Chip lead bonding positioning method and system
Through image acquisition and segmentation algorithms, the contours of the chip and lead frame are detected, and the solder joint coordinates are corrected, which solves the accuracy and quality problems caused by position deviation during chip packaging, and achieves high-precision bonding.
Patent Information
- Application Number
- CN202410983101.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-22
- Publication Date
- 2024-11-08
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
During the chip packaging process, due to the movement of the bonding console, there is a deviation between the actual position of the chip and the lead frame relative to the target position, resulting in low product packaging accuracy and quality problems, and this deviation cannot be solved through prejudgment.
The bonded image is segmented through image acquisition and segmentation algorithm, the outline of the chip and lead frame is detected, the deviation is calculated and the solder joint coordinates of the chip and lead frame are accurately determined to achieve accurate bonding.
It improves the accuracy and quality of the chip package, prevents dummy or breaks caused by solder joint deviation, and achieves high-precision bonding.
Smart Images

Figure CN118918315A_ABST
Abstract
Description
Technical Field
[0001] The present invention mainly relates to the technical field of chip packaging, and in particular to a chip wire bonding positioning method and system. Background Art
[0002] Wire bonding is the process of connecting the chip and the lead frame through wires. It can ensure the connection between the chip and the external circuit to achieve information communication, and is a key step in the chip packaging process. As chip packaging technology continues to develop towards high integration, high performance, multiple leads and fine pitch, wire bonding technology has become the most commonly used packaging method, and the bonding accuracy requirements are becoming increasingly stringent. In the wire bonding process, each step involves precise positioning issues. Therefore, wire bonding positioning has become one of the key factors affecting chip packaging accuracy.
[0003] During the chip packaging process, due to the continuous movement of the bonding console, there will be a certain deviation between the actual position of the chip and the lead frame relative to the target position. If bonding is performed according to the deviated position, it will lead to low product packaging accuracy, and more seriously affect product quality. Moreover, this deviation cannot be solved by pre-judgment. Therefore, it is urgent to accurately position the bonding point of the chip during the bonding process to improve the accuracy of chip packaging. Summary of the invention
[0004] The technical problem to be solved by the present invention is to provide a chip wire bonding positioning method and system in view of the deficiencies in the prior art.
[0005] The technical solution of the present invention to solve the above technical problems is as follows:
[0006] A chip wire bonding positioning method comprises the following steps:
[0007] The image of the chip to be bonded is captured by a preset camera to obtain an image to be bonded;
[0008] Segmenting the image to be bonded by an image segmentation algorithm to obtain a chip image including a plurality of chip solder joints and a lead frame image including a plurality of pins;
[0009] Performing contour detection on the chip image to obtain a chip contour, and calculating a deviation between the chip contour and a pre-imported template contour to obtain a compensation difference, and correcting the coordinates of each chip solder joint in the template image corresponding to the template contour according to the compensation difference to obtain a plurality of actual chip solder joint coordinates;
[0010] Performing contour detection on the lead frame image to obtain a plurality of pin contours, and calculating actual pin solder joint coordinates corresponding to the plurality of pin contours based on each pin contour;
[0011] The actual chip solder joint coordinates are connected to the corresponding actual pin solder joint coordinates respectively to complete the bonding of the chip and the lead frame.
[0012] Another technical solution of the present invention to solve the above technical problems is as follows:
[0013] A chip wire bonding positioning system, comprising:
[0014] An image acquisition module is used to acquire an image of the chip to be bonded by using a preset camera to obtain an image to be bonded;
[0015] An image recognition module is used to segment the image to be bonded by an image segmentation algorithm to obtain a chip image containing a plurality of chip solder joints and a lead frame image containing a plurality of pins;
[0016] A chip positioning module is used to perform contour detection on the chip image to obtain a chip contour, calculate the deviation between the chip contour and a pre-imported template contour, obtain a compensation difference, and correct the coordinates of each chip solder joint in the template image corresponding to the template contour according to the compensation difference to obtain multiple actual chip solder joint coordinates;
[0017] A lead frame positioning module, used to perform contour detection on the lead frame image to obtain a plurality of pin contours, and calculate the actual pin solder joint coordinates corresponding to the plurality of pin contours based on each pin contour;
[0018] The wire bonding module is used to respectively connect the coordinates of the plurality of actual chip solder joints with the corresponding coordinates of the actual pin solder joints to complete the bonding between the chip and the lead frame.
[0019] The beneficial effects of the present invention are as follows: in the chip bonding packaging process, the image to be bonded is segmented by an image segmentation algorithm, thereby improving the image recognition rate to identify the image for solder joint positioning. The image where the solder joint is located is detected with high precision by an edge detection algorithm to correct the position deviation of the chip solder joint and the pin solder joint in the lead frame, thereby preventing the solder joint from being too biased and not firm, thereby improving the bonding accuracy and quality, and achieving precise bonding. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] Figure 1 A flow chart of a chip wire bonding positioning method provided by an embodiment of the present invention;
[0021] Figure 2 A module block diagram of a chip wire bonding positioning system provided in an embodiment of the present invention. DETAILED DESCRIPTION
[0022] The principles and features of the present invention are described below in conjunction with the accompanying drawings. The examples given are only used to explain the present invention and are not used to limit the scope of the present invention.
[0023] like Figure 1 As shown, a chip wire bonding positioning method provided by an embodiment of the present invention includes the following steps:
[0024] The image of the chip to be bonded is captured by a preset camera to obtain an image to be bonded;
[0025] Segmenting the image to be bonded by an image segmentation algorithm to obtain a chip image including a plurality of chip solder joints and a lead frame image including a plurality of pins;
[0026] Performing contour detection on the chip image to obtain a chip contour, and calculating a deviation between the chip contour and a pre-imported template contour to obtain a compensation difference, and correcting the coordinates of each chip solder joint in the template image corresponding to the template contour according to the compensation difference to obtain a plurality of actual chip solder joint coordinates;
[0027] Performing contour detection on the lead frame image to obtain a plurality of pin contours, and calculating actual pin solder joint coordinates corresponding to the plurality of pin contours based on each pin contour;
[0028] The actual chip solder joint coordinates are connected to the corresponding actual pin solder joint coordinates respectively to complete the bonding of the chip and the lead frame.
[0029] Specifically, the segmentation of the image to be bonded is performed based on edge features through the image segmentation algorithm. First, the edge of the target in the image to be bonded (such as points, lines, target contours, etc.) is detected using the edge detection algorithm, and then the spatial relationship of the pixel points is used to connect the edges into a closed contour of the detection target according to the set conditions, and the chip image and the lead frame image are segmented.
[0030] It should be understood that in the chip design stage, the location of the bonding area needs to be clearly marked in the circuit diagram or layout. These areas usually include the chip's solder joints (i.e., the chip solder joints in the template image) and the corresponding solder joints on the lead frame, and the bonding area (i.e., the template outline) is marked with specific marks to ensure that each solder joint in the chip and the corresponding solder joint on the lead frame have a unique identifier, and finally generate a design file with the lead interconnection position for matching and identification during wire bonding. The design file is imported into the control unit of the wire bonding machine. When the chip is wire bonded, the control unit controls the motion mechanism to wire bond the solder joints in the chip with the corresponding solder joints on the lead frame according to the design file. At this time, the chip wire bonding positioning method of this scheme is used to accurately position the chip's image to be bonded, and the obtained multiple actual chip solder joint coordinates and the actual pin solder joint coordinates of the corresponding pins are sent to the control unit to control the motion mechanism to accurately bond the chip.
[0031] In an embodiment of the present invention, an industrial camera is used to collect images of the chip to be bonded, and the current positions of the chip and the lead frame are identified. A compensation value is calculated based on the current position of the chip and the template position to correct the coordinates of the solder joints to be bonded in the chip. The coordinates of the solder joints in each pin are calculated based on the current position of the lead frame, that is, the optimal position for wire bonding of each pin of the lead frame is calculated, which not only saves bonding materials, but also prevents the solder joints from being too biased and not firm, causing cold solder joints or fractures. It is able to dynamically compensate for mechanical structure deviations and motion errors caused by the bonding process, thereby improving the accuracy and quality of bonding.
[0032] Preferably, the performing contour detection on the chip image to obtain the chip contour is specifically:
[0033] The chip image is converted into a chip grayscale image by an edge detection algorithm, and edge detection is performed on the chip grayscale image to obtain a chip edge image, and Hough transformation is performed on the chip edge image to obtain a chip contour.
[0034] Specifically, the Canny edge detection algorithm is used to detect the edge of the chip image, and pixels with a pixel value of 255 in the chip image are considered to be edges, and multiple chip pixel points are obtained.
[0035] In the embodiment of the present invention, the Canny edge detection algorithm is used to remove non-edge pixels from the grayscale image and only retain possible edge pixels, which can reduce the number of pixels that need to be processed by the Hough transform, thereby reducing the amount of calculation.
[0036] Preferably, the chip edge image is composed of a plurality of chip pixels;
[0037] The chip edge image is subjected to Hough transformation to obtain the chip outline, specifically:
[0038] Map multiple chip pixel points into a polar coordinate system to obtain multiple straight line parameters, create corresponding accumulators according to the multiple straight line parameters, calculate the number of chip pixel points that meet the corresponding straight line parameters through the multiple accumulators, convert the straight line parameters whose chip pixel numbers exceed a preset threshold into straight line equations to obtain multiple chip edge straight line equations, respectively combine the multiple chip edge straight line equations for solution to obtain multiple chip vertex coordinates, and obtain a chip outline by connecting the vertices corresponding to the multiple chip vertex coordinates one by one.
[0039] Specifically, a parameter space is created, and the parameter space (ie, polar coordinate system) is a two-dimensional matrix of ρ and θ, where ρ is the polar diameter and θ is the polar angle, ranging from 0° to 180°.
[0040] The chip edge image is mapped pixel by pixel, and multiple chip pixels are mapped into the polar coordinate system. The corresponding first straight line parameters are generated according to the first mapped chip pixels, and an accumulator is created for the first straight line parameters. The second straight line parameters corresponding to the chip pixels mapped later are compared with the first straight line parameters to determine whether the θ values of the chip pixels are the same. If so, the accumulated value of the accumulator is increased. If not, an accumulator is created for the second straight line parameters, and so on, until all chip pixels are mapped.
[0041] The straight line parameters whose accumulated values of the accumulators in the identification parameter space exceed the preset threshold value are used, and the chip pixel points in the accumulator corresponding to the straight line parameters constitute edge straight lines. Therefore, the corresponding chip edge straight line equations are calculated according to the straight line parameters, and multiple chip edge straight line equations are combined in pairs to solve four intersection points (i.e., four vertices), and the four vertices are connected to obtain the chip contour.
[0042] It should be understood that a chip is generally rectangular, so four straight line equations are obtained according to the Hough transform, and four vertices are obtained by solving them in pairs.
[0043] In the embodiment of the present invention, the Hough transform can extract shape parameters from the image, and express the straight line in a parameterized form, and the corresponding straight line equation can be directly obtained according to the straight line parameters.
[0044] Preferably, the calculation of the deviation between the chip contour and the pre-imported template contour to obtain the compensation difference is specifically as follows:
[0045] Calculating the actual center point coordinates of the chip outline according to the plurality of chip vertex coordinates, performing difference calculation between the actual center point coordinates and the template center point coordinates of the pre-imported template outline to obtain a coordinate difference;
[0046] The actual tilt angle of the chip contour is calculated according to the line parameters of any chip edge line equation, and the difference between the actual tilt angle and the template tilt angle of the pre-imported template contour is calculated to obtain the angle difference.
[0047] It should be understood that the pre-imported template image includes the template outline of the chip and the coordinates of each solder joint on the chip. The chip is rectangular, so the actual center point coordinates of the chip outline can be calculated according to the center point coordinate formula of the matrix; the inclination angle of the straight line can be calculated according to the polar coordinate angle of the straight line parameter.
[0048] In the embodiment of the present invention, the coordinate compensation amount of each solder point is calculated according to the vertices of the chip outline, and the angle compensation amount of each solder point is calculated according to the inclination angle of any side of the chip outline.
[0049] Preferably, the coordinates of each chip solder joint in the template image corresponding to the template contour are corrected according to the compensation difference to obtain a plurality of actual chip solder joint coordinates, specifically:
[0050] The coordinate difference, the angle difference and the coordinates of each chip solder joint in the template image corresponding to the template contour are calculated by a coordinate transformation expression to obtain a plurality of actual chip solder joint coordinates. The coordinate transformation expression is:
[0051]
[0052] Among them, x ′ y ′ are the horizontal and vertical coordinates of the actual chip solder joint, xy are the horizontal and vertical coordinates of the template chip solder joint, θ is the angle difference, Δx is the horizontal coordinate difference, and Δy is the vertical coordinate difference.
[0053] It should be understood that the coordinate transformation expression is applied to a conventional coordinate system in a two-dimensional plane with the positive direction of the x-axis to the right and the positive direction of the y-axis upward.
[0054] In the embodiment of the present invention, the coordinates of each solder joint on the chip are corrected according to the compensation difference to perform precise bonding.
[0055] Preferably, the contour detection is performed on the lead frame image to obtain a plurality of pin contours, specifically:
[0056] The lead frame image is converted into a lead frame grayscale image by an edge detection algorithm, and edge detection is performed on the lead frame grayscale image to obtain a plurality of pin edge images respectively composed of a plurality of pin pixel points, and a plurality of pin pixel points of each of the pin edge images are respectively fitted by a least square method to obtain a plurality of pin edge straight line equations corresponding to each of the pin edge images;
[0057] The multiple pin edge straight line equations corresponding to each of the pin edge images are solved separately to obtain multiple pin vertex coordinates, and the vertices corresponding to the multiple pin vertex coordinates corresponding to each of the pin edge images are connected one by one to obtain the pin contours corresponding to the multiple pin edge images.
[0058] It should be understood that the edge detection algorithm may be a Canny edge detection algorithm, a Sobel edge detection algorithm or a Prewitt edge detection algorithm to identify the pixel coordinates of edge feature points in the lead frame image.
[0059] In the embodiment of the present invention, a large amount of nonlinear data is fitted by the least square method, each pixel point is expressed by a linear equation, and the straight line equation of the edge of each pin in the lead frame is obtained to obtain the contour of each pin.
[0060] Preferably, the plurality of pin edge straight line equations of the pin edge image include pin edge straight line equations corresponding to two opposite long sides;
[0061] The actual pin soldering point coordinates corresponding to the plurality of pin contours are calculated based on the respective pin contours, specifically:
[0062] Solve the pin edge straight line equations corresponding to two opposite long sides of any one of the pin edge images to obtain the center coordinates of all inscribed circles between the two long sides, calculate the distances between the center coordinates of all inscribed circles and the corresponding actual chip solder joint coordinates, obtain multiple solder joint distances, screen the multiple solder joint distances according to a preset threshold range, obtain the optimal solder joint distance, the center coordinates of the inscribed circle corresponding to the optimal solder joint distance are the actual pin solder joint coordinates in the pin edge image, and so on, until the actual pin solder joint coordinates corresponding to all pin contours are calculated.
[0063] It should be understood that the pin outline is generally a quadrilateral, and two opposite long sides are selected according to the pin outline so that the coordinates of the center of the inscribed circle between the two long sides can be solved according to the straight line equations corresponding to the two long sides.
[0064] In the embodiment of the present invention, the coordinates of the center of the inscribed circle in the pin outline are obtained and set as the coordinates of the pin solder joint, so that the solder joint is within the pin surface and in the center position, preventing the connection at the edge of the pin during bonding and causing cold solder joint or breakage.
[0065] like Figure 2 As shown, a chip wire bonding positioning system provided by an embodiment of the present invention includes:
[0066] An image acquisition module is used to acquire an image of the chip to be bonded by using a preset camera to obtain an image to be bonded;
[0067] An image recognition module is used to segment the image to be bonded by an image segmentation algorithm to obtain a chip image containing a plurality of chip solder joints and a lead frame image containing a plurality of pins;
[0068] A chip positioning module is used to perform contour detection on the chip image to obtain a chip contour, calculate the deviation between the chip contour and a pre-imported template contour, obtain a compensation difference, and correct the coordinates of each chip solder joint in the template image corresponding to the template contour according to the compensation difference to obtain multiple actual chip solder joint coordinates;
[0069] A lead frame positioning module, used to perform contour detection on the lead frame image to obtain a plurality of pin contours, and calculate the actual pin solder joint coordinates corresponding to the plurality of pin contours based on each pin contour;
[0070] The wire bonding module is used to respectively connect the coordinates of the plurality of actual chip solder joints with the corresponding coordinates of the actual pin solder joints to complete the bonding between the chip and the lead frame.
[0071] The above-mentioned chip wire bonding positioning system can refer to the implementation content and beneficial effects of the chip wire bonding positioning method specifically described above, which will not be repeated here.
[0072] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.
[0073] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims
1. A chip wire bonding positioning method, characterized in that: The steps include: The image of the chip to be bonded is captured by a preset camera to obtain an image to be bonded; Segmenting the image to be bonded by an image segmentation algorithm to obtain a chip image including a plurality of chip solder joints and a lead frame image including a plurality of pins; Performing contour detection on the chip image to obtain a chip contour, and calculating a deviation between the chip contour and a pre-imported template contour to obtain a compensation difference, and correcting the coordinates of each chip solder joint in the template image corresponding to the template contour according to the compensation difference to obtain a plurality of actual chip solder joint coordinates; Performing contour detection on the lead frame image to obtain a plurality of pin contours, and calculating actual pin solder joint coordinates corresponding to the plurality of pin contours based on each pin contour; The actual chip solder joint coordinates are connected to the corresponding actual pin solder joint coordinates respectively to complete the bonding of the chip and the lead frame.
2. The chip wire bonding positioning method according to claim 1, characterized in that: The chip image is subjected to contour detection to obtain the chip contour, specifically: The chip image is converted into a chip grayscale image by an edge detection algorithm, and edge detection is performed on the chip grayscale image to obtain a chip edge image, and Hough transformation is performed on the chip edge image to obtain a chip contour.
3. The chip wire bonding positioning method according to claim 2, characterized in that: The chip edge image is composed of a plurality of chip pixels; The chip edge image is subjected to Hough transformation to obtain the chip outline, specifically: Map multiple chip pixel points into a polar coordinate system to obtain multiple straight line parameters, create corresponding accumulators according to the multiple straight line parameters, calculate the number of chip pixel points that meet the corresponding straight line parameters through the multiple accumulators, convert the straight line parameters whose chip pixel numbers exceed a preset threshold into straight line equations to obtain multiple chip edge straight line equations, respectively combine the multiple chip edge straight line equations for solution to obtain multiple chip vertex coordinates, and obtain a chip outline by connecting the vertices corresponding to the multiple chip vertex coordinates one by one.
4. The chip wire bonding positioning method according to claim 3, characterized in that: The calculation of the deviation between the chip contour and the pre-imported template contour to obtain the compensation difference is specifically as follows: Calculating the actual center point coordinates of the chip outline according to the plurality of chip vertex coordinates, performing difference calculation between the actual center point coordinates and the template center point coordinates of the pre-imported template outline to obtain a coordinate difference; The actual tilt angle of the chip contour is calculated according to the line parameters of any chip edge line equation, and the difference between the actual tilt angle and the template tilt angle of the pre-imported template contour is calculated to obtain the angle difference.
5. The chip wire bonding positioning method according to claim 4, characterized in that: The coordinates of each chip solder joint in the template image corresponding to the template contour are corrected according to the compensation difference to obtain multiple actual chip solder joint coordinates, specifically: The coordinate difference, the angle difference and the coordinates of each chip solder joint in the template image corresponding to the template contour are calculated by a coordinate transformation expression to obtain a plurality of actual chip solder joint coordinates. The coordinate transformation expression is: Among them, x ′ y ′ are the horizontal and vertical coordinates of the actual chip solder joint, xy are the horizontal and vertical coordinates of the template chip solder joint, θ is the angle difference, Δx is the horizontal coordinate difference, and Δy is the vertical coordinate difference.
6. The chip wire bonding positioning method according to claim 1, characterized in that: The contour detection is performed on the lead frame image to obtain multiple pin contours, specifically: The lead frame image is converted into a lead frame grayscale image by an edge detection algorithm, and edge detection is performed on the lead frame grayscale image to obtain a plurality of pin edge images respectively composed of a plurality of pin pixel points, and a plurality of pin pixel points of each of the pin edge images are respectively fitted by a least square method to obtain a plurality of pin edge straight line equations corresponding to each of the pin edge images; The multiple pin edge straight line equations corresponding to each of the pin edge images are solved separately to obtain multiple pin vertex coordinates, and the vertices corresponding to the multiple pin vertex coordinates corresponding to each of the pin edge images are connected one by one to obtain the pin contours corresponding to the multiple pin edge images.
7. The chip wire bonding positioning method according to claim 6, characterized in that: The plurality of pin edge straight line equations of the pin edge image include pin edge straight line equations corresponding to two opposite long sides; The actual pin soldering point coordinates corresponding to the plurality of pin contours are calculated based on the respective pin contours, specifically: Solve the pin edge straight line equations corresponding to two opposite long sides of any one of the pin edge images to obtain the center coordinates of all inscribed circles between the two long sides, calculate the distances between the center coordinates of all inscribed circles and the corresponding actual chip solder joint coordinates, obtain multiple solder joint distances, screen the multiple solder joint distances according to a preset threshold range, obtain the optimal solder joint distance, the center coordinates of the inscribed circle corresponding to the optimal solder joint distance are the actual pin solder joint coordinates in the pin edge image, and so on, until the actual pin solder joint coordinates corresponding to all pin contours are calculated.
8. A chip wire bonding positioning system, characterized in that: include: An image acquisition module is used to acquire an image of the chip to be bonded by using a preset camera to obtain an image to be bonded; An image recognition module is used to segment the image to be bonded by an image segmentation algorithm to obtain a chip image containing a plurality of chip solder joints and a lead frame image containing a plurality of pins; A chip positioning module is used to perform contour detection on the chip image to obtain a chip contour, calculate the deviation between the chip contour and a pre-imported template contour, obtain a compensation difference, and correct the coordinates of each chip solder joint in the template image corresponding to the template contour according to the compensation difference to obtain multiple actual chip solder joint coordinates; A lead frame positioning module, used to perform contour detection on the lead frame image to obtain a plurality of pin contours, and calculate the actual pin solder joint coordinates corresponding to the plurality of pin contours based on each pin contour; The wire bonding module is used to respectively connect the coordinates of the plurality of actual chip solder joints with the corresponding coordinates of the actual pin solder joints to complete the bonding between the chip and the lead frame.
9. The chip wire bonding positioning system according to claim 8, characterized in that: In the chip positioning module, the chip image is subjected to contour detection to obtain the chip contour, specifically: The chip image is converted into a chip grayscale image by an edge detection algorithm, and edge detection is performed on the chip grayscale image to obtain a chip edge image, and Hough transformation is performed on the chip edge image to obtain a chip contour.
10. The chip wire bonding positioning system according to claim 9, characterized in that: In the chip positioning module, the chip edge image is composed of a plurality of chip pixels; The chip edge image is subjected to Hough transformation to obtain the chip outline, specifically: Map multiple chip pixel points into a polar coordinate system to obtain multiple straight line parameters, create corresponding accumulators according to the multiple straight line parameters, calculate the number of chip pixel points that meet the corresponding straight line parameters through the multiple accumulators, convert the straight line parameters whose chip pixel numbers exceed a preset threshold into straight line equations to obtain multiple chip edge straight line equations, respectively combine the multiple chip edge straight line equations for solution to obtain multiple chip vertex coordinates, and obtain a chip outline by connecting the vertices corresponding to the multiple chip vertex coordinates one by one.
Citation Information
Cited By
Electronic component pin detection method and device and computer storage medium
CN121304772A
Dynamic compensation method and device for displacement error of motion platform of high-speed wire bonding equipment
CN121772803A
High-speed wire bonding equipment motion platform displacement error dynamic compensation method and device
CN121772803B
Bonding quality real-time detection system and method
CN121925099A
