A terminal pressing imaging quality detection method, system, medium and product

Abstract

This invention discloses a method, system, medium, and product for detecting terminal crimping imaging quality, relating to the field of imaging quality detection. In this method, deep learning feature extraction is combined with image processing technology. Through the synergistic effect of a dynamic threshold adjustment strategy library and an edge detection morphological processing rule library, a complete adaptive detection process is formed. This invention can automatically adjust detection parameters based on real-time image brightness conditions and terminal specification information, ensuring stable detection accuracy under different lighting environments and diverse terminal types. This reduces the problems of unstable detection accuracy and poor versatility caused by fixed parameters in existing technologies, improving the reliability and practicality of terminal crimping quality detection.

Description

Technical Field

[0001] This application belongs to the field of imaging quality detection, and particularly relates to a terminal crimping imaging quality detection method, system, medium and product. Background Art

[0002] Terminal crimping is a key process in the electronics manufacturing industry, directly affecting the reliability of electrical connections and product quality. As electronic products develop towards miniaturization and high density, the precision requirements for terminal crimping processes are getting higher and higher, posing higher requirements for crimping quality detection technologies. There is an urgent need to adopt high-precision and high-efficiency automated detection technologies to ensure the stability and consistency of terminal crimping quality.

[0003] Existing terminal crimping quality detection technologies mainly adopt machine vision detection methods based on fixed thresholds, performing edge detection and morphological analysis on the terminal crimping area through preset image processing parameters. These methods usually use traditional image processing algorithms, such as Canny edge detection, morphological operations, etc., combined with predefined geometric feature parameters to judge the crimping quality, and determine whether the crimping is qualified by comparing with a standard template.

[0004] However, fixed image processing parameters cannot adapt to imaging changes under different lighting conditions, resulting in unstable detection accuracy; at the same time, there is a lack of personalized parameter adjustment mechanisms for different terminal specifications and crimping processes, affecting the generality of detection, and this situation needs to be further improved. Summary of the Invention

[0005] This application provides a terminal crimping imaging quality detection method, system, medium and product, which is used to solve the problems of unstable detection accuracy and poor generality in the prior art. This method realizes high-precision quality detection under different lighting conditions and terminal specifications through image brightness adaptive adjustment and personalized image processing parameter calculation.

[0006] In a first aspect, this application provides a terminal crimping imaging quality detection method, including: Receiving the input of image data during the terminal crimping process, performing image preprocessing and brightness analysis, extracting the terminal crimping area features through a pre-trained deep convolutional neural network model, and calculating the crimping feature vector and image brightness parameters; According to the image brightness parameters, performing parameter matching in a preset dynamic threshold adjustment strategy library, and calculating the brightness adaptability adjustment multiple; Reading the terminal specification information and crimping process standards, performing feature analysis, and calculating personalized image processing parameters based on a preset edge detection and morphological processing rule library in combination with the brightness adaptability adjustment multiple; Based on the personalized image processing parameters, edge detection and morphological processing are used to locate the boundary of the terminal pressing area to obtain the pressing boundary contour. Based on the pressing boundary contour and the pressing feature vector, a quality assessment analysis is performed to generate the final terminal pressing quality inspection result.

[0007] In the above embodiments, this application combines deep learning feature extraction with image processing technology. Through the synergistic effect of a dynamic threshold adjustment strategy library and an edge detection morphology processing rule library, a complete adaptive detection process is formed. This application can automatically adjust the detection parameters according to real-time image brightness conditions and terminal specification information, ensuring stable detection accuracy under different lighting environments and diverse terminal types. This reduces the problems of unstable detection accuracy and poor versatility caused by fixed parameters in the prior art, and improves the reliability and practicality of terminal pressing quality detection. In one implementation, the step of matching parameters in a preset dynamic threshold adjustment strategy library based on the image brightness parameters to calculate the brightness adaptive adjustment factor specifically includes: Calculate the Euclidean distance between the image brightness parameter and each standard brightness condition in the dynamic threshold adjustment strategy library, and select the brightness condition with the smallest distance as the matching benchmark. Based on the brightness deviation coefficient calculation rule corresponding to the matching benchmark, the difference calculation is performed between the current image brightness parameter and the preset standard brightness intensity to obtain the brightness change quantification index. The brightness change quantification index is input into the corresponding nonlinear mapping function in the strategy library to calculate the adjustment factor of the image binarization threshold, contrast enhancement parameter and brightness correction parameter.

[0008] In the above embodiments, this application adopts a multi-parameter collaborative adjustment mechanism, and dynamically optimizes the binarization threshold, contrast enhancement parameter and brightness correction parameter to ensure that the best image processing effect can be obtained under different brightness conditions. The distance matching-based parameter selection method improves the accuracy and response speed of brightness adaptive adjustment, and improves the stability and accuracy of the entire detection process under changing lighting conditions. In one implementation, the step of reading terminal specification information and pressing process standards, performing feature analysis, and calculating personalized image processing parameters based on a preset edge detection and morphological processing rule library, combined with the brightness adaptive adjustment factor, specifically includes: Extract terminal size features based on terminal specification information, and extract crimping depth features based on crimping process standards; Based on the terminal size features and pressing depth features, the corresponding reference image processing parameters are selected from the preset edge detection and morphological processing rule library; Based on a preset terminal pressing detection parameter coordination rule library, the reference image processing parameters are coordinated and optimized to obtain a parameter coordination scheme. The terminal pressing detection parameter coordination rule library is a structured database that records the image processing parameter coordination strategies under different combinations of terminal specifications and pressing depths. The coordination strategies include four types: size compensation, depth enhancement, edge sharpening, and noise suppression. The brightness adaptive adjustment factor is multiplied by the coordinated and optimized image processing parameters to obtain personalized image processing parameters.

[0009] In the above embodiments, this application uses four coordination strategies based on the terminal crimping detection parameter coordination rule base to automatically select the most suitable parameter optimization method for different combinations of terminal specifications and crimping depths. Among them, size compensation ensures the detection accuracy of terminals of different specifications, depth enhancement improves the recognition ability of crimping depth, edge sharpening improves the boundary positioning effect, and noise suppression reduces interference information. Finally, combined with the correction of the brightness adaptive adjustment factor, personalized image processing parameters are formed that simultaneously adapt to changes in illumination, terminal specifications, and crimping processes, thereby improving the adaptability and detection accuracy of the detection system under complex working conditions. In one embodiment, after the step of performing parameter matching in a preset dynamic threshold adjustment strategy library based on the image brightness parameters to calculate the brightness adaptive adjustment factor, the method further includes: Detect whether the rate of change of the image brightness parameter exceeds a preset rate of change threshold; When the rate of change exceeds the preset rate of change threshold, a fast response mode is activated to shorten the parameter adjustment cycle to the preset fast response cycle. In the fast response mode, multiple consecutive frames of images are acquired and fused, and features are re-extracted based on the fused images.

[0010] In the above implementation, when the brightness change rate is detected to exceed the preset threshold, the system can automatically switch to a fast response mode, thereby improving the system's response speed to changes in illumination by shortening the parameter adjustment cycle; by using a multi-frame image fusion processing mechanism to reduce random noise and illumination unevenness in a single frame image, and by integrating temporal information to improve the stability and accuracy of feature extraction, the robustness and practicality of the system are enhanced. In one embodiment, prior to the step of extracting terminal pressing region features using a pre-trained deep convolutional neural network model, the method further includes: Based on the preset terminal material identifier, the terminal material is identified to obtain the material type; Based on the material type, retrieve the corresponding network configuration parameters from the pre-established multi-material neural network parameter library; The input layer normalization parameters and convolutional layer weights of the deep convolutional neural network model are adjusted based on the network configuration parameters.

[0011] In the above embodiments, terminals made of different materials will exhibit different optical properties and texture features during imaging. By adjusting the input layer normalization parameters, the image feature distribution of different materials can be processed in a targeted manner, while the adjustment of the convolutional layer weights ensures that the network can better extract the key features of terminals made of various materials; thus improving the feature extraction accuracy and generalization ability of the neural network model in multi-material terminal detection scenarios. In one embodiment, after the step of generating the final terminal crimping quality test result, the method further includes: Collect detection data within a preset time period, statistically analyze the deviation trend between the terminal pressing shape and the standard template, and establish a pressing tool wear compensation model based on the deviation trend; The standard template is dynamically corrected and the quality inspection benchmark parameters are updated based on the pressure tool wear compensation model. When the detected tool wear exceeds a preset threshold, a maintenance recommendation for the pressing equipment is marked in the quality inspection result.

[0012] In the above embodiments, this application can identify systematic deviations caused by tool wear by statistically analyzing the deviation trend between the terminal pressing shape and the standard template within a preset time period, and distinguish them from random quality problems. The mechanism of dynamically correcting the standard template and updating the detection benchmark parameters ensures that the quality inspection standard can be adaptively adjusted with changes in equipment status, reducing misjudgments and missed detections caused by tool wear. When the degree of tool wear reaches a preset threshold, the system can proactively provide equipment maintenance suggestions, improving the continuity and stability of the production line. In a second aspect, embodiments of this application provide a terminal pressing imaging quality detection system, comprising: one or more processors and a memory; the memory is coupled to one or more processors, the memory is used to store computer program code, the computer program code including computer instructions, and one or more processors call the computer instructions to cause the system to perform the method described in the first aspect and any possible implementation thereof. Thirdly, embodiments of this application provide a computer-readable storage medium including instructions that, when executed on a system, cause the system to perform the method described in the first aspect and any possible implementation thereof. Fourthly, embodiments of this application provide a computer program product that, when run on a system, causes the system to perform the method described in the first aspect and any possible implementation thereof. One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages: 1. This application provides a terminal crimping imaging quality detection method, which combines deep learning feature extraction with image processing technology. Through the synergistic effect of a dynamic threshold adjustment strategy library and an edge detection morphological processing rule library, a complete adaptive detection process is formed. This application can automatically adjust the detection parameters according to real-time image brightness conditions and terminal specification information, ensuring stable detection accuracy under different lighting environments and diverse terminal types. It reduces the problems of unstable detection accuracy and poor versatility caused by fixed parameters in the prior art, and improves the reliability and practicality of terminal crimping quality detection.

[0013] 2. This application provides a terminal pressing imaging quality detection method. Based on four coordination strategies of the terminal pressing detection parameter coordination rule library, it can automatically select the most suitable parameter optimization method for different combinations of terminal specifications and pressing depths. Among them, size compensation ensures the detection accuracy of terminals of different specifications, depth enhancement improves the recognition ability of pressing depth, edge sharpening improves the boundary positioning effect, and noise suppression reduces interference information. Finally, combined with the correction of the brightness adaptive adjustment factor, personalized image processing parameters are formed to adapt to changes in illumination, terminal specifications, and pressing process, thereby improving the adaptability and detection accuracy of the detection system under complex working conditions.

[0014] 3. This application provides a terminal pressing imaging quality detection method. When the brightness change rate is detected to exceed a preset threshold, the system can automatically switch to a fast response mode, thereby improving the system's response speed to illumination changes by shortening the parameter adjustment cycle. The system reduces random noise and illumination unevenness in single-frame images through a multi-frame image fusion processing mechanism, and improves the stability and accuracy of feature extraction by integrating temporal information, thereby enhancing the robustness and practicality of the system. Attached Figure Description

[0015] Figure 1 This is a schematic flowchart of a terminal pressing imaging quality detection method in an embodiment of this application.

[0016] Figure 2 This is another schematic flowchart of a terminal pressing imaging quality detection method in an embodiment of this application.

[0017] Figure 3 This is a schematic diagram of the physical device structure of a terminal pressing imaging quality detection system provided in an embodiment of this application. Detailed Implementation

[0018] The terminology used in the following embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification and appended claims of this application, the singular expressions “a,” “an,” “the,” “the,” “the,” and “this” are intended to include the plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this application refers to any or all possible combinations including one or more of the listed items.

[0019] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature, and in the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more. In the fields of electronics manufacturing and connector production, the quality of the terminal crimping process directly affects the reliability of electrical connections and the overall performance of the product. As electronic products develop towards miniaturization and high density, the precision requirements for the terminal crimping process are becoming increasingly stringent, and traditional manual inspection methods can no longer meet the needs of modern industrial production.

[0020] In related technologies, machine vision inspection methods based on fixed thresholds are mainly used to analyze the terminal pressing area through preset image processing parameters. However, these methods suffer from unstable detection accuracy and are unable to adapt to different lighting conditions and changes in terminal specifications, which seriously affects the accuracy and versatility of the inspection.

[0021] This application is primarily applied to production scenarios requiring high-reliability electrical connections, such as automotive wiring harnesses, electronic device connectors, and aerospace cables, especially for real-time online inspection of terminal crimping quality in automated production lines. In these applications, the diverse terminal types, complex crimping processes, and frequent changes in lighting conditions place extremely high demands on the adaptability and accuracy of the inspection technology. To address these technical problems, this application provides a terminal crimping imaging quality inspection method, system, medium, and product. An embodiment is described below in conjunction with… Figure 1 The following describes a terminal pressing imaging quality detection method according to an embodiment of this application: Please see Figure 1 This is a flowchart illustrating a terminal pressing imaging quality detection method in an embodiment of this application.

[0022] S101: Receive image data input during the terminal pressing process, perform image preprocessing and brightness analysis, extract terminal pressing area features through a pre-trained deep convolutional neural network model, and calculate the pressing feature vector and image brightness parameters.

[0023] In this embodiment, image preprocessing refers to the process of denoising, enhancing, and standardizing the original image. Brightness analysis quantifies the illumination characteristics of the image by calculating the average gray value, standard deviation, and histogram distribution. The pressing feature vector is a high-dimensional numerical vector output by the deep convolutional neural network model, used to characterize the shape, texture, and boundary features of the terminal pressing area. The deep convolutional neural network model pre-establishes a feature mapping table for the terminal pressing area. This mapping table records the correspondence between image features and quality levels under different pressing states, including three main categories: acceptable pressing, excessive pressing, and insufficient pressing.

[0024] S102. Based on the image brightness parameters, perform parameter matching in the preset dynamic threshold adjustment strategy library to calculate the brightness adaptive adjustment factor.

[0025] In this embodiment, the dynamic threshold adjustment strategy library is a pre-established structured database that stores image processing parameter optimization strategies under different brightness conditions. The brightness adaptive adjustment factor is a correction coefficient calculated based on the deviation between the current image brightness and the standard brightness, which is used to dynamically adjust the parameter settings of subsequent image processing algorithms.

[0026] Specifically, the system first calculates the Euclidean distance between the image brightness parameters and each standard brightness condition in the dynamic threshold adjustment strategy library, selecting the brightness condition with the smallest distance as the matching benchmark. Then, based on the brightness deviation coefficient calculation rules corresponding to the matching benchmark, the system performs a difference operation between the current image brightness parameters and the preset standard brightness intensity to obtain a brightness change quantification index. Finally, the brightness change quantification index is input into the corresponding nonlinear mapping function in the strategy library to calculate the adjustment multiples of the image binarization threshold, contrast enhancement parameter, and brightness correction parameter. The dynamic threshold adjustment strategy library pre-establishes a brightness deviation compensation table, which records the parameter adjustment rules under brightness change ranges from 0.2 to 2.5 times the standard brightness condition. Through a combination of piecewise linear interpolation and an exponential decay function, the specific adjustment multiples of the binarization threshold, contrast enhancement, and brightness correction are calculated.

[0027] S103. Read the terminal specification information and pressing process standards, perform feature analysis, and calculate personalized image processing parameters based on the preset edge detection and morphological processing rule library and the brightness adaptive adjustment factor.

[0028] In this embodiment, the terminal specification information includes the terminal's length, width, thickness, and material type. The pressing process standards include the technical requirements for pressing depth, pressing angle, and pressing force. The edge detection and morphological processing rule base is a pre-established parameter configuration database that stores the optimal combination of image processing parameters for different terminal types and process requirements.

[0029] Specifically, the system pre-establishes a mapping table between terminal specifications and processing parameters. After reading the current terminal specification information, the system queries the corresponding baseline parameters from the rule base, and then fine-tunes the parameters in conjunction with the pressing process standards.

[0030] S104. Based on personalized image processing parameters, edge detection and morphological processing are used to locate the boundary of the terminal pressing area to obtain the pressing boundary contour.

[0031] In this embodiment, edge detection is an image processing technique that identifies target boundaries by calculating image gradients and applying threshold segmentation. Morphological processing is a method that uses mathematical morphological principles to perform opening, closing, erosion, and dilation operations on the image. The overlapping boundary contour is a set of coordinate points describing the shape of the outer boundary of the overlapping region of the terminal. The system pre-establishes an edge detection operator selection table and selects the corresponding operator for edge detection based on different image characteristics and noise levels.

[0032] S105. Based on the pressing boundary profile and pressing feature vector, perform quality assessment analysis to generate the final terminal pressing quality inspection results.

[0033] In this embodiment, the quality assessment analysis is an analytical process that combines the geometric features of the boundary contour and deep learning feature vectors to judge the quality. The terminal pressing quality inspection results include quality level determination, defect type identification, and confidence score. The system has pre-established a quality assessment rule base, which contains judgment criteria for three main assessment dimensions: shape regularity, dimensional conformity, and surface quality. Each dimension has clearly defined quantitative indicators and threshold ranges.

[0034] Specifically, the system first extracts geometric features from the pressing boundary contour, calculating the contour's perimeter, area, roundness, and rectangularity indices. Then, it compares and analyzes these features with a standard template to derive the shape deviation value. Quality assessment employs a combination of multi-level decision trees and support vector machines, using the contour's geometric features and a 512-dimensional pressing feature vector as input. A pre-trained classification model outputs a quality level. Simultaneously, the system establishes a defect type identification database, including four main defect types: insufficient pressing, excessive pressing, offset pressing, and surface scratches. Each defect corresponds to a specific feature pattern and discrimination threshold. The final detection results are output in structured data format.

[0035] In one embodiment, terminal specification information and pressing process standards are read, feature analysis is performed, and personalized image processing parameters are calculated based on a preset edge detection and morphological processing rule library, combined with a brightness adaptive adjustment factor. Specifically, this includes: extracting terminal size features based on terminal specification information and pressing depth features based on pressing process standards; selecting corresponding baseline image processing parameters from the preset edge detection and morphological processing rule library based on terminal size features and pressing depth features; coordinating and optimizing the baseline image processing parameters based on a preset terminal pressing detection parameter coordination rule library to obtain a parameter coordination scheme; the terminal pressing detection parameter coordination rule library refers to a structured database that records image processing parameter coordination strategies under different combinations of terminal specifications and pressing depths, and the coordination strategies include four types: size compensation, depth enhancement, edge sharpening, and noise suppression; multiplying the brightness adaptive adjustment factor with the coordinated and optimized image processing parameters yields personalized image processing parameters.

[0036] The terminal size features refer to the quantified geometric parameters extracted from the terminal specification information, including the numerical range of terminal length, width, and thickness, and the terminal cross-sectional area. The pressing depth feature is the pressing deformation requirement determined according to the pressing process standard, expressed as the percentage change in terminal thickness before and after pressing. The edge detection and morphological processing rule base is a pre-built parameter configuration database that stores the optimal image processing parameter combinations for different terminal specifications, including standardized configurations of edge detection operator type, threshold setting, and morphological kernel size. The terminal pressing detection parameter coordination rule base adopts a multi-dimensional data table structure, using terminal specification and pressing depth as index keys, recording the coordination relationships and optimization strategies between parameters. Among them, the size compensation strategy is used to adjust the detection sensitivity of terminals of different sizes, the depth enhancement strategy optimizes gradient detection parameters for different pressing depths, the edge sharpening strategy improves boundary positioning accuracy, and the noise suppression strategy reduces the impact of environmental interference.

[0037] Specifically, the system first parses the geometric parameters in the terminal specification information, converting the terminal length, width, and thickness into normalized dimensional feature vectors. Simultaneously, it calculates the target deformation rate based on the pressing depth requirements in the pressing process standards, forming pressing depth feature parameters. The edge detection and morphological processing rule base adopts a two-level index structure. The first level performs coarse classification based on the terminal cross-sectional area range, and the second level performs fine classification based on the aspect ratio. The system queries the corresponding baseline image processing parameters using a feature matching algorithm. The terminal pressing detection parameter coordination rule base coordinates parameters through a preset decision tree model. Based on the combination of terminal size features and pressing depth features, it automatically selects the appropriate coordination strategy and calculates parameter correction coefficients. Finally, it uses the brightness adaptive adjustment factor as a global correction factor, multiplies it with the coordinated parameters, and generates a personalized image processing parameter set adapted to the current detection conditions.

[0038] In some embodiments, optionally, terminal size feature extraction can employ a fast lookup table method based on terminal type encoding, pre-establishing a direct mapping relationship between terminal models and size features to avoid real-time calculations and improve processing efficiency. Optionally, the edge detection and morphological processing rule base can adopt a hierarchical storage structure, storing parameters of commonly used terminal specifications in a high-speed cache and parameters of less commonly used specifications in an extended database, dynamically adjusting the cache content based on access frequency statistics.

[0039] In some embodiments, there may be cases where the combination of terminal specifications and pressing depth exceeds the coverage of the rule base. In this regard, this application adopts the nearest neighbor interpolation algorithm to select the most similar parameter combination in the rule base as the benchmark, and calculates suitable coordination parameters through linear interpolation.

[0040] In the above embodiments, the coordinated use of a preset dynamic threshold adjustment strategy library and an edge detection morphological processing rule library enables adaptive parameter adjustment for different terminal specifications and pressing processes, improving the adaptability and accuracy of the detection system under complex working conditions. By utilizing four strategy types in the terminal pressing detection parameter coordination rule library, the system can specifically address detection challenges under different terminal sizes and pressing depth combinations. Combined with dynamic correction of the brightness adaptability adjustment factor, it ensures optimal configuration of detection parameters under various lighting conditions.

[0041] In the above embodiments, although the basic detection process can meet the needs of most application scenarios, in actual production environments, problems such as drastic changes in lighting conditions, mixed production of terminals made of multiple materials, and accuracy drift caused by long-term operation of production equipment may be encountered. However, the above basic embodiments do not fully consider the impact of these complex working conditions on detection accuracy and system stability. To further improve the robustness and practicality of the system, this application also provides another terminal pressing imaging quality detection method. The following is in conjunction with... Figure 2 Another terminal pressing imaging quality detection method in the embodiments of this application is described below: Please see Figure 2 This is another flowchart illustrating a terminal pressing imaging quality detection method in an embodiment of this application.

[0042] S201. Identify the terminal material according to the preset terminal material identifier to obtain the material type; retrieve the corresponding network configuration parameters from the pre-established multi-material neural network parameter library according to the material type; adjust the input layer normalization parameters and convolutional layer weights of the deep convolutional neural network model based on the network configuration parameters.

[0043] In this embodiment, the terminal material identification is achieved through barcode scanning, RFID reading, or production planning system-acquired material information encoding. The material types include four main categories: copper, tin-plated, silver-plated, and aluminum. The multi-material neural network parameter library is a pre-established database of material-specific parameter configurations, storing the optimal parameter combinations for neural network models under different material types. The network configuration parameters include the normalized mean and standard deviation of the input layer, and the weight matrix and bias vector of the convolutional layer. These parameters are specifically optimized for the optical properties and imaging characteristics of different materials.

[0044] Specifically, the system pre-establishes a material type identification mapping table, which records the codes of different materials and their corresponding optical characteristic parameters, including reflectivity range, color features, and texture patterns. The multi-material neural network parameter library adopts a hierarchical storage structure, maintaining an independent parameter set for each material type and automatically selecting the corresponding network configuration based on the material identification results. The system dynamically adjusts the input layer normalization parameters of the neural network model according to the material type, unifying the image data of different materials to the same numerical distribution range. Simultaneously, it updates the weight parameters of the convolutional layers to adapt to the feature extraction requirements of different materials, ensuring that the neural network maintains optimal feature extraction performance when processing multiple material terminals.

[0045] S202: Receive image data input during the terminal pressing process, perform image preprocessing and brightness analysis, extract terminal pressing area features through an adjusted deep convolutional neural network model, and calculate pressing feature vector and image brightness parameters.

[0046] S203. Based on the image brightness parameters, perform parameter matching in the preset dynamic threshold adjustment strategy library to calculate the brightness adaptive adjustment factor.

[0047] S204. Detect whether the rate of change of the image brightness parameter exceeds the preset rate of change threshold; when the rate of change exceeds the preset rate of change threshold, start the fast response mode and shorten the parameter adjustment cycle to the preset fast response cycle; in the fast response mode, acquire multiple consecutive frames of images for fusion processing, and re-extract features based on the fused image.

[0048] In this embodiment, the rate of change of image brightness parameters is an index obtained by calculating the ratio of the brightness difference between consecutive frames to the time interval. The preset rate of change threshold is set according to the lighting stability requirements of the production environment. The fast response mode is a special processing mode activated by the system when it detects a sudden change in lighting. The preset fast response period shortens the normal parameter adjustment interval from the standard period to one-quarter to one-half of the original. The multi-frame image fusion processing uses a combination of weighted averaging and temporal filtering to reduce the impact of lighting changes on detection accuracy.

[0049] Specifically, the system establishes a brightness change rate monitoring mechanism, calculating the rate of change of brightness parameters between the current frame and the previous frame in real time. When the rate of change exceeds a preset threshold, a fast response mode is automatically triggered. In fast response mode, the system shortens the parameter adjustment cycle and simultaneously initiates a multi-frame fusion algorithm to acquire 3 to 5 consecutive frames for weighted fusion. The fusion weights are dynamically allocated based on the quality scores of each frame, which are calculated comprehensively based on image sharpness, brightness uniformity, and noise level. The fused image exhibits better brightness stability and feature consistency. The system re-extracts features based on the fused image, ensuring stable detection accuracy even under drastic lighting conditions.

[0050] S205: Read terminal specification information and pressing process standards, perform feature analysis, and calculate personalized image processing parameters based on the preset edge detection and morphological processing rule library and the brightness adaptive adjustment factor.

[0051] S206. Based on personalized image processing parameters, edge detection and morphological processing are used to locate the boundary of the terminal pressing area to obtain the pressing boundary contour.

[0052] S207. Based on the pressing boundary profile and pressing feature vector, perform quality assessment analysis to generate the final terminal pressing quality inspection results.

[0053] S208. Collect the test data within a preset time period, statistically analyze the deviation trend between the terminal pressing shape and the standard template, and establish a pressing tool wear compensation model based on the deviation trend; dynamically correct the standard template according to the pressing tool wear compensation model and update the quality inspection benchmark parameters; when the identified tool wear degree exceeds the preset threshold, mark the pressing equipment maintenance suggestion in the quality inspection results.

[0054] In this embodiment, the pressing tool wear compensation model is a predictive model based on statistical learning methods, used to describe the relationship between the degree of pressing tool wear and detection deviation. The preset time period refers to a pre-set production cycle. Deviation trend analysis identifies the degree of tool wear by calculating the systematic deviation changes between the pressing shape parameters and the standard template. The dynamic correction of the standard template is a process of adjusting the detection benchmark in real time based on the prediction results of the wear compensation model.

[0055] Specifically, the system establishes a data collection and analysis mechanism, continuously recording the geometric parameters of the terminal crimping shape within a preset time period, including crimping depth, width, and symmetry indicators. Statistical analysis methods are used to identify the systematic deviation trends of these parameters relative to the standard template. The crimping tool wear compensation model employs an algorithm combining linear regression and exponential smoothing. It predicts the tool wear degree based on the rate of change of the deviation trend and the cumulative deviation, generating corresponding compensation parameters. The system dynamically adjusts the geometric parameters of the standard template and the quality inspection threshold settings based on the compensation model's output. When the calculated tool wear degree exceeds a preset threshold percentage, the system automatically adds a maintenance suggestion to the inspection results, reminding operators to replace or maintain the tool promptly, achieving a system upgrade from passive detection to proactive prevention.

[0056] In the above embodiments, by introducing a material adaptive recognition mechanism, a rapid response mode for changes in illumination, and equipment status monitoring, the adaptability and stability of the terminal pressing imaging quality inspection system in complex production environments are improved. Dynamic access to a multi-material neural network parameter library ensures accurate identification of terminals made of different materials. The rapid response mode effectively addresses drastic changes in illumination conditions by shortening the parameter adjustment cycle and employing multi-frame fusion processing. Furthermore, the establishment of a pressing tool wear compensation model enables a shift from passive detection to proactive prevention. This embodiment not only improves the material compatibility, environmental adaptability, and long-term stability of the basic solution but also reduces detection errors caused by environmental changes and equipment wear through real-time monitoring and dynamic adjustment mechanisms, thereby improving overall production efficiency and product quality consistency. The system in the embodiments of this invention is described below from the perspective of hardware processing. Please refer to [link / reference needed]. Figure 3 This is a schematic diagram of the physical device structure of a terminal pressing imaging quality detection system provided in an embodiment of this application.

[0057] It should be noted that, Figure 3 The structure of the system shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the present invention.

[0058] like Figure 3 As shown, the system includes a CPU, which can perform various appropriate actions and processes based on a program stored in the ROM or a program loaded into the RAM from a storage portion, such as executing the methods described in the above embodiments. The RAM also stores various programs and data required for system operation. The CPU, ROM, and RAM are interconnected via a bus. I / O interfaces are also connected to the bus.

[0059] The following components are connected to the I / O interface: input sections including cameras, infrared sensors, etc.; output sections including liquid crystal displays (LCDs) and speakers, etc.; storage sections including hard drives, etc.; and communication sections including network interface cards such as LAN (Local Area Network) cards and modems, etc. The communication section performs communication processing via a network such as the Internet. Drives are also connected to the I / O interface as needed. Removable media, such as disks, optical disks, magneto-optical disks, semiconductor memories, etc., are installed on the drive as needed so that computer programs read from them can be installed into the storage section as needed.

[0060] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing computer programs for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication component, and / or installed from a removable medium. When the computer program is executed by a CPU, it performs the various functions defined in the present invention.

[0061] It should be noted that the computer-readable medium shown in the embodiments of the present invention can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In the present invention, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, wherein a computer-readable computer program is carried. The transmitted data signal can take many forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof.

[0062] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. Each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0063] In another aspect, the present invention also provides a computer-readable storage medium, which may be included in the system described in the above embodiments; or it may exist independently and not assembled into the system. The storage medium carries one or more computer programs that, when executed by a processor of a system, cause the system to implement the methods provided in the above embodiments.

[0064] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit it. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

[0065] As used in the above embodiments, depending on the context, the term "when..." can be interpreted as "if...", "after...", "in response to determining...", or "in response to detecting...". Similarly, depending on the context, the phrase "when determining..." or "if (the stated condition or event) is interpreted as "if determining...", "in response to determining...", "when (the stated condition or event) is detected", or "in response to detecting (the stated condition or event)".

[0066] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive), etc.

[0067] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A method for detecting the imaging quality of terminal pressing, characterized in that, include: The system receives image data input during the terminal pressing process, performs image preprocessing and brightness analysis, extracts terminal pressing area features through a pre-trained deep convolutional neural network model, and calculates pressing feature vectors and image brightness parameters. Based on the image brightness parameters, parameter matching is performed in a preset dynamic threshold adjustment strategy library to calculate the brightness adaptive adjustment factor. Read the terminal specification information and pressing process standards, perform feature analysis, and calculate personalized image processing parameters based on the preset edge detection and morphological processing rule library and the brightness adaptive adjustment factor. Based on the personalized image processing parameters, edge detection and morphological processing are used to locate the boundary of the terminal pressing area to obtain the pressing boundary contour. Based on the pressing boundary contour and the pressing feature vector, a quality assessment analysis is performed to generate the final terminal pressing quality inspection result.

2. The method according to claim 1, characterized in that, The step of matching parameters in a preset dynamic threshold adjustment strategy library based on the image brightness parameters and calculating the brightness adaptive adjustment factor specifically includes: Calculate the Euclidean distance between the image brightness parameter and each standard brightness condition in the dynamic threshold adjustment strategy library, and select the brightness condition with the smallest distance as the matching benchmark. Based on the brightness deviation coefficient calculation rule corresponding to the matching benchmark, the difference calculation is performed between the current image brightness parameter and the preset standard brightness intensity to obtain the brightness change quantification index. The brightness change quantification index is input into the corresponding nonlinear mapping function in the strategy library to calculate the adjustment factor of the image binarization threshold, contrast enhancement parameter and brightness correction parameter.

3. The method according to claim 2, characterized in that, The process involves reading terminal specification information and pressing process standards, performing feature analysis, and calculating personalized image processing parameters based on a preset edge detection and morphological processing rule library, combined with the brightness adaptive adjustment factor. Specifically, this includes: Extract terminal size features based on terminal specification information, and extract crimping depth features based on crimping process standards; Based on the terminal size features and pressing depth features, the corresponding reference image processing parameters are selected from the preset edge detection and morphological processing rule library; Based on a preset terminal pressing detection parameter coordination rule library, the reference image processing parameters are coordinated and optimized to obtain a parameter coordination scheme. The terminal pressing detection parameter coordination rule library is a structured database that records the image processing parameter coordination strategies under different combinations of terminal specifications and pressing depths. The coordination strategies include four types: size compensation, depth enhancement, edge sharpening, and noise suppression. The brightness adaptive adjustment factor is multiplied by the coordinated and optimized image processing parameters to obtain personalized image processing parameters.

4. The method according to claim 1, characterized in that, After the step of matching parameters in a preset dynamic threshold adjustment strategy library based on the image brightness parameters to calculate the brightness adaptive adjustment factor, the method further includes: Detect whether the rate of change of the image brightness parameter exceeds a preset rate of change threshold; When the rate of change exceeds the preset rate of change threshold, a fast response mode is activated to shorten the parameter adjustment cycle to the preset fast response cycle. In the fast response mode, multiple consecutive frames of images are acquired and fused, and features are re-extracted based on the fused images.

5. The method according to claim 1, characterized in that, Before the step of extracting terminal pressing region features using a pre-trained deep convolutional neural network model, the method further includes: Based on the preset terminal material identifier, the terminal material is identified to obtain the material type; Based on the material type, retrieve the corresponding network configuration parameters from the pre-established multi-material neural network parameter library; The input layer normalization parameters and convolutional layer weights of the deep convolutional neural network model are adjusted based on the network configuration parameters.

6. The method according to claim 1, characterized in that, After the step of generating the final terminal crimping quality test result, the method further includes: Collect detection data within a preset time period, statistically analyze the deviation trend between the terminal pressing shape and the standard template, and establish a pressing tool wear compensation model based on the deviation trend; The standard template is dynamically corrected and the quality inspection benchmark parameters are updated based on the pressure tool wear compensation model. When the detected tool wear exceeds a preset threshold, a maintenance recommendation for the pressing equipment is marked in the quality inspection result.

7. A terminal pressing imaging quality detection system, characterized in that, The system includes: One or more processors and a memory; the memory is coupled to the one or more processors, the memory being used to store computer program code, the computer program code including computer instructions, the one or more processors invoking the computer instructions to cause the system to perform the method as described in any one of claims 1-6.

8. A computer-readable storage medium comprising instructions, characterized in that, When the instructions are executed on the system, the system performs the method as described in any one of claims 1-6.

9. A computer program product, characterized in that, When the computer program product is run on the system, the system performs the method as described in any one of claims 1-6.

Images