A method and system for monitoring the appearance quality of finished RFID tag antennas

Abstract

The present invention relates to the technical field of quality monitoring, and specifically includes a method and system for monitoring the appearance quality of a finished RFID tag antenna, comprising: collecting point clouds and images to evaluate the printing quality of an antenna inside an RFID tag, identifying deformation and correcting to obtain a first antenna quality, predicting the deformation trend in combination with temperature and external force, and correcting to obtain a second antenna quality; analyzing the electrical quality change trend, and outputting an appearance quality monitoring result in combination with the second antenna quality information, so as to solve the technical problem that the appearance quality of the finished RFID tag antenna ignores the performance changes in the entire process from production to service, and the accuracy of the RFID antenna appearance quality monitoring is limited, so as to achieve quality monitoring of the entire process from production to service, evaluate the long-term change trend of the antenna printing quality, deformation, electrical performance, and environmental influences in actual application scenarios, improve the accuracy of the RFID antenna appearance quality monitoring, and reduce the risk of tag failure and data transmission errors caused by RFID defective products.

Description

Technical Field

[0001] The present invention relates to the technical field related to quality monitoring, and in particular to a method and system for monitoring the appearance quality of a finished product of an RFID tag antenna. Background Art

[0002] RFID, also known as electronic tag, is a communication technology that uses radio signals to identify specific targets and read and write related data. Since RFID technology does not require mechanical or optical contact between the identification system and the specific target, and has the advantages of high-speed identification, resistance to harsh environments and strong confidentiality, it has been widely used in logistics, warehouse management, medical and identity recognition, food processing, and real-time monitoring of production data and quality tracking in industrial manufacturing processes.

[0003] In RFID devices, antennas are the main energy transmission functional layer. They are electronic circuits designed according to the functions required by radio frequency identification. Conductive silver paste or conductive carbon paste is placed on PVC, PC or PET antenna substrates by etching, hot stamping or conductive ink printing, and then packaged with surface layer, protective layer and bottom layer. The processing quality of the antenna wiring pattern directly affects whether the signal can be transmitted between the tag and the reader to the greatest extent, and even affects the normal operation of the RFID device.

[0004] In summary, the prior art has the technical problems that the appearance quality of the finished RFID tag antenna ignores the performance changes in the entire process from production to service, and the accuracy of RFID antenna appearance quality monitoring is limited. Summary of the invention

[0005] The present application provides a method and system for monitoring the appearance quality of finished RFID tag antennas, aiming to solve the technical problems in the prior art that the appearance quality of finished RFID tag antennas ignores the performance changes throughout the entire process from production to service, and the accuracy of RFID antenna appearance quality monitoring is limited.

[0006] In view of the above problems, the technical solution to implement this application is:

[0007] In one aspect, the present application provides a method for monitoring the appearance quality of a finished RFID tag antenna, wherein the method comprises: performing scanning point cloud acquisition and image acquisition on the antenna in the RFID tag that has been produced and printed, to obtain an antenna point cloud and an antenna image;

[0008] Fitting is performed in the antenna point cloud to obtain the antenna fitting point cloud, and the antenna printing quality information is calculated;

[0009] Performing deformation recognition on the antenna image to obtain antenna deformation information, and performing correction calculation on the antenna printing quality information to obtain first antenna quality information;

[0010] According to the application scenario of the RFID tag, temperature information and external force information in the application scenario are collected, and combined with the antenna deformation information, the deformation trend of the antenna is predicted to obtain the antenna deformation trend information;

[0011] According to the antenna deformation trend information, a correction calculation is performed on the first antenna quality information to obtain second antenna quality information;

[0012] According to the temperature information and the antenna printing quality information, the antenna electrical quality change trend analysis is performed to obtain the antenna electrical quality information, and the antenna appearance quality monitoring result of the antenna in the RFID tag is output in combination with the second antenna quality information.

[0013] On the other hand, the present application provides a system for monitoring the appearance quality of a finished RFID tag antenna, wherein the system comprises: an acquisition module for scanning point cloud acquisition and image acquisition of the antenna in the RFID tag after production and printing, to obtain an antenna point cloud and an antenna image;

[0014] A point cloud fitting module is used to fit the antenna point cloud, obtain the antenna fitting point cloud, and calculate the antenna printing quality information;

[0015] A deformation recognition module, used to perform deformation recognition on the antenna image, obtain antenna deformation information, perform correction calculation on the antenna printing quality information, and obtain first antenna quality information;

[0016] A trend prediction module, for predicting the deformation trend of the antenna according to the application scenario of the RFID tag, collecting temperature information and external force information in the application scenario, and combining the antenna deformation information to obtain the antenna deformation trend information;

[0017] A correction calculation module, used to perform correction calculation on the first antenna quality information according to the antenna deformation trend information to obtain second antenna quality information;

[0018] The trend analysis module is used to perform antenna electrical quality change trend analysis based on the temperature information and antenna printing quality information, obtain antenna electrical quality information, and output the antenna appearance quality monitoring result of the antenna in the RFID tag in combination with the second antenna quality information.

[0019] In summary, one or more technical solutions provided in this application solve the technical problem that the appearance quality of the finished RFID tag antenna ignores the performance changes in the whole process from production to service, and the accuracy of RFID antenna appearance quality monitoring is limited. It realizes quality monitoring of the whole process from production to service, and evaluates the long-term change trend of antenna printing quality, deformation, electrical performance and environmental influences in actual application scenarios, thereby improving the accuracy of RFID antenna appearance quality monitoring and reducing the risk of tag failure and data transmission errors caused by defective RFID products. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 A possible flow chart of a method for monitoring the appearance quality of a finished product of an RFID tag antenna is provided for this application;

[0021] Figure 2 A schematic diagram of a possible flow chart for calculating and obtaining antenna printing quality information in a method for monitoring the appearance quality of a finished product of an RFID tag antenna is provided for this application;

[0022] Figure 3 A possible structural schematic diagram of a finished product appearance quality monitoring system for an RFID tag antenna is provided for this application.

[0023] Description of reference numerals: acquisition module M100 , point cloud fitting module M200 , deformation recognition module M300 , trend prediction module M400 , correction calculation module M500 , trend analysis module M600 . DETAILED DESCRIPTION

[0024] Embodiment 1

[0025] The present application is described in detail below in conjunction with the accompanying drawings. Figure 1 As shown, the present application provides a method for monitoring the appearance quality of a finished RFID tag antenna, wherein the method comprises:

[0026] S1: Scan the antenna in the printed RFID tag to collect point cloud and image, and obtain the antenna point cloud and antenna image;

[0027] In the production stage of RFID tag antenna products, common non-destructive testing includes: ultrasonic testing (using ultrasonic waves to penetrate antenna materials and judging whether there are cracks, inclusions, voids and other defects inside according to the reflected wave signal), magnetic particle testing (under the action of the magnetic field, magnetic particles gather at the magnetic discontinuities on the surface or near the surface of the antenna, showing defects such as cracks, folds, inclusions, etc.), eddy current testing (using the principle of electromagnetic field induction to detect defects on the surface and near the surface of conductive materials, suitable for surface quality testing of metal antennas); electrical performance testing includes: standing wave ratio (VSWR) measurement (using an antenna analyzer to measure the voltage standing wave ratio of the antenna input port to judge the matching degree between the antenna and the feeder, indirectly reflecting the overall performance of the antenna), gain and pattern measurement (in an anechoic chamber or open field, using directional receiving equipment and signal sources to measure the gain value and pattern distribution of the antenna at a specific frequency to evaluate the antenna radiation performance);

[0028] The above methods all focus on the detection in the factory state, which can ensure that the quality of the finished RFID tag antennas is up to standard, but the performance changes during the service of the antenna are limited. Since the RFID tag antenna is mostly used to be attached to the packaging surface, the label needs to be repeatedly peeled off from the packaging surface to which it is attached. In the process of peeling the label off from the packaging surface to which it is attached, it is inevitable that the frequent peeling action may cause physical damage such as scratches, creases, and wrinkles on the surface coating of the tag antenna, affecting the electrical performance of the antenna;

[0029] In short, factory inspection usually focuses on the initial state of the new tag antenna, but ignores the performance degradation caused by factors such as repeated peeling and sticking during the service life of the tag. Therefore, it is necessary to introduce more refined monitoring technologies and methods to accurately evaluate and manage the appearance quality of RFID tag antennas during service, and reduce the risk of tag failure and data transmission errors caused by the appearance of the finished RFID tag antenna.

[0030] Based on this, this application collects point clouds and images to conduct a refined assessment of the antenna printing quality, identify and correct the antenna deformation information, combine temperature and external force to predict the deformation trend, and further correct the antenna quality assessment. At the same time, it also conducts an in-depth analysis of the electrical quality change trend. Combining these multi-dimensional quality information, it outputs accurate and comprehensive appearance quality monitoring results, significantly improves the accuracy of RFID antenna appearance quality monitoring, effectively reduces the risk of tag failure and data transmission errors caused by RFID defective products, and provides strong quality assurance for the production and use of RFID tags;

[0031] Specifically, the basic components of the RFID tag include a tag, a reader and an antenna, wherein the tag is composed of a coupling element and a chip, each tag corresponds to a unique electronic code and is attached to an object to mark the target object; the reader is used to read or write tag information, and the antenna is used to transmit radio frequency signals between the tag and the reader;

[0032] The antenna in the RFID tag is printed with conductive ink to form a preset image. A dedicated 3D scanning device or sensor is used to scan the antenna in the RFID tag to collect antenna point cloud. Electrical testing can be used to scan point by point, and points with qualified electrical performance are used as data points to form a point cloud. Unqualified points (points that are not printed well or outside the pattern) are not used as data points. In this way, an antenna point cloud is formed. Each point in the antenna point cloud represents the physical position of a small area on the antenna.

[0033] Use a high-resolution camera to capture images of the antenna and obtain a two-dimensional visual image of the antenna. It is important to ensure that the shooting angle and lighting conditions are appropriate so as to clearly show the shape, color, texture and other appearance characteristics of the antenna.

[0034] S2: fitting the antenna point cloud to obtain the antenna fitting point cloud, and calculating the antenna printing quality information;

[0035] Furthermore, fitting is performed in the antenna point cloud to obtain the antenna fitting point cloud, and the antenna printing quality information is calculated, such as Figure 2 As shown, the present application method includes:

[0036] S21: obtaining a standard fitting anchor frame according to the design pattern of the antenna in the RFID tag;

[0037] S22: using the standard fitting anchor frame to perform iterative fitting in the antenna point cloud to obtain the antenna fitting point cloud;

[0038] S23: Calculate and obtain the antenna printing quality information based on the fitted antenna image formed by the antenna fitting point cloud and the standard fitting anchor frame.

[0039] In general, the collected antenna point cloud data is preprocessed to remove noise points and perform smoothing to improve data quality. According to the design drawings of the antenna in the RFID tag, the fitting parameters and constraints are set, and the original point cloud is fitted using a point cloud fitting algorithm (such as the least squares method, the ICP algorithm, etc.) to obtain an antenna fitting point cloud that is as close to the design model as possible. By comparing and analyzing the fitting point cloud with the design drawings of the antenna in the RFID tag, the deviation between the two (such as deformation error, dimensional error, etc.) is calculated, and combined with the preset printing quality evaluation standards (such as tolerance range, defect level, etc.), the printing quality information of the antenna is evaluated, including but not limited to: printing accuracy, flatness, and consistency.

[0040] The antenna fitting point cloud of the actual printing is obtained by fitting, that is, the antenna shape whose actual electrical performance meets the requirements, fitting is performed in the antenna point cloud to obtain the antenna fitting point cloud, and the antenna printing quality information is calculated. Specifically, according to the design drawings or digital models of the antenna in the RFID tag, the key features such as the expected shape, size, structure, etc. of the antenna are clarified. The design pattern should include the precise outline of the antenna, the direction of the internal circuit, the location of the connection point, etc.;

[0041] Based on the design pattern, create a standard fitting anchor frame. The anchor frame is usually a bounding box composed of a set of predefined reference points, which is aligned with the key feature points or boundaries of the design pattern. It serves as a reference for subsequent point cloud fitting. The design of the anchor frame should ensure that it can accurately reflect the main geometric features of the antenna.

[0042] Fitting to obtain the antenna fitting point cloud actually printed, based on the antenna fitting point cloud, combined with the preset standard shape, calculate the printing quality, further, based on the standard fitting anchor frame corresponding to the design pattern, pre-process the antenna fitting point cloud, including using a point cloud fitting algorithm (such as ICP, RANSAC, GMM, etc.), with the standard fitting anchor frame as a reference, iteratively fitting the original point cloud, the point cloud fitting algorithm adjusts the position of each point in the point cloud in each iteration, so that the overall shape of the point cloud is as close to the design pattern as possible, during the iteration process, by calculating the distance, angle and other error indicators between the point cloud and the anchor frame, evaluate the fitting effect, and adjust the fitting parameters according to the error feedback until the preset fitting accuracy or the upper limit of the number of iterations is reached;

[0043] After completing the iterative fitting, the antenna fitting point cloud that is highly consistent with the design pattern is obtained. The fitting point cloud can accurately reflect the actual printed shape of the antenna. After that, the fitting antenna image (a two-dimensional or three-dimensional image generated by the fitting point cloud) is compared and analyzed with the standard fitting anchor frame, and the key features in the fitting image (such as line width, spacing, corner angle, etc.) are compared with the corresponding features in the design pattern, and the corresponding deviations are calculated.

[0044] Furthermore, according to pre-set printing quality evaluation standards (such as tolerance range, defect level, etc.), the calculated deviation is converted into antenna printing quality indicators, including: printing accuracy (such as line width error, spacing error, etc.), flatness, consistency, number and severity of defects, etc.; various printing quality indicators are combined to calculate the overall printing quality information of the antenna. The antenna printing quality information reflects the influence of factors such as the process level, material properties, and equipment status of the antenna in the actual printing process on the quality of the final product, and provides data support for subsequent antenna deformation identification, electrical performance analysis, etc.

[0045] Furthermore, the standard fitting anchor frame is used to perform iterative fitting in the antenna point cloud. The method of the present application includes:

[0046] S221: using the standard fitting anchor frame to perform random fitting in the antenna point cloud to obtain a first fitting result;

[0047] S222: Obtain a proportion of antenna points falling within the first fitting result to obtain first fitting accuracy information;

[0048] S223: Continue to perform random fitting until the number of convergences is reached, and output a fitting result with the maximum fitting accuracy information as the antenna fitting point cloud.

[0049] Using the standard fitting anchor frame, iterative fitting is performed in the antenna point cloud. Specifically, a convergence condition is determined, for example, an upper limit of the number of iterations (for example, 100 times) and / or a fitting accuracy threshold is set as a criterion for judging whether convergence is reached; initialization parameters, exemplarily, initial parameters of random fitting are set, such as a search range, a step size, a random seed, etc., to ensure the repeatability and controllability of each random fitting process;

[0050] In the antenna point cloud, a standard fitting anchor frame is used for the first random fitting. The random fitting may involve transformations such as rotation, scaling, and translation of the anchor frame, so that the anchor frame matches the relevant features in the point cloud as much as possible. The proportion of antenna points that fall within the first fitting result is calculated. For example, the proportion is obtained by counting the number of points in the point cloud that are within the boundary of the fitting anchor frame or meet a certain distance threshold and dividing it by the total number of points, which reflects the consistency between the fitting result and the point cloud data, that is, the first fitting accuracy information.

[0051] Starting from the second iteration, random fitting is repeated, and the accuracy information of the corresponding fitting result (i.e., the proportion of antenna points falling within the fitting result) is calculated after each fitting; in each round of iteration, the current highest fitting accuracy information and its corresponding fitting result are saved, and if the accuracy of the new fitting result is higher than the saved best result, the best fitting result is updated; the fitting accuracy information is continuously monitored during the iteration process, and the iteration is stopped when the preset upper limit of the number of iterations (e.g., 100 times) is reached;

[0052] The fitting result with the maximum fitting accuracy information in the entire iterative process is output as the final antenna fitting point cloud. The antenna fitting point cloud has undergone multiple random fittings and optimizations, and the internal point distribution is closest to the standard fitting anchor frame, representing the best fitting state of the original antenna point cloud. In summary, through multiple random fittings and comparisons of accuracy information, it is possible to find the antenna point cloud configuration that best fits the standard fitting anchor frame under given convergence conditions, thereby obtaining the most accurate antenna fitting point cloud for subsequent printing quality evaluation.

[0053] S3: performing deformation recognition on the antenna image to obtain antenna deformation information, performing correction calculation on the antenna printing quality information to obtain first antenna quality information;

[0054] Furthermore, the antenna image is deformed and identified, antenna deformation information is obtained, and correction calculation is performed on the antenna printing quality information. The method of the present application includes:

[0055] S31: acquiring a sample antenna image set according to the antenna appearance detection data record, and identifying and marking the deformation degree of the antenna in the sample antenna image to obtain a sample antenna deformation information set;

[0056] S32: constructing an antenna deformation identifier by using the sample antenna image set and the sample antenna deformation information set, identifying the antenna image, and obtaining the antenna deformation information;

[0057] S33: performing classification decision in a deformation performance impact decision-maker according to the antenna deformation information to obtain a deformation performance impact coefficient, wherein the deformation performance impact decision-maker is constructed by a sample antenna deformation information set and a sample deformation performance impact coefficient set;

[0058] S34: using the deformation performance influence coefficient to perform correction calculation on the antenna printing quality information to obtain the first antenna quality information.

[0059] In general, the collected antenna images are processed (such as grayscale, filtering, edge detection, etc.) to enhance image features for subsequent deformation recognition. Image analysis algorithms (such as template matching, feature point detection, deep learning, etc.) are applied to the processed images for deformation recognition. Correspondingly, the recognition content includes deformation phenomena such as bending, twisting, local depression or bulge that may cause the antenna performance to deteriorate.

[0060] Antenna deformation information, to evaluate the specific impact of deformation on printing quality, it is necessary to consider factors such as deformation location, degree, direction, and potential impact on antenna electrical performance (such as impedance, gain, directivity, etc.); based on the deformation impact assessment result, the first antenna quality information is corrected and calculated to obtain a more accurate antenna quality assessment result, i.e., the first antenna quality information, which reflects the actual printing quality status of the antenna after considering the deformation factor;

[0061] As is known, antennas are generally printed on substrates, and deformation of the substrate will also affect the quality of the antenna. The antenna image is deformed to obtain antenna deformation information, and the antenna printing quality information is corrected and calculated. Specifically, a large number of antenna appearance inspection data records are collected, and representative antenna images are selected from them to form a sample antenna image set; the deformation of the antenna in the sample antenna image is identified and marked, including manual marking or using image marking tools to mark the deformation area, type (such as bending, twisting, local depression, etc.), degree and other information to form a sample antenna deformation information set;

[0062] Using a set of sample antenna images and a set of sample antenna deformation information as training data, an antenna deformation recognition model is trained. The antenna deformation recognition model may be an image classification or detection model based on deep learning, such as a convolutional neural network (CNN), YOLO, Faster R-CNN, etc. During the training process, the model learns the ability to extract features from antenna images and recognize deformations, and adjusts model parameters through back propagation and gradient optimization algorithms to gradually improve the recognition performance (such as accuracy, recall rate, F1 score, etc.) of the original model on the training set. After the training is completed, the obtained model is used as an antenna deformation identifier to perform deformation recognition on new antenna images.

[0063] Input the antenna image to be detected to the constructed antenna deformation identifier, the antenna deformation identifier extracts and analyzes the image features, and outputs the antenna deformation information, the antenna deformation information includes the deformation type, position, degree, etc.; use the pre-constructed deformation performance impact decision maker (such as a rule-based decision tree, support vector machine, logistic regression model, etc.), input the antenna deformation information, and make a classification decision. The deformation performance impact decision maker is trained based on a set of sample antenna deformation information and a set of sample deformation performance impact coefficients, and can judge the specific impact of the deformation information on the antenna performance according to the deformation information. The deformation performance impact decision maker outputs the deformation performance impact coefficient, and the deformation performance impact coefficient reflects the degree of influence of the antenna deformation on the electrical performance, mechanical strength and other key indicators of the antenna;

[0064] The obtained deformation performance influence coefficient is used to correct the original antenna printing quality information. The correction method can be to directly adjust the printing quality score, weight distribution, or re-evaluate the specific quality indicators (such as line width, spacing, consistency, etc.) according to the influence coefficient; after the correction calculation is completed, the first antenna quality information is obtained. The first antenna quality information comprehensively considers the antenna printing quality and the influence of deformation on the performance of the antenna, and more accurately reflects the actual quality status of the antenna.

[0065] The deformation recognition of antenna images, the acquisition of deformation information, and the correction calculation of antenna printing quality information provide more accurate data support for subsequent antenna quality assessment and maintenance decisions.

[0066] S4: According to the application scenario of the RFID tag, temperature information and external force information in the application scenario are collected, and combined with the antenna deformation information, the deformation trend of the antenna is predicted to obtain the antenna deformation trend information;

[0067] Furthermore, according to the application scenario of the RFID tag, the temperature information and external force information in the application scenario are collected, and the deformation trend of the antenna is predicted in combination with the antenna deformation information to obtain the antenna deformation trend information. The method of the present application includes:

[0068] S41: Acquire an application scenario of the RFID tag, and collect average application temperature information of the application scenario and average external force information received by the RFID tag;

[0069] S42: acquiring a sample temperature information set, a sample external force information set, and a sample antenna application deformation information set according to the application data record of the RFID tag, and processing and acquiring a sample antenna deformation trend information set according to the sample antenna application deformation set;

[0070] S43: Using the sample temperature information set, the sample external force information set and the sample antenna deformation trend information set, construct an antenna application deformation trend predictor, predict the average application temperature information and the average external force information, and obtain the antenna deformation trend information.

[0071] Overall, determine the actual application scenarios of RFID tags, such as warehousing logistics, retail commodity management, medical equipment tracking, etc., and understand the working environment characteristics of RFID tags (such as temperature range, humidity, vibration frequency, etc.); arrange appropriate sensor equipment in the application scenario to collect temperature information and external force information in real time. The temperature information can be obtained through temperature sensors, and the external force information can be monitored by pressure sensors, accelerometers and other equipment to monitor the pressure, vibration and other external forces on the tags;

[0072] The collected temperature information and external force information are combined with the antenna deformation information and input into the antenna application deformation trend predictor, which can be based on a machine learning algorithm trained with historical data, such as time series analysis, deep learning network, etc. The antenna application deformation trend predictor calculates the deformation trend of the antenna that may occur in the current and future period of time according to the input data, including information such as deformation degree, deformation direction, deformation speed, etc., and outputs the antenna deformation trend prediction result;

[0073] According to the application scenario, the deformation trend is predicted. For example, when RFID is used in cold chain transportation tags, low temperature and the extrusion of goods may cause the antenna to deform. According to the application scenario of the RFID tag, the temperature information and external force information in the application scenario are collected, and the deformation trend of the antenna is predicted in combination with the antenna deformation information to obtain the antenna deformation trend information. Specifically, the specific application scenario of the RFID tag is determined, such as warehouse management, logistics tracking, production line monitoring, etc., and the working environment characteristics, usage conditions and expected life of the RFID tag are clarified;

[0074] Deploy temperature sensors and pressure / force sensors in the application scenarios of RFID tags to collect the average application temperature information of the environment where the RFID tags are located and the average external force information of the RFID tags in real time or regularly. Temperature information may include ambient temperature, operating temperature, extreme temperature, etc., and external force information may include tension, pressure, torque, vibration, etc. In addition, in order to ensure the accuracy and stability of sensor data collection, it is necessary to calibrate the sensors regularly and handle sensor failures;

[0075] Collect data records of RFID tags in past applications, including temperature, external force, deformation and other information. The application data records of RFID tags come from sensor history records, maintenance logs, fault reports, etc.; Based on the collected historical data, construct a sample temperature information set, a sample external force information set and a sample antenna application deformation information set. The sample temperature information set, the sample external force information set and the sample antenna application deformation information set contain data points that are diverse and representative enough to reflect the performance of RFID tags under different working conditions;

[0076] Analyze the deformation information set of the sample antenna application to identify the development trend of antenna deformation, such as deformation rate, deformation direction, relationship between deformation and temperature / external force, etc.; based on the analysis results, process and extract the deformation trend information of the sample antenna, such as deformation curve, trend coefficient, key turning point, etc., to form a deformation trend information set of the sample antenna;

[0077] Using a sample temperature information set, a sample external force information set, and a sample antenna deformation trend information set as training data, an antenna application deformation trend predictor is trained. The antenna application deformation trend predictor is a time series prediction model, a regression model, a neural network model, etc. based on machine learning; during the training process, the ability to predict the antenna deformation trend from temperature and external force data is learned, and the model parameters are adjusted through back propagation and gradient optimization algorithms to gradually improve the prediction performance of the model on the training set; after the training is completed, the obtained model is used as an antenna application deformation trend predictor to predict future antenna deformation trends;

[0078] The current or expected average application temperature information and average external force information are input into the constructed antenna application deformation trend predictor. The predictor analyzes the input data and outputs the antenna deformation trend information. The obtained antenna deformation trend information reflects the possible deformation of the RFID tag antenna and the changing trend of the antenna deformation under the current or expected working conditions, providing an important basis for preventive maintenance, performance optimization, life prediction, etc.

[0079] S5: performing correction calculation on the first antenna quality information according to the antenna deformation trend information to obtain second antenna quality information;

[0080] Furthermore, according to the antenna deformation trend information, the first antenna quality information is corrected and calculated to obtain the second antenna quality information. The method of the present application includes:

[0081] S51: performing deformation prediction on the antenna deformation information according to the antenna deformation trend information to obtain predicted antenna deformation information;

[0082] S52: using the predicted antenna deformation information to input into the deformation performance impact decision maker to perform classification decision, obtain the predicted deformation performance impact coefficient, perform correction calculation on the first antenna quality information, and obtain the second antenna quality information.

[0083] In general, the first antenna quality information is corrected and calculated using the antenna deformation trend information. The specific method may include: combining the predicted deformation degree, direction and other parameters with the printing accuracy, flatness and other indicators in the first antenna quality information to evaluate the impact of the deformation on the antenna performance, thereby adjusting the first antenna quality information; after the correction calculation is completed, the second antenna quality information is obtained, and the second antenna quality information reflects the actual quality status of the antenna after considering the expected deformation impact;

[0084] According to the antenna deformation trend information, the first antenna quality information is corrected and calculated to obtain the second antenna quality information. Specifically, the acquired antenna deformation trend information is used in combination with current or expected working conditions to predict the future deformation of the antenna, and based on time series models, physical simulations, machine learning models and other methods, the predicted antenna deformation information is output, including deformation type, degree, time evolution, etc.;

[0085] The predicted antenna deformation information is input into a pre-built deformation performance impact decision maker, which is trained based on a set of sample antenna deformation information and a set of sample deformation performance impact coefficients, and can judge the specific impact of the deformation information on the antenna performance; the deformation performance impact decision maker makes a classification decision based on the predicted antenna deformation information, and outputs a predicted deformation performance impact coefficient, which reflects the degree of impact of future antenna deformation on key indicators such as its electrical performance and mechanical strength;

[0086] The predicted deformation performance influence coefficient is used to correct the first antenna quality information. The correction method can be to directly adjust the quality score, weight distribution, or re-evaluate specific quality indicators (such as line width, spacing, consistency, etc.) according to the influence coefficient, considering the impact of future deformation on the current quality; in the correction calculation process, it may be necessary to consider factors such as time windows, appropriately adjust the correction coefficient or introduce a confidence interval to reflect the impact of the prediction uncertainty on the quality assessment;

[0087] After the correction calculation is completed, the second antenna quality information is obtained. The second antenna quality information comprehensively considers the first antenna quality information and the impact of the predicted future deformation on its performance, and more accurately reflects the actual quality status of the antenna under future working conditions. The second antenna quality information is obtained by correcting the first antenna quality information based on the antenna deformation trend information, providing more accurate and forward-looking data support for the long-term performance management and maintenance decisions of RFID tags.

[0088] S6: Analyze the change trend of the antenna electrical quality according to the temperature information and the antenna printing quality information to obtain the antenna electrical quality information, and output the antenna appearance quality monitoring result of the antenna in the RFID tag in combination with the second antenna quality information.

[0089] Furthermore, according to the temperature information and the antenna printing quality information, the antenna electrical quality change trend analysis is performed to obtain the antenna electrical quality information. The present application method includes:

[0090] S61: Acquire a sample antenna printing quality information set of the antenna in the RFID tag, a sample temperature information set of the antenna application, and a sample antenna electrical quality information set after application;

[0091] S62: constructing an antenna electrical quality change predictor by using the sample antenna printing quality information set, the sample temperature information set and the sample antenna electrical quality information set, predicting the antenna printing quality information and the temperature information, and obtaining the antenna electrical quality information;

[0092] S63: obtaining a sample second antenna quality information set according to the antenna quality detection data record, and evaluating and obtaining a sample antenna appearance quality grade set in combination with the sample antenna electrical quality information set;

[0093] S64: Using the antenna electrical quality information and the second antenna quality information as data dimensions, constructing an antenna appearance quality classification coordinate system, inputting the sample antenna electrical quality information set and the sample second antenna quality information set into the antenna appearance quality classification coordinate system, obtaining a plurality of sample coordinate points, and using the sample antenna appearance quality level set for identification;

[0094] S65: using the antenna electrical quality information and the second antenna quality information, inputting them into the antenna appearance quality classification coordinate system for classification, and obtaining classification coordinate points;

[0095] S66: Calculate and obtain the antenna appearance quality monitoring result according to the K sample antenna appearance quality levels of the K sample coordinate points closest to the classification coordinate point, where K is an integer greater than or equal to 3.

[0096] In general, based on the temperature information and the antenna printing quality information, the electrical performance change trend of the antenna under different temperature conditions is analyzed, such as the change of parameters such as impedance, inductance, and quality factor; combined with the second antenna quality information, the impact of antenna structure deformation on electrical performance is evaluated, and the analysis results of electrical quality change trend are further improved; the second antenna quality information and the electrical quality change trend analysis results are integrated to comprehensively evaluate the appearance quality of the antenna in the RFID tag. From the application perspective, corresponding quality control strategies can also be formulated to extend the service life, such as adjusting the production process, improving the label design, etc.;

[0097] According to the temperature information and the antenna printing quality information, an antenna electrical quality change trend analysis is performed to obtain the antenna electrical quality information. Specifically, a sample antenna printing quality information set of the antenna in the RFID tag, a sample temperature information set of the antenna application, and a sample antenna electrical quality information set after application are collected. The sample antenna printing quality information set can be extracted from historical detection records. The sample antenna printing quality information set, the sample temperature information set, and the sample antenna electrical quality information set are used as training data to train an antenna electrical quality change prediction model. The antenna electrical quality change prediction model can be a time series prediction model based on machine learning, a regression model, a neural network model, etc.

[0098] During the training process, the model learns the ability to predict the electrical quality of the antenna from the printing quality information and temperature information. The model parameters are adjusted through back propagation and gradient optimization algorithms to improve the prediction performance of the model on the training set (such as mean square error, mean absolute error, R 2 After the training is completed, the obtained model is used as an antenna electrical quality change predictor to predict the future antenna electrical quality;

[0099] Input the current or expected antenna printing quality information and temperature information to the constructed antenna electrical quality change predictor. The predictor analyzes the input data and outputs antenna electrical quality information, such as impedance, resonant frequency, radiation efficiency, etc. The obtained antenna electrical quality information reflects the possible electrical performance of the RFID tag antenna under the current or expected working conditions, providing an important basis for performance optimization, life prediction, etc.

[0100] According to the antenna quality test data record, obtain the sample second antenna quality information set, and combine the sample antenna electrical quality information set to evaluate and obtain the sample antenna appearance quality grade set. The appearance quality grade may be formulated based on industry standards (GB / T 38889-2020 "Radio Interference Antenna Measurement Vehicle Antenna and System for Antennas and Receiving Systems" and other relevant standards), enterprise specifications, etc., reflecting the comprehensive performance of the antenna in terms of electrical performance, mechanical strength, environmental adaptability, etc.;

[0101] With the antenna electrical quality information and the second antenna quality information as data dimensions, the antenna appearance quality classification coordinate system is constructed. The antenna appearance quality classification coordinate system can be in the form of multi-dimensional space, radar chart, bar chart, etc., which is convenient for intuitively displaying the distribution and grade classification of antenna appearance quality; the antenna electrical quality information and the second antenna quality information of the antenna to be monitored are input into the constructed antenna appearance quality classification coordinate system to obtain the classification coordinate point; according to the K sample antenna appearance quality grades of the K sample coordinate points closest to the classification coordinate point, the appearance quality monitoring result of the antenna to be monitored is calculated, and the calculation method is such as average, median, weighted average, etc. In addition, the selection of K value should take into account the calculation accuracy and calculation complexity. The antenna electrical quality change trend analysis is performed on the temperature information and the antenna printing quality information to obtain the antenna electrical quality information, and the appearance quality grade evaluation and coordinate system calculation are combined to obtain the antenna appearance quality monitoring result, which provides support for the performance evaluation and quality control of RFID tags.

[0102] In summary, the beneficial effects of the embodiments of the present application are:

[0103] 1. Through scanning point cloud acquisition and image acquisition, combined with electrical performance testing, printing defects and unevenness can be effectively identified to improve the accuracy of quality inspection.

[0104] 2. Use image deformation recognition technology to monitor the deformation state of the antenna in actual use in real time, provide a basis for preventive maintenance, and reduce the risk of performance degradation or failure caused by deformation.

[0105] 3. Through the analysis of the electrical quality change trend of temperature and printing quality, the electrical performance of the antenna can be dynamically evaluated to provide support for quality control in the production process.

[0106] 4. The deformation prediction of the antenna deformation information is carried out according to the antenna deformation trend information to obtain the predicted antenna deformation information; the predicted antenna deformation information is input into the deformation performance impact decision device for classification decision making to obtain the predicted deformation performance impact coefficient, and the first antenna quality information is corrected and calculated to obtain the second antenna quality information. The correction calculation of the first antenna quality information according to the antenna deformation trend information to obtain the second antenna quality information provides more accurate and forward-looking data support for the long-term performance management and maintenance decision of the RFID tag.

[0107] Embodiment 2

[0108] Based on the same inventive concept as the method for monitoring the appearance quality of a finished RFID tag antenna in the aforementioned embodiment, Figure 3 As shown, the embodiment of the present application provides a system for monitoring the appearance quality of a finished product of an RFID tag antenna, wherein the system comprises:

[0109] The acquisition module M100 is used to scan the point cloud and image of the antenna in the RFID tag that has been printed, and obtain the antenna point cloud and antenna image;

[0110] The point cloud fitting module M200 is used to fit the antenna point cloud, obtain the antenna fitting point cloud, and calculate the antenna printing quality information;

[0111] The deformation recognition module M300 is used to perform deformation recognition on the antenna image, obtain antenna deformation information, perform correction calculation on the antenna printing quality information, and obtain first antenna quality information;

[0112] The trend prediction module M400 is used to predict the deformation trend of the antenna according to the application scenario of the RFID tag, collect temperature information and external force information in the application scenario, and obtain the antenna deformation trend information in combination with the antenna deformation information;

[0113] A correction calculation module M500, configured to perform correction calculation on the first antenna quality information according to the antenna deformation trend information to obtain second antenna quality information;

[0114] The trend analysis module M600 is used to perform antenna electrical quality change trend analysis based on the temperature information and antenna printing quality information, obtain antenna electrical quality information, and output the antenna appearance quality monitoring result of the antenna in the RFID tag in combination with the second antenna quality information.

[0115] Furthermore, the point cloud fitting module M200 is used to perform the following method:

[0116] According to the design pattern of the antenna in the RFID tag, a standard fitting anchor frame is obtained;

[0117] Using the standard fitting anchor frame, performing iterative fitting in the antenna point cloud to obtain the antenna fitting point cloud;

[0118] The antenna printing quality information is calculated based on the fitted antenna image formed by the antenna fitting point cloud in combination with the standard fitting anchor frame.

[0119] Furthermore, the point cloud fitting module M200 is also used to perform the following method:

[0120] Using the standard fitting anchor frame, performing random fitting in the antenna point cloud to obtain a first fitting result;

[0121] Obtaining a proportion of antenna points falling within the first fitting result to obtain first fitting accuracy information;

[0122] The random fitting is continued until the convergence times are reached, and the fitting result with the maximum fitting accuracy information is output as the antenna fitting point cloud.

[0123] Furthermore, the deformation recognition module M300 is used to perform the following method:

[0124] According to the antenna appearance detection data record, a sample antenna image set is obtained, and the deformation degree of the antenna in the sample antenna image is identified and marked to obtain a sample antenna deformation information set;

[0125] Using the sample antenna image set and the sample antenna deformation information set, constructing an antenna deformation identifier, identifying the antenna image, and obtaining the antenna deformation information;

[0126] According to the antenna deformation information, a classification decision is made in a deformation performance impact decision device to obtain a deformation performance impact coefficient, wherein the deformation performance impact decision device is constructed by a sample antenna deformation information set and a sample deformation performance impact coefficient set;

[0127] The deformation performance influence coefficient is used to perform correction calculation on the antenna printing quality information to obtain the first antenna quality information.

[0128] Furthermore, the trend prediction module M400 is used to perform the following method:

[0129] Acquire an application scenario of the RFID tag, and collect average application temperature information of the application scenario and average external force information received by the RFID tag;

[0130] According to the application data record of the RFID tag, a sample temperature information set, a sample external force information set and a sample antenna application deformation information set are obtained, and according to the sample antenna application deformation set, a sample antenna deformation trend information set is obtained by processing;

[0131] The sample temperature information set, the sample external force information set and the sample antenna deformation trend information set are used to construct an antenna application deformation trend predictor, and the average application temperature information and the average external force information are predicted to obtain the antenna deformation trend information.

[0132] Furthermore, the correction calculation module M500 is used to perform the following method:

[0133] According to the antenna deformation trend information, deformation prediction is performed on the antenna deformation information to obtain predicted antenna deformation information;

[0134] The predicted antenna deformation information is input into the deformation performance impact decision maker to perform classification decision, obtain the predicted deformation performance impact coefficient, perform correction calculation on the first antenna quality information, and obtain the second antenna quality information.

[0135] Furthermore, the trend analysis module M600 is used to perform the following method:

[0136] Obtaining a sample antenna printing quality information set of the antenna in the RFID tag, a sample temperature information set of the antenna application, and a sample antenna electrical quality information set after application;

[0137] Using the sample antenna printing quality information set, the sample temperature information set and the sample antenna electrical quality information set, constructing an antenna electrical quality change predictor, predicting the antenna printing quality information and temperature information, and obtaining the antenna electrical quality information;

[0138] According to the antenna quality detection data record, a sample second antenna quality information set is obtained, and combined with the sample antenna electrical quality information set, an evaluation is performed to obtain a sample antenna appearance quality grade set;

[0139] Taking the antenna electrical quality information and the second antenna quality information as data dimensions, constructing an antenna appearance quality classification coordinate system, inputting the sample antenna electrical quality information set and the sample second antenna quality information set into the antenna appearance quality classification coordinate system, obtaining a plurality of sample coordinate points, and using the sample antenna appearance quality grade set for identification;

[0140] Using the antenna electrical quality information and the second antenna quality information, inputting them into the antenna appearance quality classification coordinate system for classification, and obtaining classification coordinate points;

[0141] The antenna appearance quality monitoring result is obtained by calculation according to the K sample antenna appearance quality levels of the K sample coordinate points closest to the classification coordinate point, where K is an integer greater than or equal to 3.

[0142] In summary, any step can be stored as a computer instruction or program in an unlimited computer memory, and can be called and recognized by an unlimited computer processor, and no unnecessary restrictions are made here.

[0143] Furthermore, the above-mentioned technical scheme only reflects the preferred technical scheme of the technical scheme of the embodiment of the present application. Some changes that may be made to some parts thereof by technicians in this technical field all reflect the novel principles of the embodiment of the present application. Obviously, technicians in this field can make various changes and modifications to the present application without departing from the scope of the present application.

Claims

1. A method for monitoring the appearance quality of a finished RFID tag antenna, characterized in that: The method comprises: Scan the antenna in the printed RFID tag to collect point cloud and image, and obtain the antenna point cloud and antenna image; Fitting is performed in the antenna point cloud to obtain the antenna fitting point cloud, and the antenna printing quality information is calculated; Performing deformation recognition on the antenna image to obtain antenna deformation information, and performing correction calculation on the antenna printing quality information to obtain first antenna quality information; According to the application scenario of the RFID tag, temperature information and external force information in the application scenario are collected, and combined with the antenna deformation information, the deformation trend of the antenna is predicted to obtain the antenna deformation trend information; According to the antenna deformation trend information, a correction calculation is performed on the first antenna quality information to obtain second antenna quality information; Perform antenna electrical quality change trend analysis based on the temperature information and antenna printing quality information to obtain antenna electrical quality information, and output antenna appearance quality monitoring results of the antenna in the RFID tag in combination with the second antenna quality information; The method includes performing fitting in the antenna point cloud, obtaining the antenna fitting point cloud, and calculating and obtaining the antenna printing quality information, including: According to the design pattern of the antenna in the RFID tag, a standard fitting anchor frame is obtained; Using the standard fitting anchor frame, performing iterative fitting in the antenna point cloud to obtain the antenna fitting point cloud; The antenna printing quality information is calculated based on the fitted antenna image formed by the antenna fitting point cloud and combined with the standard fitting anchor frame; Performing deformation recognition on the antenna image, obtaining antenna deformation information, and performing correction calculation on the antenna printing quality information, including: According to the antenna appearance detection data record, a sample antenna image set is obtained, and the deformation degree of the antenna in the sample antenna image is identified and marked to obtain a sample antenna deformation information set; Using the sample antenna image set and the sample antenna deformation information set, constructing an antenna deformation identifier, identifying the antenna image, and obtaining the antenna deformation information; According to the antenna deformation information, a classification decision is made in a deformation performance impact decision device to obtain a deformation performance impact coefficient, wherein the deformation performance impact decision device is constructed by a sample antenna deformation information set and a sample deformation performance impact coefficient set; The deformation performance influence coefficient is used to perform correction calculation on the antenna printing quality information to obtain the first antenna quality information.

2. The method according to claim 1, characterized in that Using the standard fitting anchor frame, performing iterative fitting in the antenna point cloud, including: Using the standard fitting anchor frame, performing random fitting in the antenna point cloud to obtain a first fitting result; Obtaining a proportion of antenna points falling within the first fitting result to obtain first fitting accuracy information; The random fitting is continued until the convergence times are reached, and the fitting result with the maximum fitting accuracy information is output as the antenna fitting point cloud.

3. The method according to claim 1, characterized in that According to the application scenario of the RFID tag, temperature information and external force information in the application scenario are collected, and the deformation trend of the antenna is predicted in combination with the antenna deformation information to obtain the antenna deformation trend information, including: Acquire an application scenario of the RFID tag, and collect average application temperature information of the application scenario and average external force information received by the RFID tag; According to the application data record of the RFID tag, a sample temperature information set, a sample external force information set and a sample antenna application deformation information set are obtained, and according to the sample antenna application deformation set, a sample antenna deformation trend information set is obtained by processing; The sample temperature information set, the sample external force information set and the sample antenna deformation trend information set are used to construct an antenna application deformation trend predictor, and the average application temperature information and the average external force information are predicted to obtain the antenna deformation trend information.

4. The method according to claim 1, characterized in that: According to the antenna deformation trend information, the first antenna quality information is corrected and calculated to obtain second antenna quality information, including: According to the antenna deformation trend information, deformation prediction is performed on the antenna deformation information to obtain predicted antenna deformation information; The predicted antenna deformation information is input into the deformation performance impact decision maker to perform classification decision, obtain the predicted deformation performance impact coefficient, perform correction calculation on the first antenna quality information, and obtain the second antenna quality information.

5. The method according to claim 1, characterized in that According to the temperature information and the antenna printing quality information, an antenna electrical quality change trend analysis is performed to obtain the antenna electrical quality information, including: Obtaining a sample antenna printing quality information set of the antenna in the RFID tag, a sample temperature information set of the antenna application, and a sample antenna electrical quality information set after application; Using the sample antenna printing quality information set, the sample temperature information set and the sample antenna electrical quality information set, constructing an antenna electrical quality change predictor, predicting the antenna printing quality information and temperature information, and obtaining the antenna electrical quality information; According to the antenna quality detection data record, a sample second antenna quality information set is obtained, and combined with the sample antenna electrical quality information set, an evaluation is performed to obtain a sample antenna appearance quality grade set; Taking the antenna electrical quality information and the second antenna quality information as data dimensions, constructing an antenna appearance quality classification coordinate system, inputting the sample antenna electrical quality information set and the sample second antenna quality information set into the antenna appearance quality classification coordinate system, obtaining a plurality of sample coordinate points, and using the sample antenna appearance quality grade set for identification; Using the antenna electrical quality information and the second antenna quality information, inputting them into the antenna appearance quality classification coordinate system for classification, and obtaining classification coordinate points; The antenna appearance quality monitoring result is obtained by calculation according to the K sample antenna appearance quality levels of the K sample coordinate points closest to the classification coordinate point, where K is an integer greater than or equal to 3.

6. A system for monitoring the appearance quality of finished RFID tag antennas, characterized in that: A method for monitoring the appearance quality of a finished RFID tag antenna according to any one of claims 1 to 5, the system comprising: The acquisition module is used to scan the point cloud and image of the antenna in the RFID tag that has been printed, and obtain the antenna point cloud and antenna image; A point cloud fitting module is used to fit the antenna point cloud, obtain the antenna fitting point cloud, and calculate the antenna printing quality information; A deformation recognition module, used to perform deformation recognition on the antenna image, obtain antenna deformation information, perform correction calculation on the antenna printing quality information, and obtain first antenna quality information; A trend prediction module, for predicting the deformation trend of the antenna according to the application scenario of the RFID tag, collecting temperature information and external force information in the application scenario, and combining the antenna deformation information to obtain the antenna deformation trend information; A correction calculation module, used to perform correction calculation on the first antenna quality information according to the antenna deformation trend information to obtain second antenna quality information; The trend analysis module is used to perform antenna electrical quality change trend analysis based on the temperature information and antenna printing quality information, obtain antenna electrical quality information, and output the antenna appearance quality monitoring result of the antenna in the RFID tag in combination with the second antenna quality information.

Images