Identification detection method and system for radio frequency connector

Abstract

The invention provides an identification detection method and system for a radio frequency connector, and the method comprises the steps: obtaining the initial appearance image data of the surface of the radio frequency connector, and carrying out the processing of the image data, and obtaining a first appearance image; extracting an appearance feature set and an identification area image from the first appearance image; performing character segmentation and feature extraction on the identification area image, obtaining a preliminary manufacturing information text in combination with a pre-established text recognition model, verifying the preliminary manufacturing information text and associating the preliminary manufacturing information text with the appearance feature set to generate an associated data set; comparing the appearance feature set with a preset standard appearance database to generate defect marking data, fusing the associated data set with the defect marking data to generate comprehensive analysis data, and generating a visual chart according to the comprehensive analysis data; and further analyzing performance index change and defect severity according to the visual chart, generating risk assessment data, and generating a final detection analysis file according to the risk assessment data.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of information technology, and in particular to a recognition detection method and system for a radio frequency connector. BACKGROUND

[0002] As an indispensable core component in modern communication and electronic equipment, the radio frequency connector plays a crucial role in ensuring signal transmission quality and system stability. Its performance directly affects the operational efficiency and safety of multiple critical areas such as communication networks, aerospace, medical devices, etc. Therefore, accurate detection and performance evaluation of radio frequency connectors have become a top priority in the industry.

[0003] However, current mainstream detection methods rely heavily on manual inspection and basic testing tools, which not only have low efficiency but are also susceptible to human factors, making it difficult to ensure the accuracy and consistency of the detection results. This traditional approach often falls short when faced with complex and diverse connector types and large-scale production demands, making it difficult to meet the requirements of modern industry for high precision and high efficiency.

[0004] In this context, the field faces significant technical challenges. The first and foremost is how to quickly obtain the appearance features and manufacturing information of the connector through automated means, replacing the inefficient mode of manual identification. Due to the failure to effectively address this issue, it further leads to difficulties in fully covering the geometric parameters and working environment information of the connector during data collection, thereby affecting the accurate analysis and evaluation of performance indicators. These challenges are interconnected and collectively constitute a technical bottleneck in the intelligentization and comprehensiveness of the detection system, necessitating innovative means to break through.

[0005] Therefore, how to build an automated method and system based on image recognition and data analysis to efficiently obtain the appearance features and manufacturing information of the radio frequency connector has become a key issue in current research. SUMMARY

[0006] In view of the deficiencies of the prior art, the present application provides a recognition detection method and system for a radio frequency connector, which realizes accurate detection and risk assessment of the appearance defects of the radio frequency connector, and provides effective support for product quality control.

[0007] To achieve the above purpose, the technical solution of the present application is as follows: A recognition detection method for a radio frequency connector, the method comprising the following steps: Step 1, acquiring initial appearance image data of the surface of the radio frequency connector through an image acquisition device, and processing the initial appearance image data to obtain a first appearance image; Step 2, extracting an appearance feature set and an identification area image from the first appearance image; Step 3, character segmentation and feature extraction are performed on the identification area image, a preliminary manufacturing information text is obtained by combining a pre-established text recognition model, the preliminary manufacturing information text is checked and associated with the appearance feature set, and a first associated data set is generated; Step 4, the appearance feature set is compared with a preset standard appearance database to generate defect marking data, the first associated data set is fused with the defect marking data to generate comprehensive analysis data, and a first visualization chart is generated according to the comprehensive analysis data for presenting defect distribution and trend relationship; Step 5, according to the first visualization chart, further analyze the performance index change and the defect severity, generate risk assessment data, and generate a final detection analysis archive according to the risk assessment data for archiving and feedback.

[0008] Preferably, the initial appearance image data of the radio frequency connector surface is obtained by the image acquisition device, and the initial appearance image data is processed to obtain the first appearance image, comprising: The surface of the radio frequency connector is scanned by a high-resolution image acquisition device to obtain initial appearance image data. For the initial appearance image data, a light correction processing technique is used to adjust the uneven brightness area. Through the brightness correction of the initial appearance image data, the uniformity of the image data is ensured. According to the corrected initial appearance image data, it is determined whether the definition of the image meets the preset requirements. If not, image acquisition and processing are performed again, and the first appearance image with quality meeting the requirements is obtained through multiple processing iterations.

[0009] Preferably, the appearance feature set and the identification area image are extracted from the first appearance image, comprising: For the first appearance image, a character region positioning method is used to identify the surface manufacturing information identification area, determine the identification area image, and perform region segmentation on the identification area image to separate the specific range of manufacturing information identification, and generate clear identification area image data; The edge contour feature in the first appearance image is extracted by an edge detection technique, the geometric feature data of the surface defect is determined according to the edge contour feature, and the edge contour feature and the geometric feature data are integrated into the appearance feature set; According to the appearance feature set and the clear identification area image data, a feature mapping relationship is established for subsequent defect comparison and text analysis.

[0010] Preferably, the character segmentation and feature extraction are performed on the identification area image, a preliminary manufacturing information text is obtained by combining a pre-established text recognition model, the preliminary manufacturing information text is checked and associated with the appearance feature set, and a first associated data set is generated, comprising: According to the identification area image, a character segmentation algorithm is used to split the characters connected together into single character units, and through a character feature extraction technology, stroke width and spacing feature data of the single character units are obtained; according to the stroke width and spacing feature data, a pre-established text recognition model is combined for classification matching, and through the classification matching result, a preliminary manufacturing information text is generated; According to the preliminary manufacturing information text, a text format verification method is used to verify whether the batch number and the production date conform to the preset coding rule, and through the verification result, the error data that does not conform to the rule is removed; according to the text data after verification, a data correlation mapping technology is used to uniquely bind with the appearance feature set, and through the binding result, a first correlation data set is generated.

[0011] Preferably, the comparison of the appearance feature set with the preset standard appearance database to generate defect marking data comprises: According to the appearance feature set, a template matching method is used to match with the preset standard appearance database, and through the matching result, a deviation value of the appearance feature set and the standard data is calculated, if the deviation value exceeds a preset threshold, it is judged as an appearance defect, and according to the judgment result, defect marking data is generated.

[0012] Preferably, the fusion of the first correlation data set and the defect marking data to generate comprehensive analysis data, and the generation of a first visualization chart according to the comprehensive analysis data comprises: According to the first correlation data set and the defect marking data, a multi-dimensional data superposition technology is used to perform feature matching on the appearance defect information in the defect marking data and the manufacturing information in the first correlation data set, to determine a preliminary fused data structure, through a data structure analysis tool, field mapping processing is performed on the preliminary fused data structure, to obtain intermediate layer data containing defect positions and batch numbers, if there is a field missing or format inconsistency in the intermediate layer data, then through a data cleaning algorithm, the missing values are filled and the format is unified, the standardized data that meets the requirements is determined, according to the standardized data, a data integration framework is used to deeply associate the defect positions and the batch numbers, to obtain a basic version of the comprehensive analysis data, through a data verification mechanism, the basic version of the comprehensive analysis data is detected for integrity and accuracy, to determine a reliable data set after no error, if an abnormal value or data redundancy is found in the reliable data set, then through an abnormality detection algorithm, the records that do not conform to the logic are removed, to obtain the comprehensive analysis data; According to the comprehensive analysis data, a defect distribution heat map is used to draw a spatial distribution pattern of the surface defects of the radio frequency connector, and a batch correlation line chart is used to show the trend relationship between different batch numbers and defect rates, and through the defect distribution heat map and the batch correlation line chart, a first visualization chart is generated.

[0013] Preferably, the performance indicator change and defect severity are further analyzed by the visualization chart to generate risk assessment data, including: For the first visualization chart, a time series scatter plot technique is used to arrange the performance indicator change distribution by production date, and a color coding method is used to mark the defect severity with different colors. According to the marking result, a second visualization chart is generated. For the second visualization chart, a dynamic interactive interface technique is used to support users to view specific defect location information. The performance indicator bar chart presents the comparison result of the predicted performance value and the preset threshold value. If the predicted performance value is lower than the preset threshold value, it is marked as a high-risk item. According to the marking result, risk assessment data is generated.

[0014] Preferably, the final detection analysis archive is generated according to the risk assessment data, including: For the risk assessment data, a defect distribution heat map and a batch associated line graph are saved as a vector format file through a chart export function. According to the saving result, a final detection analysis archive is generated; For the final detection analysis archive, an archive index is established to support fast retrieval. By checking the final detection analysis archive, the integrity of the archive data is determined. If the integrity does not meet the requirements, the archive data is regenerated; According to the final detection analysis archive, a data backup mechanism is established to ensure data security. Through the classification management of the final detection analysis archive, the storage priority of the archive is determined. According to the storage priority, the storage location and access permission of the archive are adjusted.

[0015] A recognition detection system suitable for a recognition detection method of a radio frequency connector, the recognition detection system comprising: An image acquisition and preprocessing module: used for acquiring initial appearance image data of the surface of the radio frequency connector, and processing the initial appearance image data to obtain a first appearance image; A multi-modal feature extraction and decoding module: used for extracting an appearance feature set and an identification area image from the first appearance image; A data fusion and verification module: used for character segmentation and feature extraction on the identification area image, and combining a pre-established text recognition model to obtain a preliminary manufacturing information text. The preliminary manufacturing information text is verified and associated with the appearance feature set to generate a first associated data set; A defect analysis and visualization decision module: used for comparing the appearance feature set with a preset standard appearance database to generate defect marking data, and fusing the first associated data set with the defect marking data to generate comprehensive analysis data, and generating a first visualization chart according to the comprehensive analysis data; Risk assessment and archive management module: used for further analyzing performance index changes and defect severity of the first visualization chart, generating risk assessment data, and generating a final detection analysis archive according to the risk assessment data.

[0016] Preferably, the image acquisition and preprocessing module comprises: An image acquisition unit: scanning the surface of the radio frequency connector to obtain initial appearance image data; An illumination correction preprocessing unit: for the initial appearance image data, adjusting uneven brightness areas by using illumination correction processing technology, and ensuring the uniformity of the image data by correcting the brightness of the initial appearance image data; An image quality iterative optimization unit: checking the definition of the corrected initial appearance image data, and if it does not meet the requirements, re-performing image acquisition and processing, and obtaining a first appearance image with quality meeting the requirements through multiple processing iterations; The multi-modal feature extraction and decoding module comprises: A character region positioning unit: used for identifying the surface manufacturing information identification region of the obtained first appearance image and determining the identification region image; An edge detection and feature extraction unit: used for extracting edge contour features in the first appearance image, determining geometric feature data of surface defects, and integrating the edge contour features and the geometric feature data into an appearance feature set; A region segmentation and boundary optimization unit: used for region segmentation of the identification region image to separate the specific range of the manufacturing information identification, and boundary optimization of the identification region image to generate clear identification region image data; A feature mapping relationship establishment unit: used for establishing a feature mapping relationship between the appearance feature set and the clear identification region image data for subsequent defect comparison and text analysis; The data fusion and verification module comprises: A character segmentation and feature extraction unit: used for splitting characters connected together into single character units and obtaining stroke width and spacing feature data of the single character units; A text recognition and preliminary generation unit: used for classifying and matching the obtained stroke width and spacing feature data of the single character units in combination with a pre-established text recognition model, and generating the preliminary manufacturing information text through the classification matching result; A text format verification unit: used for verifying whether the batch number and the production date of the preliminary manufacturing information text conform to the preset coding rules, and removing error data that does not conform to the rules through the verification result; The cross-modal data association unit uniquely binds the checked text data and the appearance feature set, and generates the first associated data set through the binding result; The defect analysis and visualization decision module includes: The standard comparison and defect detection unit is used to match the appearance feature set with the preset standard appearance database, calculate the deviation value of the appearance feature set from the standard data through the matching result, and if the deviation value exceeds the preset threshold, judge as appearance defect, and generate the defect mark data according to the judgment result; The multi-dimensional data fusion unit is used to perform information fusion on the first associated data set and the defect mark data by using multi-dimensional data superposition technology, and associate the appearance defect information with the manufacturing information through the fusion result, and generate the comprehensive analysis data containing the defect position and batch number; The visualization chart generation unit is used to draw the spatial distribution graph of the surface defects of the radio frequency connector by the defect distribution heat map, and use the batch association line chart to show the trend relationship between different batch numbers and defect rates, and generate the first visualization chart; The risk assessment and archive management module includes: The time series visualization analysis unit is used to arrange the performance index change distribution according to the production date by using the time series scatter plot technology on the first visualization chart, and mark the defect severity with different colors by using the color coding method, and generate the second visualization chart according to the marking result; The interactive defect positioning unit is used to support users to view specific defect position information by using the dynamic interactive interface technology on the second visualization chart, and present the comparison result of the predicted performance value and the preset threshold by using the performance index column chart, and if the predicted performance value is lower than the preset threshold, mark it as a high risk item, and generate the risk assessment data according to the marking result; The multi-format chart export unit is used to save the defect distribution heat map and the batch association line chart as a vector format file by using the chart export function on the risk assessment data, and generate the final detection analysis archive according to the saving result; For the final detection analysis archive, an archive index is established to support fast retrieval The intelligent archive management unit establishes an archive index to support fast retrieval for the final detection analysis archive, determines the integrity of the archive data by checking the final detection analysis archive, and if the integrity does not meet the requirements, regenerates the archive data; The enterprise-level data governance unit is used to establish a data backup mechanism for the final detection analysis archive to ensure data security, and perform classified management on the final detection analysis archive to determine the storage priority of the archive, and adjust the storage location and access permission of the archive according to the storage priority.

[0017] The technical scheme provided by the embodiment of the present application can include the following beneficial effects: The application discloses a radio frequency connector appearance defect detection and risk assessment method, an initial appearance image is acquired through a high-resolution image acquisition device, a first appearance image is obtained through illumination correction processing, a character region positioning method is used to identify a surface manufacturing information mark region and extract appearance features, manufacturing information text is acquired in combination with a character segmentation algorithm and a text recognition model, the verified text is uniquely bound with the appearance feature set to obtain a first associated data set; the application detects appearance defects by using a template comparison method, generates defect mark data, fuses appearance defect information and manufacturing information to form comprehensive analysis data, presents defect distribution and trends through defect distribution heat maps and batch association line graphs and other visualization methods, and finally generates risk assessment data, realizing accurate detection and risk assessment of radio frequency connector appearance defects and providing effective support for product quality control. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 A flowchart of the identification detection method for the radio frequency connector of the present application; Figure 2 A structural block diagram of the identification detection system for the radio frequency connector of the present application. DETAILED DESCRIPTION

[0019] The technical scheme in the embodiments of the present application will be described clearly and in detail below with reference to the drawings in the embodiments of the present application. The described embodiments are only a part of the embodiments of the present application.

[0020] As Figure 1 , the identification detection method for the radio frequency connector of the present application, the method comprises the following steps: Step 1, acquiring initial appearance image data of the surface of the radio frequency connector through an image acquisition device, and processing the initial appearance image data to obtain a first appearance image; Step 2, extracting an appearance feature set and a mark region image from the first appearance image; Step 3, performing character segmentation and feature extraction on the mark region image, obtaining preliminary manufacturing information text in combination with a pre-established text recognition model, verifying the preliminary manufacturing information text and associating the preliminary manufacturing information text with the appearance feature set to generate a first associated data set; Step 4, comparing the appearance feature set with a preset standard appearance database to generate defect mark data, fusing the first associated data set and the defect mark data to generate comprehensive analysis data, and generating a first visualization chart according to the comprehensive analysis data, for presenting defect distribution and trend relationship; Step 5, further analyzing the performance index change and defect severity according to the first visualization chart, generating risk assessment data, and generating a final detection analysis file according to the risk assessment data for archiving and feedback.

[0021] In one embodiment, the initial appearance image data of the surface of the radio frequency connector is acquired by the image acquisition device, and the first appearance image is obtained by processing the initial appearance image data, comprising: The surface of the radio frequency connector is scanned by a high-resolution image acquisition device to obtain initial appearance image data, and the uneven brightness area is adjusted by using an illumination correction processing technology for the initial appearance image data. The uniformity of the image data is ensured by correcting the brightness of the initial appearance image data. According to the corrected initial appearance image data, it is determined whether the definition of the image meets the preset requirements. If not, image acquisition and processing are performed again, and the first appearance image with quality meeting the requirements is obtained through multiple processing iterations.

[0022] For example, when image acquisition is performed, the surface of the radio frequency connector can be scanned by a high-resolution image acquisition device at a resolution of 50 million pixels, and a line array CCD sensor is used to acquire initial appearance image data at a precision of 0.05 mm. For the initial appearance image data, the MSRCR algorithm based on the Retinex theory is used for illumination correction, the Gaussian kernel scale parameter is set to [15, 80, 200], the uneven brightness area is adjusted to a gray value deviation of less than 5%, and after the correction is completed, the first appearance image is obtained.

[0023] In one embodiment, the appearance feature set and the identification area image are extracted from the first appearance image, comprising: For the first appearance image, a character region positioning method is used to identify the surface manufacturing information identification area, determine the identification area image, and perform region segmentation on the identification area image to separate the specific range of the manufacturing information identification, and generate clear identification area image data. In this embodiment, when extracting the identification area image, a character area positioning method based on a convolutional neural network (CNN) can be used to perform feature extraction on the first appearance image, identify the surface manufacturing information identification area, and determine the identification area image. A threshold-based image segmentation technique is used to separate the text part from the identification area image to obtain clear identification area image data. When using the threshold-based image segmentation technique, a global threshold method can be used to manually select a threshold value by observing the gray scale histogram of the image according to experience. For example, if the gray scale values of most text pixels in the image are concentrated between 100-200, and the gray scale values of background pixels are concentrated between 0-50, a threshold value of 75 can be selected. A local threshold method such as an adaptive threshold method can also be used, which can achieve adaptive threshold segmentation by calculating the threshold value according to the local area characteristics of the image. The clear identification area image data is analyzed by the Tesseract optical character recognition engine to extract manufacturing information text data, including model number, batch number, and other information.

[0024] The edge profile features in the first appearance image are extracted by an edge detection technique, and the geometric feature data of the surface defects are determined according to the edge profile features. The edge profile features and the geometric feature data are integrated into an appearance feature set. According to the appearance feature set and the clear identification area image data, a feature mapping relationship is established for subsequent defect comparison and text analysis.

[0025] In one embodiment, the identification area image is subjected to character segmentation and feature extraction, and a preliminary manufacturing information text is obtained by combining a pre-established text recognition model. The preliminary manufacturing information text is verified and associated with the appearance feature set to generate a first associated data set, including: For the identification area image, a character segmentation algorithm is used to separate the characters connected together into single character units, and through character feature extraction technology, the stroke width and spacing feature data of the single character units are obtained. According to the stroke width and spacing feature data, a pre-established text recognition model is used for classification matching, and the preliminary manufacturing information text is generated through the classification matching result. In this embodiment, the characters connected together are split by using a character segmentation algorithm based on the projection method, for example, vertical projection analysis is performed on the image, the threshold is set to 5 pixels, and the characters are segmented into single character units. By extracting the features of the single character units, the stroke width and spacing feature data are obtained by using the edge detection algorithm, for example, the Canny algorithm is used to extract the edge, the average stroke width is calculated to be 3 pixels, and the average spacing is calculated to be 2 pixels. The extracted stroke width and spacing feature data are classified and matched by using a pre-established text recognition model based on a convolutional neural network, for example, the ResNet model is used to classify the characters, and the preliminary manufacturing information text "2023-09-15-B123" is obtained.

[0026] For the preliminary manufacturing information text, a text format verification method is used to verify whether the batch number and the production date conform to the preset encoding rule, and through the verification result, the error data that does not conform to the rule is removed; according to the text data after verification, a data correlation mapping technology is used to uniquely bind with the appearance feature set, and through the binding result, the first correlation data set is generated.

[0027] In this embodiment, a regular expression verification method is used to verify the batch number and the production date in the preliminary manufacturing information text, for example, a regular expression "^\d{4}-\d{2}-\d{2}-B\d{3} ” is used to judge whether it conforms to the preset encoding rule. If the verification result shows that the batch number or the production date does not conform to the preset encoding rule, for example, "2023-09-15-B12" does not conform to the rule, then the error data is removed, and the text data after verification "2023-09-15-B123" is obtained. Through the data correlation mapping technology, the text data after verification is uniquely bound with the appearance feature set, for example, a hash algorithm is used to generate a unique identifier, and the first correlation data set is obtained.

[0028] In one embodiment, the appearance feature set is compared with a preset standard appearance database to generate defect marking data, including: For the appearance feature set, a template matching method is used to match with the preset standard appearance database, through the matching result, the deviation value of the appearance feature set and the standard data is calculated, if the deviation value exceeds the preset threshold, it is judged as an appearance defect, according to the judgment result, the defect marking data is generated.

[0029] In this embodiment, a template matching method based on Euclidean distance is used to match the appearance feature set with the standard appearance database, for example, the Euclidean distance is calculated to be 0.8, and the comparison deviation value is obtained. If the comparison deviation value exceeds the preset threshold of 0.5, it is judged as an appearance defect, and the defect marking data "defect type: scratch, position: upper right corner" is generated.

[0030] By integrating the defect mark data with the first association data set, using a decision tree-based comprehensive analysis algorithm, the final manufacturing information and defect analysis result "batch: B123, production date: 2023-09-15, defect: scratch" is obtained.

[0031] In one embodiment, the first association data set and the defect mark data are fused to generate comprehensive analysis data, and a first visualization chart is generated according to the comprehensive analysis data, comprising: For the first association data set and the defect mark data, a multi-dimensional data superposition technology is used to perform feature matching on the appearance defect information in the defect mark data and the manufacturing information in the first association data set to determine a preliminary fused data structure. Through a data structure analysis tool, field mapping processing is performed on the preliminary fused data structure to obtain intermediate layer data containing defect positions and batch numbers. If there are missing fields or inconsistent formats in the intermediate layer data, a data cleaning algorithm is used to fill in the missing values and unify the format to determine standardized data that meets the requirements. According to the standardized data, a data integration framework is used to deeply associate the defect positions and batch numbers to obtain a basic version of the comprehensive analysis data. Through a data verification mechanism, the integrity and accuracy of the basic version of the comprehensive analysis data are detected to determine a reliable data set. If abnormal values or data redundancy are found in the reliable data set, an abnormality detection algorithm is used to remove records that do not meet the logic to obtain comprehensive analysis data. In the example, when the first associated data set is fused with the defect marking data, the original data combination containing appearance defect information and preliminary manufacturing information is extracted from the first associated data set and the defect marking data, for example, manufacturing records and corresponding defect marking data of batch numbers B20231001 to B20231010 are obtained from the database, according to the original data combination, the appearance defect information in the defect marking data and the manufacturing information in the first associated data set are matched by using multi-dimensional data superposition technology, for example, KNN algorithm is used to associate defect position coordinates with manufacturing batch numbers, to determine the preliminary fused data structure, through field mapping processing of the preliminary fused data structure by using data structure analysis tool, for example, the defect position field is mapped to latitude and longitude coordinates, and the batch number field is mapped to production date, to obtain intermediate layer data containing defect position and batch number, if there are field missing or format inconsistency in the intermediate layer data, the missing values are filled and the format is unified by using data cleaning algorithm, for example, the missing defect position data is completed by using mean filling method, to judge the standardized data meeting the requirements, according to the standardized data, the defect position and the batch number are deeply associated by using data integration framework, for example, the association rule mining algorithm is used to analyze the association between the defect position and the batch number, to obtain the basic version of the comprehensive analysis data, the integrity and accuracy of the basic version of the comprehensive analysis data are detected by using data verification mechanism, for example, the CRC check algorithm is used to verify the integrity of the data, to determine the reliable data set, if abnormal values or data redundancy are found in the reliable data set, the records not meeting the logic are removed by using abnormal detection algorithm, for example, the isolated forest algorithm is used to identify and remove abnormal batch data, to obtain the comprehensive analysis data.

[0032] For the comprehensive analysis data, a spatial distribution analysis algorithm is used to grid the defect position, to obtain preliminary spatial data of defect distribution; by calculating the heat value of the preliminary spatial data, a color mapping technology is used to generate a defect distribution heat map, to determine the defect dense area on the surface of the radio frequency connector; according to the result of the defect distribution heat map, the defect quantity statistical value of each grid area is obtained, to obtain the quantization index data of spatial distribution; by collecting the defect rate data of different batch numbers, a time series analysis method is used to arrange the relationship between batch and defect rate, to obtain trend data associated with batch; according to the trend data associated with batch, a line chart drawing technology is used to generate a batch association line chart, to determine the trend characteristics of the defect rate changing with batch; by integrating the defect distribution heat map and the batch association line chart, a multi-layer superposition technology is used to generate a first visual chart.

[0033] In this embodiment, the spatial distribution analysis algorithm is used to grid the defect position, the surface is divided into 10x10 grid units, the preliminary spatial data of defect distribution is obtained, the defect density value of each grid unit is calculated by calculating the heat value of the preliminary spatial data and using the Gaussian kernel density estimation algorithm, and the density value is mapped to a gradient heat map from blue to red using color mapping technology. Determine the defect dense area of the surface of the radio frequency connector, according to the result of the defect distribution heat map, obtain the defect quantity statistical value of each grid area, for example, the defect quantity of grid A1 is 15, the defect quantity of grid B2 is 8, obtain the quantitative index data of spatial distribution, collect the defect rate data of different batch numbers, and arrange the relationship between batch and defect rate by using time series analysis method, for example, the defect rate of batch 001 is 2.5%, the defect rate of batch 002 is 3.8%, and the trend data associated with the batch is obtained. According to the trend data associated with the batch, the batch association line chart is generated by using line chart drawing technology, the horizontal axis is the batch number, the vertical axis is the defect rate, the trend characteristics of the defect rate changing with the batch are determined, and the first visual chart is generated by using multi-layer superposition technology through view integration of the defect distribution heat map and the batch association line chart.

[0034] In one embodiment, the performance indicator change and defect severity are further analyzed by the visual chart to generate risk assessment data, including: The production date and corresponding performance indicator data of the radio frequency connector are obtained to obtain a preliminary time series data set, the change distribution of the performance indicator is arranged in order of production date according to the preliminary time series data set by using time series scatter plot technology, a basic visual graph is generated, the defect severity data of each data point is analyzed by the defect detection algorithm, the classification result of the defect severity is obtained, and each data point is assigned a different color label according to the classification result of the defect severity by using color coding method. Get a time series scatter plot with color markers, associate the time series scatter plot with color markers with the original performance indicator data through the data integration module to generate a second visual chart, bind each data point in the chart with the corresponding defect detailed information according to the second visual chart by using data storage technology, and obtain an interactive data structure. Through the dynamic interaction interface technology, the interactive data structure is rendered, the specific defect position information is extracted and displayed when the chart is clicked, the user query result is obtained, the comparison graph of predicted performance value and preset threshold is drawn by using performance indicator column chart generation technology according to the user query result, and the comparison analysis chart is obtained. If the predicted performance value is lower than the preset threshold, the corresponding data point is marked as a high-risk item by the risk marking algorithm, and the final risk assessment data is determined.

[0035] In this embodiment, the production date and corresponding performance indicators of the radio frequency connector, such as insertion loss (range 0.5dB-2.0dB) and voltage standing wave ratio (range 1.2-1.8), are extracted to form a CSV format dataset containing timestamp and numerical value. Time series scatter plot technology calls Python's Matplotlib library to draw a scatter plot with production date as horizontal axis and performance indicator as vertical axis. The horizontal axis interval is set to 7 days, and the vertical axis is graded according to 0.1dB accuracy. The defect detection algorithm uses the YOLOv5 model to analyze the quality inspection image associated with each data point, and outputs the defect severity classification (0-no defect, 1-mild, 2-moderate, 3-severe) with an accuracy of 98%. The color coding module maps the classification results to green (0 level), yellow (1 level), orange (2 level), and red (3 level). The scatter plot color channel value is modified through the OpenCV library. The data integration module uses Pandas to merge the color marking data and the original performance indicators to generate a JSON structure containing the RGB field. The data storage technology uses MongoDB to establish a document-based database to associate defect coordinates (such as X:12.5mm, Y:8.3mm) and optical detection raw data with the production date as the primary key. The dynamic interactive interface is developed based on D3.js, which listens for mouse click events and triggers AJAX requests to retrieve 200dpi microscopic images corresponding to the coordinates from the database. The performance indicator bar chart is generated through Echarts, with a preset threshold line of insertion loss 1.5dB. If the predicted value 1.7dB exceeds the threshold, the risk marking algorithm updates the risk_flag field to 1 in MySQL and triggers an enterprise WeChat alarm push.

[0036] In one embodiment, the final detection analysis file is generated based on the risk assessment data, including: For the risk assessment data, save the defect distribution heat map and batch associated line chart as a vector format file through the chart export function. According to the saving result, generate the final detection analysis file; For the final detection analysis file, establish an index to support fast retrieval. By checking the final detection analysis file, determine the integrity of the file data. If the integrity does not meet the requirements, regenerate the file data; According to the final detection analysis file, establish a data backup mechanism to ensure data security. Through the classification management of the final detection analysis file, determine the storage priority of the file. According to the storage priority, adjust the storage location and access rights of the file.

[0037] Referring to Figure 2 As shown in the drawings, an identification detection system for an identification detection method of a radio frequency connector, the identification detection system comprising: An image acquisition and preprocessing module is configured to acquire initial appearance image data of the surface of the RF connector and to process the initial appearance image data to obtain a first appearance image. The image acquisition and preprocessing module includes: An image acquisition unit is configured to scan the surface of the RF connector to obtain initial appearance image data. An illumination correction preprocessing unit is configured to adjust uneven brightness regions of the initial appearance image data by using an illumination correction processing technique, and to ensure uniformity of the image data by correcting the brightness of the initial appearance image data. An image quality iterative optimization unit is configured to check the definition of the corrected initial appearance image data, and if the definition does not meet the requirements, to re-acquire and process the image, and to obtain a first appearance image with quality meeting the requirements through multiple processing iterations.

[0038] A multi-modal feature extraction and decoding module is configured to extract an appearance feature set and an identification region image from the first appearance image. The multi-modal feature extraction and decoding module includes: A character region positioning unit is configured to identify the first appearance image surface manufacturing information identification region to determine the identification region image. An edge detection and feature extraction unit is configured to extract edge contour features in the first appearance image, to determine geometric feature data of surface defects, and to integrate the edge contour features and the geometric feature data into an appearance feature set. A region segmentation and boundary optimization unit is configured to perform region segmentation on the identification region image to separate the specific range of the manufacturing information identification, to optimize the boundary of the identification region image, and to generate clear identification region image data. A feature mapping relationship establishment unit is configured to establish a feature mapping relationship between the appearance feature set and the clear identification region image data, which is used for subsequent defect comparison and text analysis.

[0039] A data fusion and verification module is configured to perform character segmentation and feature extraction on the identification region image, to obtain preliminary manufacturing information text by combining a pre-established text recognition model, to verify the preliminary manufacturing information text, and to associate the preliminary manufacturing information text with the appearance feature set to generate a first association data set. The data fusion and verification module includes: A character segmentation and feature extraction unit is configured to split characters connected together into single character units and to obtain stroke width and spacing feature data of the single character units. A text recognition and preliminary generation unit is configured to classify and match the stroke width and spacing feature data of the obtained single character units by combining a pre-established text recognition model, and to generate the preliminary manufacturing information text through the classification and matching results. The text format checking unit is configured to check whether the batch number and the production date of the preliminary manufacturing information text conform to preset coding rules, and remove error data that does not conform to the rules according to the checking result. The cross-modal data association unit is configured to uniquely bind the checked text data and the appearance feature set, and generate the first associated data set according to the binding result.

[0040] The defect analysis and visualization decision module is configured to compare the appearance feature set with a preset standard appearance database to generate defect marking data, fuse the first associated data set and the defect marking data to generate comprehensive analysis data, and generate a first visualization chart according to the comprehensive analysis data. The defect analysis and visualization decision module includes: The standard comparison and defect detection unit is configured to match the appearance feature set with the preset standard appearance database, calculate a deviation value of the appearance feature set from the standard data according to a matching result, determine that there is an appearance defect if the deviation value exceeds a preset threshold, and generate the defect marking data according to a determination result. The multi-dimensional data fusion unit is configured to fuse the first associated data set and the defect marking data by using a multi-dimensional data superposition technology, associate appearance defect information and manufacturing information according to a fusion result, and generate the comprehensive analysis data including a defect position and a batch number. The visualization chart generation unit is configured to draw a spatial distribution graph of surface defects of the radio frequency connector by using a defect distribution heat map, display a trend relationship between different batch numbers and defect rates by using a batch association line chart, and generate the first visualization chart.

[0041] The risk assessment and archive management module is configured to further analyze performance index changes and defect severity according to the first visualization chart, generate risk assessment data, and generate a final detection analysis archive according to the risk assessment data. The risk assessment and archive management module includes: The time series visualization analysis unit is configured to arrange performance index change distribution according to production dates by using a time series scatter plot technology, mark defect severity by using different colors according to a color coding method, and generate a second visualization chart according to a marking result. The interactive defect positioning unit is configured to support a user to view specific defect position information by using a dynamic interactive interface technology according to the second visualization chart, present a comparison result of a predicted performance value and a preset threshold by using a performance index column chart, mark a high-risk item if the predicted performance value is lower than the preset threshold, and generate the risk assessment data according to a marking result. A multi-format chart exporting unit is configured to save a defect distribution heat map and a batch-associated line chart as a vector format file through a chart exporting function, and generate a final detection analysis archive according to a saving result; and an archive index is established for the final detection analysis archive to support fast retrieval An intelligent archive management unit is configured to establish an archive index for the final detection analysis archive to support fast retrieval, and determine the integrity of archive data through verification of the final detection analysis archive; if the integrity does not meet the requirements, the archive data is regenerated; An enterprise-level data governance unit is configured to establish a data backup mechanism for the final detection analysis archive to ensure data security, and perform classified management on the final detection analysis archive, determine a storage priority of the archive, and adjust a storage location and access authority of the archive according to the storage priority.

[0042] Through comprehensive application of the above system, the detection efficiency is greatly improved, the interference of human factors is reduced, and the accuracy and reliability of the detection result are significantly improved.

[0043] The above specific embodiments further illustrate the purpose, technical solutions and beneficial effects of the present application. It should be understood that the above description is merely a specific embodiment of the present application and is not intended to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for identifying and detecting radio frequency connectors, characterized in that, The method includes the following steps: Step 1: Acquire initial appearance image data of the RF connector surface using an image acquisition device, and process the initial appearance image data to obtain the first appearance image; Step 2: Extract the set of appearance features and the image of the marked area from the first appearance image; Step 3: Perform character segmentation and feature extraction on the image of the identified area, combine it with the pre-established text recognition model to obtain preliminary manufacturing information text, verify the preliminary manufacturing information text and associate it with the appearance feature set to generate the first associated dataset; Step 4: Compare the set of appearance features with the preset standard appearance database to generate defect marking data. Merge the first associated dataset with the defect marking data to generate comprehensive analysis data. Generate the first visualization chart based on the comprehensive analysis data to present the distribution and trend relationship of defects. Step 5: Further analyze the changes in performance indicators and the severity of defects based on the first visualization chart, generate risk assessment data, and generate the final detection analysis file based on the risk assessment data for archiving and feedback.

2. The identification and detection method for radio frequency connectors as described in claim 1, characterized in that, The process of acquiring initial appearance image data of the RF connector surface through an image acquisition device and processing the initial appearance image data to obtain a first appearance image includes: The surface of the RF connector is scanned using a high-resolution image acquisition device to obtain initial appearance image data. For the initial appearance image data, illumination correction processing technology is used to adjust areas with uneven brightness. By correcting the brightness of the initial appearance image data, the uniformity of the image data is ensured. Based on the corrected initial appearance image data, determine whether the image clarity meets the preset requirements. If not, re-acquire and process the image, and obtain a first appearance image that meets the quality requirements through multiple processing iterations.

3. The identification and detection method for radio frequency connectors as described in claim 1, characterized in that, The step of extracting the set of appearance features and the image of the marked region from the first appearance image includes: For the first appearance image, the character region positioning method is used to identify the surface manufacturing information marking area, determine the marking area image, and perform region segmentation on the marking area image to separate the specific range of the manufacturing information marking, generating clear marking area image data; Edge contour features are extracted from the first appearance image using edge detection technology. Based on the edge contour features, the geometric feature data of the surface defects are determined. The edge contour features and geometric feature data are then integrated into an appearance feature set. Based on the set of appearance features and the image data of clearly marked areas, a feature mapping relationship is established for subsequent defect comparison and text parsing.

4. The identification and detection method for radio frequency connectors as described in claim 3, characterized in that, The process of segmenting and extracting characters from the identified region image, combining it with a pre-established text recognition model to obtain preliminary manufacturing information text, verifying the preliminary manufacturing information text and associating it with the appearance feature set to generate a first associated dataset includes: For the image of the identified region, a character segmentation algorithm is used to break down connected characters into individual character units. Then, character feature extraction technology is used to obtain the stroke width and spacing feature data of each individual character unit. Based on the stroke width and spacing feature data, a pre-established text recognition model is used for classification and matching. Based on the classification and matching results, preliminary manufacturing information text is generated. For the initial manufacturing information text, a text format verification method is used to verify whether the batch number and production date conform to the preset coding rules. Based on the verification results, erroneous data that does not conform to the rules are removed. Based on the verified text data, a data association mapping technology is used to uniquely bind it to the appearance feature set. Based on the binding results, the first associated dataset is generated.

5. The identification and detection method for radio frequency connectors as described in claim 1, characterized in that, The step of comparing the set of appearance features with a preset standard appearance database to generate defect marking data includes: For the set of appearance features, a template comparison method is used to match it with a preset standard appearance database. Based on the matching results, the deviation value between the set of appearance features and the standard data is calculated. If the deviation value exceeds a preset threshold, it is judged as an appearance defect. Based on the judgment result, defect marking data is generated.

6. The identification and detection method for radio frequency connectors as described in claim 5, characterized in that, The process of fusing the first associated dataset with defect marker data to generate comprehensive analysis data, and generating a first visualization chart based on the comprehensive analysis data, includes: For the first associated dataset and the defect marking data, multidimensional data overlay technology is used to fuse information. Through the fusion result, the appearance defect information is associated with the manufacturing information. Based on the association result, comprehensive analysis data containing defect location and batch number is generated. Based on the comprehensive analysis data, a spatial distribution graph of surface defects of the RF connector is drawn using a defect distribution heatmap. A batch correlation line graph is used to show the trend relationship between different batch numbers and defect rates. The first visualization chart is generated by combining the defect distribution heatmap and the batch correlation line graph.

7. The identification and detection method for radio frequency connectors as described in claim 6, characterized in that, The process involves further analyzing performance indicator changes and defect severity through visual charts to generate risk assessment data, including: For the first visualization chart, time series scatter plot technology is used to arrange the distribution of performance index changes by production date. Different colors are used to mark the severity of defects through color coding. Based on the marking results, a second visualization chart is generated. For the second visualization chart, dynamic interactive interface technology is used to allow users to view specific defect location information. The performance index bar chart presents the comparison results between the predicted performance value and the preset threshold. If the predicted performance value is lower than the preset threshold, it is marked as a high-risk item. Based on the marking results, risk assessment data is generated.

8. The identification and detection method for radio frequency connectors as described in claim 7, characterized in that, The process of generating the final detection and analysis file based on the risk assessment data includes: For the risk assessment data, the defect distribution heatmap and batch correlation line chart are saved as vector format files using the chart export function. Based on the saved results, the final inspection and analysis file is generated. For the final test and analysis archives, an archive index is established to support fast retrieval. The integrity of the archive data is determined by verifying the final test and analysis archives. If the integrity does not meet the requirements, the archive data is regenerated. Based on the final test and analysis files, a data backup mechanism is established to ensure data security. By classifying and managing the final test and analysis files, the storage priority of the files is determined, and the storage location and access permissions of the files are adjusted according to the storage priority.

9. An identification and detection system applicable to the identification and detection method for radio frequency connectors according to any one of claims 1-8, characterized in that, The identification and detection system includes: Image acquisition and preprocessing module: used to acquire initial appearance image data of the surface of the RF connector, and process the initial appearance image data to obtain a first appearance image; Multimodal feature extraction and decoding module: used to extract the set of appearance features and the image of the marked region from the first appearance image; Data fusion and verification module: used to perform character segmentation and feature extraction on the image of the identified area, and combine it with a pre-established text recognition model to obtain preliminary manufacturing information text, verify the preliminary manufacturing information text and associate it with the appearance feature set to generate a first associated dataset; Defect analysis and visualization decision module: used to compare the set of appearance features with a preset standard appearance database to generate defect marking data, and to merge the first associated dataset with the defect marking data to generate comprehensive analysis data, and generate a first visualization chart based on the comprehensive analysis data; Risk assessment and record management module: used to further analyze the performance index changes and defect severity of the first visualization chart, generate risk assessment data, and generate the final detection analysis record based on the risk assessment data.

10. The identification and detection system for an identification and detection method for radio frequency connectors as described in claim 9, characterized in that, The image acquisition and preprocessing module includes: Image acquisition unit: scans the surface of the RF connector to acquire initial appearance image data; Illumination correction preprocessing unit: For the initial appearance image data, illumination correction processing technology is used to adjust the areas with uneven brightness. By correcting the brightness of the initial appearance image data, the uniformity of the image data is ensured. Image quality iterative optimization unit: verifies the clarity of the corrected initial appearance image data. If it does not meet the requirements, the image is re-acquired and processed. Through multiple processing iterations, a first appearance image with the required quality is obtained. The multimodal feature extraction and decoding module includes: Character region positioning unit: used to identify the manufacturing information marking area on the surface of the acquired first appearance image and determine the marking area image; Edge detection and feature extraction unit: used to extract edge contour features from the first appearance image, determine the geometric feature data of surface defects, and integrate the edge contour features and the geometric feature data into an appearance feature set; Region segmentation and boundary optimization unit: used to segment the identification region image to separate the specific range of the manufacturing information identification, and optimize the boundary of the identification region image to generate clear identification region image data; Feature mapping relationship establishment unit: used to establish a feature mapping relationship between the set of appearance features and the image data of clearly marked areas, for subsequent defect comparison and text parsing; The data fusion and verification module includes: Character segmentation and feature advance unit: used to split connected characters into individual character units and obtain the stroke width and spacing feature data of the individual character units; Text recognition and preliminary generation unit: This unit is used to classify and match the stroke width and spacing feature data of the acquired individual character units with a pre-established text recognition model, and generate preliminary manufacturing information text based on the classification and matching results. Text format verification unit: used to verify whether the batch number and production date of the preliminary manufacturing information text conform to the preset coding rules, and to remove erroneous data that does not conform to the rules based on the verification results; Cross-modal data association unit: uniquely binds the verified text data to the appearance feature set, and generates a first association dataset based on the binding result; The defect analysis and visualization decision-making module includes: Standard comparison and defect detection unit: used to match the set of appearance features with a preset standard appearance database, calculate the deviation value between the set of appearance features and the standard data based on the matching result, and if the deviation value exceeds a preset threshold, it is judged as an appearance defect. Based on the judgment result, defect marking data is generated. Multidimensional data fusion unit: used to fuse information between the first associated dataset and the defect marking data using multidimensional data overlay technology, and to associate the appearance defect information with the manufacturing information through the fusion result, generating comprehensive analysis data containing defect location and batch number; Visualization chart generation unit: used to draw the spatial distribution of defects on the surface of RF connectors using a defect distribution heatmap, and to show the trend relationship between different batch numbers and defect rates using a batch correlation line chart, generating the first visualization chart; The risk assessment and record management module includes: Time series visualization analysis unit: Used to arrange the distribution of performance index changes by production date using time series scatter plot technology on the first visualization chart, and to mark the severity of defects with different colors using a color coding method. Based on the marking results, a second visualization chart is generated. Interactive Defect Location Unit: Used to support users in viewing specific defect location information using dynamic interactive interface technology on the second visualization chart, and presents the comparison results of predicted performance value and preset threshold through performance index bar chart. If the predicted performance value is lower than the preset threshold, it is marked as a high-risk item. Based on the marking results, risk assessment data is generated. Multi-format chart export unit: This unit exports risk assessment data to vector format files, including defect distribution heatmaps and batch correlation line graphs. Based on the exported data, it generates a final inspection and analysis archive. An index is then created for this final inspection and analysis archive to support rapid retrieval. Intelligent record management unit: For the final inspection and analysis records, a record index is established to support fast retrieval. The integrity of the record data is determined by verifying the final inspection and analysis records. If the integrity does not meet the requirements, the record data is regenerated. Enterprise-level data governance unit: Used to establish a data backup mechanism for the final test and analysis files to ensure data security, as well as to classify and manage the final test and analysis files, determine the storage priority of the files, and adjust the storage location and access permissions of the files according to the storage priority.

Images