Bonding copper wire winding automatic abnormity monitoring method based on mechanical vision

Through the automatic abnormality monitoring method of bonded copper wire winding based on mechanical vision, the problem of inaccurate monitoring during the winding of high-speed bonded copper wire is solved, and high accuracy monitoring of the winding process of bonded copper wire is achieved, reducing the number of Rework and scrapping have improved the quality of semiconductor packaging products.

CN118943046AInactive Publication Date: 2024-11-12SHENZHEN ZHONGBAO NEW MATERIAL TECH CO LTD
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202410953542.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-16
Publication Date
2024-11-12
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The prior art is difficult to accurately monitor abnormalities during the winding process of high-speed bonded copper wire, resulting in poor quality of semiconductor packaging products and high rework and scrapping costs.

Method used

Using a method based on mechanical vision, we use the method to acquire and preprocess the wound images, perform edge detection and background separation, identify copper wire winding parameters, build a normal trajectory model, set an abnormality detection threshold, identify abnormal areas, and generate an abnormality monitoring report.

Benefits of technology

It improves the accuracy of automatic abnormality monitoring of bonded copper wire winding, can identify abnormal areas in real time, determine the root cause of problems, reduce rework and scrapping, and improve production efficiency and product quality.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118943046A_ABST
    Figure CN118943046A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of bonding copper wire winding monitoring, and provides a bonding copper wire winding automatic abnormity monitoring method based on mechanical vision, and the method comprises the steps: collecting a winding image of a bonding copper wire in a winding process, and carrying out the preprocessing of the winding image, and obtaining a preprocessing image; performing edge detection on the target copper wire in the preprocessed image to obtain an edge detection result, performing background separation processing on the bonding copper wire based on the edge detection result to obtain an independent copper wire image, identifying a copper wire winding parameter of the independent copper wire image, and evaluating a winding effect of the bonding copper wire; calculating a matching coefficient of the copper wire winding parameter and the abnormal detection threshold value, and when the matching coefficient is greater than a preset matching coefficient, identifying an abnormal area of the bonding copper wire in the winding process; and determining an abnormal monitoring report of the bonding copper wire in the winding process, and outputting the abnormal monitoring report by a visual notification mechanism. According to the invention, the accuracy of automatic abnormal monitoring of bonding copper wire winding is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of bonding copper wire winding monitoring, and in particular to a method for automatic abnormality monitoring of bonding copper wire winding based on machine vision. Background Art

[0002] Abnormal bonding copper wire winding monitoring refers to the real-time monitoring and identification of abnormal problems that occur during the winding of the bonding copper wire during the semiconductor packaging process. Abnormal bonding copper wire winding monitoring can avoid the production of defective products, thereby improving the overall yield of semiconductor packaging and reducing rework and scrap costs. For example, installing a force sensor on the bonding head can control the tension of the bonding copper wire during the winding process in real time. If the tension exceeds the preset range, the sensor will send a signal to indicate that adjustment is needed.

[0003] At present, in the semiconductor packaging industry, bonding copper wire winding abnormality monitoring is an important monitoring technology and is widely used in microelectronics and integrated circuits to ensure the product quality of semiconductor packaging. However, due to the continuous increase in bonding speed during the bonding copper wire winding abnormality monitoring process, the image capture and image analysis capabilities provided by the monitoring equipment can no longer keep up with the rhythm of high-speed bonding, resulting in inaccurate monitoring of the bonding copper wire during the winding process. Summary of the invention

[0004] The present invention provides a method for automatic abnormality monitoring of bonding copper wire winding based on machine vision, and its main purpose is to improve the accuracy of automatic abnormality monitoring of bonding copper wire winding.

[0005] To achieve the above object, the present invention provides a method for automatic abnormality monitoring of bonding copper wire winding based on machine vision, comprising:

[0006] Acquiring a bonding copper wire to be wound, collecting a winding image of the bonding copper wire during the winding process, and preprocessing the winding image to obtain a preprocessed image;

[0007] Performing edge detection on the target copper wire in the preprocessed image to obtain an edge detection result, performing background separation processing on the bonding copper wire based on the edge detection result to obtain an independent copper wire image, identifying copper wire winding parameters of the independent copper wire image, and evaluating the winding effect of the bonding copper wire based on the copper wire winding parameters;

[0008] When the winding effect is greater than the expected effect, a normal trajectory model of the bonding copper wire during the winding process is constructed, and an abnormality detection threshold of the normal trajectory model is set, and a matching coefficient between the copper wire winding parameter and the abnormality detection threshold is calculated, and when the matching coefficient is greater than a preset matching coefficient, an abnormal area of ​​the bonding copper wire during the winding process is identified;

[0009] Analyze the abnormal factors of the abnormal area, determine the abnormal monitoring report of the bonding copper wire during the winding process based on the abnormal factors, construct a visual notification mechanism for the abnormal monitoring report, and output the abnormal monitoring report through the visual notification mechanism.

[0010] Optionally, the collecting a winding image of the bonding copper wire during the winding process includes:

[0011] Querying an image acquisition device of the bonding copper wire during the winding process, and setting a lighting condition of the image acquisition device;

[0012] Analyzing a winding cycle of the copper bonding wire, and setting an image capture trigger mechanism of the copper bonding wire during the winding process based on the winding cycle;

[0013] Under the lighting condition, based on the image capture trigger mechanism, a winding image of the bonding copper wire during the winding process is collected.

[0014] Optionally, analyzing the winding period of the bonding copper wire includes:

[0015] querying the material properties of the bonding copper wire, and measuring the winding parameters of the bonding copper wire based on the material properties;

[0016] Based on the winding parameters, a time series of the winding of the bonding copper wire is constructed to obtain the winding period of the bonding copper wire.

[0017] Optionally, performing background separation processing on the bonding copper wire based on the edge detection result to obtain an independent copper wire image includes:

[0018] Based on the edge detection result, determining the copper wire contour of the preprocessed image;

[0019] Performing denoising processing on the copper wire profile to obtain a denoised copper wire profile;

[0020] The foreground of the denoised copper wire contour is extracted to obtain the independent copper wire image.

[0021] Optionally, the identifying the copper wire winding parameters of the independent copper wire image includes:

[0022] Acquire a valid line segment of the independent copper wire image, and measure an appearance coefficient of the valid line segment;

[0023] Based on the appearance coefficient, the winding angle of the copper wire winding is calculated using the following formula:

[0024] θ=arctan(Δy / Δx)

[0025] Among them, θ represents the winding angle of the copper wire, Δy represents the height change of the copper wire when it is winding up or down, and Δx represents the straight-line distance that the copper wire moves on the winding axis.

[0026] The winding density of the copper wire winding is calculated by the following formula:

[0027]

[0028] Among them, ρ represents the copper wire winding density, m represents the total mass of the copper wire wound by the copper wire winding, D represents the outer diameter of the copper wire after winding, d represents the inner diameter of the winding drum of the copper wire winding, H represents the height of the copper wire winding,

[0029] The copper wire winding parameters of the independent copper wire image are generated according to the winding angle and the winding density.

[0030] Optionally, the evaluating the winding effect of the bonding copper wire based on the copper wire winding parameter includes:

[0031] Based on the copper wire winding parameters, constructing a winding standard for the bonding copper wire;

[0032] Based on the winding standard, analyzing the degree of structural deformation of the bonding copper wire;

[0033] When the degree of structural deformation meets the winding standard, the structural stability of the bonding copper wire is evaluated to obtain an evaluation result.

[0034] Optionally, the constructing of a normal trajectory model of the bonding copper wire during the winding process includes:

[0035] Constructing a simulation model of the winding scene;

[0036] Setting boundary conditions of the simulation model, and based on the boundary conditions, identifying movement behavior data of the bonding copper wire during the winding process;

[0037] The behavior results corresponding to the motion behavior data are analyzed to obtain the normal trajectory model.

[0038] Optionally, the identifying an abnormal area of ​​the bonding copper wire during the winding process includes:

[0039] Analyzing the variation trend of the copper wire winding parameters;

[0040] Based on the change trend, constructing a winding abnormality time series of the bonding copper wire, and analyzing an image abnormality time series of the winding image;

[0041] After comparing the abnormal winding positions of the abnormal winding time series and the abnormal image time series, an abnormal area of ​​the bonding copper wire during the winding process is determined.

[0042] Optionally, analyzing the abnormal factor of the abnormal area includes:

[0043] Checking abnormal characteristics of the abnormal area;

[0044] Based on the abnormal features, collecting abnormal data of the abnormal area; classifying the abnormal data to obtain classified data;

[0045] Based on the classified data, an abnormal factor of the abnormal area is analyzed.

[0046] Optionally, determining an abnormal monitoring report of the bonding copper wire during the winding process based on the abnormal factor includes:

[0047] Setting the visualization type of the abnormal monitoring report and configuring the notification mechanism of the visualization type;

[0048] Based on the notification mechanism, the abnormal monitoring report is output.

[0049] The embodiment of the present invention obtains the bonding copper wire to be wound, collects the winding image of the bonding copper wire during the winding process, and pre-processes the winding image to obtain a pre-processed image, which can remove the interference elements of the bonding copper wire winding image to make the image clearer. The embodiment of the present invention performs edge detection on the target copper wire in the pre-processed image to obtain an edge detection result, which can provide the key features of the bonding copper wire winding image. The embodiment of the present invention constructs a normal trajectory model of the bonding copper wire during the winding process when the winding effect is greater than the expected effect, which can effectively evaluate the winding quality of the bonding copper wire. The embodiment of the present invention determines the root cause of the abnormal winding of the bonding copper wire by analyzing the abnormal factor of the abnormal area. The abnormal factor is a variable factor that deviates significantly from the normal winding mode during the winding process of the bonding copper wire. Therefore, the method of automatic abnormal monitoring of bonding copper wire winding based on machine vision provided by the embodiment of the present invention can improve the accuracy of automatic abnormal monitoring of bonding copper wire winding. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 A schematic flow chart of a method for automatic abnormality monitoring of bonding copper wire winding based on machine vision provided by one embodiment of the present invention;

[0051] Figure 2 A functional module diagram of a system for automatic abnormal monitoring of bonding copper wire winding based on machine vision provided in one embodiment of the present invention.

[0052] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings in conjunction with the embodiments. DETAILED DESCRIPTION

[0053] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0054] The embodiment of the present application provides a method for automatic abnormal monitoring of bonding copper wire winding based on mechanical vision. In the embodiment of the present application, the execution subject of the method for automatic abnormal monitoring of bonding copper wire winding based on mechanical vision includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided in the embodiment of the present application. In other words, the method for automatic abnormal monitoring of bonding copper wire winding based on mechanical vision can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc. The server can be an independent server, or it can be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and big data and artificial intelligence platforms.

[0055] Reference Figure 1 FIG. 1 is a flow chart of a method for automatic abnormality monitoring of copper bonding wire winding based on machine vision according to an embodiment of the present invention. In this embodiment, the method for automatic abnormality monitoring of copper bonding wire winding based on machine vision includes steps S1-S4.

[0056] S1. Obtain a bonding copper wire to be wound, collect a winding image of the bonding copper wire during the winding process, and pre-process the winding image to obtain a pre-processed image.

[0057] The embodiment of the present invention obtains a bonding copper wire to be wound, collects a winding image of the bonding copper wire during the winding process, and preprocesses the winding image to obtain a preprocessed image, thereby removing interference elements of the winding image of the bonding copper wire to make the image clearer. The bonding copper wire refers to a lead used to connect a semiconductor chip and a package substrate pin, and the winding image refers to an image of the bonding copper wire during the winding process composed of different image frames. The preprocessed image refers to an image that is obtained by simplifying the data and improving the quality of the winding image to meet the needs of abnormality detection.

[0058] As an embodiment of the present invention, the capturing of the winding image of the bonding copper wire during the winding process includes: querying an image capturing device of the bonding copper wire during the winding process, and setting a lighting condition of the image capturing device; analyzing the winding cycle of the bonding copper wire, and based on the winding cycle, setting an image capture trigger mechanism of the bonding copper wire during the winding process; under the lighting condition, based on the image capture trigger mechanism, capturing the winding image of the bonding copper wire during the winding process.

[0059] Among them, the image acquisition device refers to a device used to acquire, record and convert image information, such as a CCD camera, the lighting condition refers to the optical configuration condition of the image acquisition device, the winding cycle refers to the time and number of cycles required for the bonding copper wire to start winding, complete a full number of winding turns, and then return to the starting position, and the image capture trigger mechanism refers to a method of sending a signal to the image acquisition device to capture an image.

[0060] Optionally, the image acquisition device for querying the bonding copper wire during the winding process can be implemented using a high-resolution industrial camera, such as a CCD, and the setting of the lighting conditions of the image acquisition device can be obtained by adjusting the lens and the light.

[0061] Furthermore, in another optional embodiment of the present invention, the analysis of the winding period of the bonding copper wire includes: querying the material properties of the bonding copper wire, and measuring the winding parameters of the bonding copper wire based on the material properties; based on the winding parameters, constructing a time series of the winding of the bonding copper wire to obtain the winding period of the bonding copper wire.

[0062] Among them, the material properties refer to the physical properties of the copper wire that affect the winding of the bonding copper wire, including the diameter of the copper wire and the ductility of the copper wire. The winding parameters refer to the parameters that affect the winding quality of the bonding copper wire, including the winding time and the winding speed. The time series refers to the time sequence arrangement of the bonding copper wire during the winding process.

[0063] Optionally, measuring the winding parameters of the bonding copper wire based on the material properties can be achieved using a winding machine, and constructing a time series of the winding of the bonding copper wire based on the winding parameters to obtain the winding period of the bonding copper wire can be obtained by recording time intervals.

[0064] Optionally, the preprocessing of the warped image may be implemented using an image processing library, such as ImageProcessing Toolbox of MATLAB.

[0065] S2. Perform edge detection on the target copper wire in the preprocessed image to obtain an edge detection result, and based on the edge detection result, perform background separation processing on the bonding copper wire to obtain an independent copper wire image; identify the copper wire winding parameters of the independent copper wire image, and evaluate the winding effect of the bonding copper wire based on the copper wire winding parameters.

[0066] The embodiment of the present invention performs edge detection on the target copper wire in the preprocessed image to obtain an edge detection result, which can provide key features of the bonding copper wire winding image. The target copper wire refers to the processed image copper wire in the preprocessed image, and the edge detection result refers to the boundary recognition result of the object with obvious changes in brightness or color in the image.

[0067] Optionally, the edge detection of the target copper wire in the pre-processed image may be implemented using a Canny edge detection algorithm.

[0068] Furthermore, the embodiment of the present invention performs background separation processing on the bonding copper wire based on the edge detection result to obtain an independent copper wire image, which can help track the winding trajectory of the bonding copper wire. The independent copper wire image refers to a background-free copper wire image separated from the image background.

[0069] As another embodiment of the present invention, based on the edge detection result, background separation processing is performed on the bonding copper wire to obtain an independent copper wire image, including: based on the edge detection result, determining the copper wire contour of the preprocessed image; denoising the copper wire contour to obtain a denoised copper wire contour; and foreground extraction is performed on the denoised copper wire contour to obtain the independent copper wire image.

[0070] The copper wire contour refers to a pixel sequence contour with a clear boundary between the bonding copper wire and the image background, and the denoised copper wire contour refers to the copper wire contour after image noise is eliminated.

[0071] Optionally, determining the copper wire contour of the preprocessed image based on the edge detection result can be achieved by utilizing a contour tracing algorithm, denoising the copper wire contour to obtain a denoised copper wire contour can be obtained by contour smoothing, and foreground extraction of the denoised copper wire contour to obtain the independent copper wire image can be obtained by clipping.

[0072] The embodiment of the present invention can help analyze key features of the copper wire by identifying the copper wire winding parameters of the independent copper wire image, where the copper wire winding parameters refer to physical parameters of the bonding copper wire itself, such as diameter and length.

[0073] As an embodiment of the present invention, the identifying of the copper wire winding parameters of the independent copper wire image includes: obtaining the effective line segments of the independent copper wire image and measuring the appearance coefficient of the effective line segments; based on the appearance coefficient, calculating the winding angle of the copper wire winding, calculating the winding density of the copper wire winding, and generating the copper wire winding parameters of the independent copper wire image according to the winding angle and the winding density.

[0074] Among them, the effective line segment refers to the continuous and complete copper wire segment of the independent copper wire image, the appearance coefficient refers to the appearance characteristic parameters of the copper wire segment itself, such as line width and line length parameters, the winding angle refers to the angle of the copper wire relative to the winding axis during the winding process, and the winding density refers to the tightness of the copper wire winding.

[0075] Further, in an optional embodiment of the present invention, based on the appearance coefficient, the winding angle of the copper wire winding is calculated using the following formula:

[0076] θ=arctan(Δy / Δx)

[0077] Among them, θ represents the winding angle of the copper wire, Δy represents the height change of the copper wire when it is winding up or down, and Δx represents the straight-line distance that the copper wire moves on the winding axis.

[0078] Further, in an optional embodiment of the present invention, the winding density of the copper wire winding is calculated using the following formula:

[0079]

[0080] Among them, ρ represents the copper wire winding density, m represents the total mass of the copper wire in the copper wire winding, D represents the outer diameter of the copper wire after winding, d represents the inner diameter of the winding drum of the copper wire winding, and H represents the height of the copper wire winding.

[0081] Furthermore, the embodiment of the present invention can ensure that the copper wire can be wound under normal conditions and improve the winding quality of the bonding copper wire by evaluating the winding effect of the bonding copper wire based on the copper wire winding parameters. The key characteristic parameters refer to important indicator parameters that affect the winding of the bonding copper wire, such as winding spacing, winding speed and tension.

[0082] As another embodiment of the present invention, the evaluating the winding effect of the bonding copper wire based on the copper wire winding parameters includes: constructing a winding standard for the bonding copper wire based on the copper wire winding parameters; analyzing the structural deformation degree of the bonding copper wire based on the winding standard; when the structural deformation degree meets the winding standard, evaluating the structural stability of the bonding copper wire to obtain an evaluation result.

[0083] Among them, the winding standard refers to the standard indicator for measuring the winding quality of the bonding copper wire, the degree of structural deformation refers to the degree of abnormal structural change of the bonding copper wire during the winding process, and the structural stability refers to the mechanical properties of the bonding copper wire during the winding process, such as the tensile and bending ability of the copper wire after winding.

[0084] Optionally, the analysis of the degree of structural deformation of the bonding copper wire based on the winding standard can be achieved using a measuring device, such as a microscope. When the degree of structural deformation meets the winding standard, the structural stability of the bonding copper wire is evaluated, and the evaluation result can be obtained through copper wire stretching and copper wire bending tests.

[0085] S3. When the winding effect is greater than the expected effect, a normal trajectory model of the bonding copper wire during the winding process is constructed, and an abnormal detection threshold of the normal trajectory model is set, and a matching coefficient between the copper wire winding parameters and the abnormal detection threshold is calculated. When the matching coefficient is greater than a preset matching coefficient, an abnormal area of ​​the bonding copper wire during the winding process is identified.

[0086] The embodiment of the present invention can effectively evaluate the winding quality of the bonding copper wire by constructing a normal trajectory model of the bonding copper wire during the winding process when the winding effect is greater than the expected effect. The expected effect refers to the ideal winding result of the bonding copper wire, which can be set based on different business scenarios. The normal trajectory model refers to a trajectory model that describes how the copper wire forms an orderly and defect-free trajectory during the winding process.

[0087] As an embodiment of the present invention, the construction of the normal trajectory model of the bonding copper wire during the winding process includes: constructing a simulation model of the winding scene; setting boundary conditions of the simulation model, and based on the boundary conditions, identifying the motion behavior data of the bonding copper wire during the winding process; analyzing the behavior results corresponding to the motion behavior data to obtain the normal trajectory model.

[0088] Among them, the simulation model refers to a geometric model established to simulate the winding of the bonding copper wire, the boundary conditions refer to the constraint conditions for the winding of the bonding copper wire, such as the rotation speed of the winding wheel, the ambient temperature, etc., the motion behavior data refers to the motion state data of the bonding copper wire during the winding process, and the behavior result refers to the deformation result of the copper wire caused by the motion behavior of the bonding copper wire.

[0089] Optionally, when the winding effect is greater than the expected effect, constructing a simulation model of the winding scene can be implemented using CAD, setting the boundary conditions of the simulation model can be obtained through the simulation model, calculating the motion behavior data of the bonding copper wire during the winding process based on the boundary conditions can be implemented by solving the motion equation, and analyzing the behavior results corresponding to the motion behavior data to obtain the normal trajectory model can be obtained through stress-strain analysis.

[0090] Furthermore, the embodiment of the present invention can identify the behavioral deviation of the bonding copper wire during the winding process by setting the abnormal detection threshold of the normal trajectory model. The abnormal detection threshold is used to determine whether the behavior of the bonding copper wire during the winding process deviates from the normal detection standard threshold, which can be obtained by statistical methods.

[0091] Furthermore, the embodiment of the present invention can measure the degree of deviation of the parameter value from the normal range by calculating the matching coefficient between the copper wire winding parameter and the abnormal detection threshold, and the matching coefficient refers to a coefficient measuring the degree of fit between the copper wire winding parameter and the abnormal detection threshold.

[0092] Optionally, the calculation of the matching coefficient between the copper wire winding parameter and the abnormality detection threshold may be achieved using a weighted average method.

[0093] The embodiment of the present invention identifies the abnormal area of ​​the bonding copper wire during the winding process when the matching coefficient is greater than the preset matching coefficient. The winding parameters can be adjusted by analyzing the abnormal area to improve the winding quality. The abnormal area refers to the abnormal winding area of ​​the bonding copper wire. The preset matching coefficient can be based on different business scenario devices, such as 0.8.

[0094] As an embodiment of the present invention, the identifying of the abnormal area of ​​the bonding copper wire during the winding process includes: analyzing the changing trend of the copper wire winding parameters; constructing a winding abnormality time series of the bonding copper wire based on the changing trend, and analyzing the image abnormality time series of the winding image; after comparing the abnormal winding position of the winding abnormality time series with the image abnormality time series, determining the abnormal area of ​​the bonding copper wire during the winding process.

[0095] Among them, the change trend refers to the direction of the evolution of various parameters related to winding over time during the winding process, such as acceleration and tension; the winding abnormality time series refers to the order in which the winding abnormality parameters are arranged according to the time parameters; the image abnormality time series refers to the abnormal time sequence of the winding image abnormalities; the abnormal winding position refers to the bonding copper wire appearing in an unexpected position due to process problems, equipment failures, material defects or operational errors during the winding process.

[0096] Optionally, when the matching coefficient is greater than a preset matching coefficient, analyzing the changing trend of the copper wire winding parameters can be obtained by a simple moving average method, constructing the winding abnormality time series of the bonding copper wire based on the changing trend can be achieved using a state space model, and analyzing the image abnormality time series of the winding image can be obtained by a computer vision algorithm.

[0097] Furthermore, after comparing the abnormal winding positions of the abnormal winding time series and the abnormal image time series, the abnormal area of ​​the bonding copper wire during the winding process is determined, including: synchronizing the abnormal winding time series and the abnormal image time series to obtain the synchronization time; based on the synchronization time, determining the abnormal time point of the bonding copper wire during the winding process; locating the image frame corresponding to the abnormal time point to obtain the abnormal area of ​​the bonding copper wire during the winding process.

[0098] Among them, the synchronization time refers to the same time when two independent time series data sets overlap on the time axis, the abnormal time point refers to a specific time point in the winding process of the bonding copper wire when the winding quality decreases, the equipment fails, or the process parameters exceed the preset range due to some reason, and the corresponding image frame refers to an image screen that can directly display the abnormal state of the bonding copper wire.

[0099] Optionally, the synchronous processing of the winding abnormality time series and the image abnormality time series to obtain the synchronization time can be achieved by utilizing the image acquisition device trigger mechanism; the determination of the abnormal time point of the bonding copper wire in the winding process based on the synchronization time can be obtained through DTW; the positioning of the corresponding image frame of the abnormal time point to obtain the abnormal area of ​​the bonding copper wire in the winding process can be obtained through time series matching.

[0100] S4. Analyze the abnormal factors of the abnormal area, determine the abnormal monitoring report of the bonding copper wire during the winding process based on the abnormal factors, build a visual notification mechanism for the abnormal monitoring report, and output the abnormal monitoring report through the visual notification mechanism.

[0101] The embodiment of the present invention can determine the root cause of the abnormal bonding copper wire winding problem by analyzing the abnormal factor of the abnormal area, wherein the abnormal factor is a variable factor that deviates significantly from a normal winding pattern during the bonding copper wire winding process.

[0102] As an embodiment of the present invention, the analysis of the abnormal factors of the abnormal area includes: checking the abnormal characteristics of the abnormal area; based on the abnormal characteristics, collecting abnormal data of the abnormal area; classifying the abnormal data to obtain classified data; and analyzing the abnormal factors of the abnormal area based on the classified data.

[0103] Among them, the abnormal feature refers to the behavioral feature of the bonding copper wire that deviates from the normal mode during the winding process, the abnormal data refers to the abnormal behavior data corresponding to the abnormal feature, and the classified data refers to classifying the abnormal data into different categories of data.

[0104] Optionally, the checking of abnormal features of the abnormal area may be obtained through result observation, and the collecting of abnormal data of the abnormal area based on the abnormal features may be achieved through data statistics.

[0105] Furthermore, the embodiment of the present invention can report the abnormal situation of the winding of the bonding copper wire in real time by determining the abnormal monitoring report of the bonding copper wire during the winding process based on the abnormal factor, thereby improving the efficiency of solving abnormal problems. The abnormal monitoring report refers to an alarm set for monitoring and detecting the abnormal winding situation of the bonding copper wire.

[0106] As an embodiment of the present invention, the abnormal monitoring report of the bonding copper wire in the winding process is determined based on the abnormal factor, including: setting the abnormal warning threshold of the bonding copper wire in the winding process based on the abnormal factor; setting the maximum width limit value of the abnormal warning threshold, and based on the maximum width limit value, constructing the alarm triggering condition of the bonding copper wire in the winding process, and obtaining the abnormal monitoring report of the bonding copper wire in the winding process.

[0107] Among them, the abnormal warning threshold refers to the alarm threshold set to prevent the bonding copper wire from winding abnormally during the winding process, the maximum tolerance value refers to the bonding copper wire winding abnormality value that the abnormal warning threshold can allow, and the alarm triggering condition refers to the condition for automatically triggering an alarm notification when a winding abnormality is set in the bonding copper wire winding abnormality monitoring system.

[0108] Optionally, the abnormal warning threshold of the bonding copper wire during the winding process is set based on the abnormal factor, which can be achieved by utilizing historical experimental data records, and the maximum margin value of the abnormal warning threshold can be obtained by setting a static margin threshold and a dynamic-static margin threshold.

[0109] Furthermore, the embodiment of the present invention constructs a visual notification mechanism for the abnormal monitoring report, and the visual notification mechanism outputs the abnormal monitoring results of the bonding copper wire in the winding process, that is, the abnormal detection report, which can effectively manage and solve abnormal problems, thereby improving production efficiency and product quality. The visual notification mechanism refers to a notification mechanism that converts the analysis results into charts, dashboards, maps or other graphic forms. The abnormal monitoring results refer to the results of monitoring the abnormal winding of the bonding copper wire, which refers to the identification of event results that deviate significantly from normal behavior through real-time analysis of data or processes.

[0110] As an embodiment of the present invention, the visualization notification mechanism for constructing the abnormal monitoring report and outputting the abnormal monitoring report by the visualization notification mechanism include: setting the visualization type of the abnormal monitoring report and configuring the notification mechanism of the visualization type; and outputting the abnormal monitoring report based on the notification mechanism.

[0111] The visualization type refers to the conversion of monitoring data into a graphic or image format, such as an icon and a display, and the notification mechanism refers to a mechanism for delivering information or reminding a user of a status change.

[0112] Optionally, the setting of the visualization type of the abnormal monitoring report can be obtained through an instrument display, and the configuration of the notification mechanism of the visualization type can be implemented using a sound alarm.

[0113] The embodiment of the present invention obtains the bonding copper wire to be wound, collects the winding image of the bonding copper wire during the winding process, and pre-processes the winding image to obtain a pre-processed image, which can remove the interference elements of the bonding copper wire winding image to make the image clearer. The embodiment of the present invention performs edge detection on the target copper wire in the pre-processed image to obtain an edge detection result, which can provide the key features of the bonding copper wire winding image. The embodiment of the present invention constructs a normal trajectory model of the bonding copper wire during the winding process when the winding effect is greater than the expected effect, which can effectively evaluate the winding quality of the bonding copper wire. The embodiment of the present invention determines the root cause of the abnormal winding of the bonding copper wire by analyzing the abnormal factor of the abnormal area. The abnormal factor is a variable factor that deviates significantly from the normal winding mode during the winding process of the bonding copper wire. Therefore, the method of automatic abnormal monitoring of bonding copper wire winding based on machine vision provided by the embodiment of the present invention can improve the accuracy of automatic abnormal monitoring of bonding copper wire winding.

[0114] like Figure 2 , is a functional module diagram of a system for automatic abnormal monitoring of bonding copper wire winding based on machine vision provided by an embodiment of the present invention.

[0115] The system 100 for automatic abnormal monitoring of bonding copper wire winding based on machine vision of the present invention can be installed in an electronic device. According to the functions implemented, the system 100 for automatic abnormal monitoring of bonding copper wire winding based on machine vision can include an image processing module 101, a winding effect evaluation module 102, an abnormal area identification module 103 and a detection report output module 104. The module described in the present invention can also be called a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0116] In this embodiment, the functions of each module / unit are as follows:

[0117] The image processing module 101 is used to obtain the bonding copper wire to be wound, collect the winding image of the bonding copper wire during the winding process, and pre-process the winding image to obtain a pre-processed image;

[0118] The winding effect evaluation module 102 is used to perform edge detection on the target copper wire in the preprocessed image to obtain an edge detection result, perform background separation processing on the bonding copper wire based on the edge detection result to obtain an independent copper wire image, identify the copper wire winding parameters of the independent copper wire image, and evaluate the winding effect of the bonding copper wire based on the copper wire winding parameters;

[0119] The abnormal area identification module 103 is used to construct a normal trajectory model of the bonding copper wire during the winding process when the winding effect is greater than the expected effect, set an abnormal detection threshold of the normal trajectory model, calculate a matching coefficient between the copper wire winding parameter and the abnormal detection threshold, and identify an abnormal area of ​​the bonding copper wire during the winding process when the matching coefficient is greater than a preset matching coefficient;

[0120] The detection report output module 104 is used to analyze the abnormal factors of the abnormal area, determine the abnormal monitoring report of the bonding copper wire during the winding process based on the abnormal factors, construct a visual notification mechanism for the abnormal monitoring report, and output the abnormal monitoring report by the visual notification mechanism.

[0121] In detail, each module described in the system 100 for automatically monitoring abnormalities of bonding copper wire winding based on machine vision described in the embodiment of the present application is used in the same manner as described above. Figure 1 The same technical means are used as described in the method for automatic abnormal monitoring of bonding copper wire winding based on machine vision, and can produce the same technical effects, so they will not be repeated here.

[0122] In the several embodiments provided by the present invention, it should be understood that the provided methods and systems can be implemented in other ways. For example, the method embodiments described above are only illustrative, for example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.

[0123] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for automatic abnormal monitoring of bonding copper wire winding based on machine vision, characterized in that: The method comprises: Acquiring a bonding copper wire to be wound, collecting a winding image of the bonding copper wire during the winding process, and preprocessing the winding image to obtain a preprocessed image; Performing edge detection on the target copper wire in the preprocessed image to obtain an edge detection result, performing background separation processing on the bonding copper wire based on the edge detection result to obtain an independent copper wire image, identifying copper wire winding parameters of the independent copper wire image, and evaluating the winding effect of the bonding copper wire based on the copper wire winding parameters; When the winding effect is greater than the expected effect, a normal trajectory model of the bonding copper wire during the winding process is constructed, and an abnormality detection threshold of the normal trajectory model is set, and a matching coefficient between the copper wire winding parameter and the abnormality detection threshold is calculated, and when the matching coefficient is greater than a preset matching coefficient, an abnormal area of ​​the bonding copper wire during the winding process is identified; Analyze the abnormal factors of the abnormal area, determine the abnormal monitoring report of the bonding copper wire during the winding process based on the abnormal factors, construct a visual notification mechanism for the abnormal monitoring report, and output the abnormal monitoring report through the visual notification mechanism.

2. The method for automatic abnormality monitoring of bonding copper wire winding based on machine vision according to claim 1, characterized in that: The collecting of the winding image of the bonding copper wire during the winding process includes: Querying an image acquisition device of the bonding copper wire during the winding process, and setting a lighting condition of the image acquisition device; Analyzing a winding cycle of the copper bonding wire, and setting an image capture trigger mechanism of the copper bonding wire during the winding process based on the winding cycle; Under the lighting condition, based on the image capture trigger mechanism, a winding image of the bonding copper wire during the winding process is collected.

3. The method for automatic abnormality monitoring of bonding copper wire winding based on machine vision as claimed in claim 2, characterized in that: The analyzing the winding cycle of the bonding copper wire comprises: querying the material properties of the bonding copper wire, and measuring the winding parameters of the bonding copper wire based on the material properties; Based on the winding parameters, a time series of the winding of the bonding copper wire is constructed to obtain the winding period of the bonding copper wire.

4. The method for automatic abnormality monitoring of bonding copper wire winding based on machine vision according to claim 1, characterized in that: The method of performing background separation processing on the bonding copper wire based on the edge detection result to obtain an independent copper wire image includes: Based on the edge detection result, determining the copper wire contour of the preprocessed image; Performing denoising processing on the copper wire profile to obtain a denoised copper wire profile; The foreground of the denoised copper wire contour is extracted to obtain the independent copper wire image.

5. The method for automatic abnormality monitoring of bonding copper wire winding based on machine vision as claimed in claim 1, characterized in that: The step of identifying the copper wire winding parameters of the independent copper wire image comprises: Acquire a valid line segment of the independent copper wire image, and measure an appearance coefficient of the valid line segment; Based on the appearance coefficient, the winding angle of the copper wire winding is calculated using the following formula: θ=arctan(Δy / Δx) Among them, θ represents the winding angle of the copper wire, Δy represents the height change of the copper wire when it is winding up or down, and Δx represents the straight-line distance that the copper wire moves on the winding axis. The winding density of the copper wire winding is calculated by the following formula: Among them, ρ represents the copper wire winding density, m represents the total mass of the copper wire wound by the copper wire winding, D represents the outer diameter of the copper wire after winding, d represents the inner diameter of the winding drum of the copper wire winding, H represents the height of the copper wire winding, The copper wire winding parameters of the independent copper wire image are generated according to the winding angle and the winding density.

6. The method for automatic abnormality monitoring of bonding copper wire winding based on machine vision according to claim 1, characterized in that: The step of evaluating the winding effect of the bonding copper wire based on the copper wire winding parameters includes: Based on the copper wire winding parameters, constructing a winding standard for the bonding copper wire; Based on the winding standard, analyzing the degree of structural deformation of the bonding copper wire; When the degree of structural deformation meets the winding standard, the structural stability of the bonding copper wire is evaluated to obtain an evaluation result.

7. The method for automatic abnormality monitoring of bonding copper wire winding based on machine vision according to claim 1, characterized in that: The constructing of a normal trajectory model of the bonding copper wire during the winding process includes: Constructing a simulation model of the winding scene; Setting boundary conditions of the simulation model, and based on the boundary conditions, identifying movement behavior data of the bonding copper wire during the winding process; The behavior results corresponding to the motion behavior data are analyzed to obtain the normal trajectory model.

8. The method for automatic abnormality monitoring of bonding copper wire winding based on machine vision as claimed in claim 1, characterized in that: The identifying of an abnormal area of ​​the bonding copper wire during the winding process includes: Analyzing the variation trend of the copper wire winding parameters; Based on the change trend, constructing a winding abnormality time series of the bonding copper wire, and analyzing an image abnormality time series of the winding image; After comparing the abnormal winding positions of the abnormal winding time series and the abnormal image time series, an abnormal area of ​​the bonding copper wire during the winding process is determined.

9. The method for automatic abnormality monitoring of bonding copper wire winding based on machine vision as claimed in claim 1, characterized in that: The analyzing the abnormal factors of the abnormal area includes: Checking abnormal characteristics of the abnormal area; Based on the abnormal features, collecting abnormal data of the abnormal area; classifying the abnormal data to obtain classified data; Based on the classified data, an abnormal factor of the abnormal area is analyzed.

10. The method for automatic abnormality monitoring of bonding copper wire winding based on machine vision according to claim 1, characterized in that: The step of determining an abnormal monitoring report of the bonding copper wire during the winding process based on the abnormal factor includes: Setting the visualization type of the abnormal monitoring report and configuring the notification mechanism of the visualization type; Based on the notification mechanism, the abnormal monitoring report is output.

Citation Information

Cited By

  • Abnormal winding detection method for protective sleeve based on machine vision

    CN119648710A