obtained by shooting again, and a device adjustment instruction is obtained by analysis, otherwise, the background baseline image update frequency is not adjusted, wherein the picture display quality value is used to represent the clarity of the current atomization image to the atomization process display; S3, based on the device adjustment instruction analysis, if the device adjustment instruction is to execute the camera equipment adjustment, the corresponding camera equipment adjustment is performed, otherwise, the camera equipment adjustment is not performed; S4, the atomization quality inspection video is obtained by the camera equipment, and is input into the preset quality inspection model for initial quality inspection processing, to obtain the atomization duration of the electronic atomization liquid, and then the atomization quality inspection ruling instruction of the electronic atomization liquid is obtained; S5, based on the atomization quality inspection ruling instruction analysis, if the atomization quality inspection ruling instruction is qualified, the next round of multi-dimensional quality inspection is continued, and a comprehensive quality inspection ruling instruction is obtained, otherwise, the liquid supply speed is adjusted again, and a second atomization quality inspection ruling instruction of the electronic atomization liquid is obtained, and then the comprehensive quality inspection ruling instruction is obtained.

2. The method of claim 1, wherein the method is for electronic atomization liquid production. The atomization execution complexity value is obtained by the following method: The atomization execution complexity parameter of the electronic atomization liquid is obtained, wherein the atomization execution complexity parameter comprises the initial temperature of the electronic atomization liquid, the near-core saturation of the electronic atomization liquid and the specific heat capacity of the electronic atomization liquid; The historical sample matrix and the historical liquid supply adjustment amount corresponding to the historical sample matrix are collected according to the preset time window, and a sparse constraint term and a time difference smoothing term are superimposed, so as to construct a target function of the square sum of the deviation between the historical liquid supply adjustment amount and the historical sample matrix, and mark it as a first target function; The first target function is solved based on the time series regression and sparse smoothing method, to obtain an ideal time series coefficient sequence, wherein the ideal time series coefficient sequence is used to represent the sensitivity of each atomization execution complexity parameter to the liquid supply adjustment amount under the current time window; The atomization execution complexity value is obtained by analyzing the atomization execution complexity parameter at the current time and substituting it into the ideal time series coefficient sequence. 3.The method of claim 1, wherein the method comprises: determining a type of the electronic atomization liquid; and determining a type of the atomization device based on the type of the electronic atomization liquid. The liquid supply speed of the atomizer is adjusted by the following method: The temperature difference between the inside and outside of the atomizer is obtained, and is matched with the database to obtain a liquid supply speed influence coefficient; The liquid supply speed adjustment value is obtained based on the atomization execution complexity value analysis; The liquid supply speed comprehensive adjustment value is obtained by multiplicative coupling processing based on the liquid supply speed adjustment value and the liquid supply speed influence coefficient, so as to adjust the liquid supply speed of the atomizer.

4. The method of claim 1, wherein the method is for electronic atomization liquid production. The picture display quality value is obtained by the following method: The picture display quality parameter of the atomization quality inspection picture is obtained, and feature extraction is performed to obtain an image display feature vector, wherein the picture display quality parameter comprises the smoke edge sharpness, the smoke color saturation and the brightness range. The preset image quality annotation value in the database is acquired, a second target function is constructed in combination with the feature vectors of the images, and an elastic net regression method is used to solve the second target function on the training sample set by superimposing a sparse constraint term and a smooth constraint term, so as to obtain an ideal weight vector, which is used to represent the contribution degree of each type of feature to the image display quality. The image display quality value is obtained by standardizing the image display quality parameters of the atomization quality inspection picture acquired in real time.

5. The method of claim 1, wherein the method is for electronic atomization liquid production. The corresponding camera adjustment is performed, and the specific method is as follows: The image display quality parameters are subjected to difference ratio processing with the preset theoretical set in the database, to obtain each difference ratio, and the camera adjustment execution parameters and the camera adjustment execution sequence are obtained based on the difference ratios, so that the camera adjustment is completed; The theoretical set includes a smoke edge sharpness theoretical value, a smoke color saturation theoretical value and a brightness range theoretical value, and the difference ratios include a smoke edge sharpness difference ratio, a smoke color saturation difference ratio and a brightness range difference ratio; The camera adjustment execution parameters are obtained, and the specific method is as follows: Based on the difference ratio analysis, if the absolute value of the difference ratio of a certain image display quality parameter of the atomization quality inspection picture is greater than the corresponding preset difference ratio threshold, the image display quality parameter corresponding to the absolute value of the difference ratio is marked as an abnormal image display quality parameter, wherein if the abnormal image display quality parameter is the smoke edge sharpness, based on the difference ratio analysis of the smoke edge sharpness, if the difference ratio is positive, the automatic focusing accuracy of the camera adjustment is reduced, otherwise the automatic focusing accuracy of the camera adjustment is improved; If the abnormal image display quality parameter is the smoke color saturation, based on the difference ratio analysis of the smoke color saturation, if the difference ratio is positive, the gamma value of the camera adjustment is reduced, otherwise the gamma value of the camera adjustment is improved; If the abnormal image display quality parameter is the brightness range, the low-light fusion ratio of the camera adjustment is increased, and the high-light fusion ratio of the camera adjustment is increased.

6. The method of claim 5, wherein the method is for electronic atomization liquid production. The camera adjustment execution sequence is obtained, and the specific method is as follows: Based on the absolute numerical value of each difference ratio, an absolute numerical value descending order sequence of the difference ratios is obtained, and based on the absolute numerical value descending order sequence of the difference ratios and the abnormal image display quality parameters corresponding to the difference ratios, the camera adjustment execution sequence is obtained.

7. The method of claim 1, wherein the method is for electronic atomization liquid production. The atomization quality inspection judgment instruction of the electronic atomization liquid is obtained, and the specific method is as follows: A preset fogging duration standard interval in the database is acquired, and compared with the fogging duration of the electronic atomization liquid, if the fogging duration of the electronic atomization liquid is within the fogging duration standard interval, the atomization quality inspection judgment instruction of the electronic atomization liquid is qualified; The internal temperature of the atomizer and the environmental temperature are acquired; If the fogging duration of the electronic atomization liquid is greater than the maximum value of the fogging duration standard interval, and the internal temperature of the atomizer is greater than the environmental temperature, the atomization quality inspection judgment instruction of the electronic atomization liquid is abnormal, and the liquid supply speed is adjusted twice. If the atomization duration of the electronic atomized liquid is greater than the maximum value of the atomization duration standard interval and the temperature inside the atomizer is below the ambient temperature, the atomization quality inspection decision instruction of the electronic atomized liquid is unqualified, and the secondary adjustment of the liquid supply speed is not performed. If the atomization duration of the electronic atomized liquid is less than the minimum value of the atomization duration standard interval and the temperature inside the atomizer is above the ambient temperature, the atomization quality inspection decision instruction of the electronic atomized liquid is unqualified, and the secondary adjustment of the liquid supply speed is not performed. If the atomization duration of the electronic atomized liquid is less than the minimum value of the atomization duration standard interval and the temperature inside the atomizer is less than the ambient temperature, the atomization quality inspection decision instruction of the electronic atomized liquid is abnormal, and the secondary adjustment of the liquid supply speed is performed. 8.The method of claim 7, wherein the method comprises: determining whether the liquid is suitable for electronic atomization based on the at least one of the first information and the second information. The method for performing the secondary adjustment of the liquid supply speed is as follows: The temperature inside the atomizer is subtracted from the ambient temperature to obtain a temperature influence difference value; The temperature influence difference value is matched with the database to obtain a secondary adjustment coefficient of the liquid supply speed; The atomization duration of the electronic atomized liquid is subtracted from the atomization duration standard interval to obtain a minimum atomization duration difference value, and the minimum atomization duration difference value is matched with the database to obtain a secondary adjustment amount of the liquid supply speed; The atomization duration of the electronic atomized liquid is compared with the atomization duration standard interval. If the atomization duration of the electronic atomized liquid is greater than the maximum value of the atomization duration standard interval, the liquid supply speed is adjusted based on the secondary adjustment amount of the liquid supply speed and the secondary adjustment coefficient of the liquid supply speed. If the atomization duration of the electronic atomized liquid is less than the minimum value of the atomization duration standard interval, the liquid supply speed is adjusted based on the secondary adjustment amount of the liquid supply speed and the secondary adjustment coefficient of the liquid supply speed. Otherwise, the adjustment of the liquid supply speed is not performed. 9.The system for applying the method for intelligently inspecting atomization quality of electronic atomization liquid production according to any one of claims 1-8, wherein, It comprises: a primary adjustment module of the liquid supply speed, a primary adjustment module of the camera configuration, a camera equipment adjustment module, a multi-dimensional quality inspection analysis module, and a secondary adjustment module of the liquid supply speed; The primary adjustment module of the liquid supply speed is configured to obtain the atomization execution complex parameter of the electronic atomized liquid after the atomizer receives the atomization execution instruction, analyze the atomization execution complex value, and thereby adjust the liquid supply speed of the atomizer. The primary adjustment module of the camera configuration is configured to initialize the camera equipment, obtain the atomization quality inspection picture by shooting, and analyze the picture display quality value, thereby obtaining the picture display quality decision instruction. If the picture display quality decision instruction is unqualified, the background baseline image update frequency is adjusted, and the atomization quality inspection optimized image is obtained by shooting again after the adjustment. The equipment adjustment instruction is analyzed. Otherwise, the background baseline image update frequency adjustment is not performed. The camera equipment adjustment module is configured to analyze the equipment adjustment instruction. If the equipment adjustment instruction is to perform the camera equipment adjustment, the corresponding camera equipment adjustment is performed. Otherwise, the camera equipment adjustment is not performed. The multi-dimensional quality inspection analysis module is configured to obtain the atomization duration of the electronic atomized liquid by inputting the atomization quality inspection video obtained by the camera equipment into the preset quality inspection model for initial quality inspection processing, and thereby obtain the atomization quality inspection decision instruction of the electronic atomized liquid. The liquid supply speed secondary adjustment module is configured to, based on the atomization quality inspection adjudication instruction analysis, if the atomization quality inspection adjudication instruction is quality inspection qualified, continue the next round of multi-dimensional quality inspection to obtain a comprehensive quality inspection adjudication instruction, otherwise execute liquid supply speed secondary adjustment to obtain an atomization quality inspection secondary adjudication instruction of the electronic atomization liquid, thereby obtaining the comprehensive quality inspection adjudication instruction.

10. An intelligent atomization quality inspection device for electronic atomized liquid production, characterized in that, The intelligent atomization quality inspection device for electronic atomization liquid production comprises: a processor; a memory, wherein the memory has computer readable instructions stored thereon, and the computer readable instructions are executed by the processor to implement the method in any one of claims 1 to 8.

Images