Efficient Mark point identification method and system for circuit board test

Abstract

The invention discloses an efficient Mark point recognition method and system for circuit board testing, and the system comprises a high-precision carrier, a visual collection module, a data processing module, and a motion control module. The visual acquisition module only identifies a first Mark point in the jointed board and a second Mark point which is diagonally distributed with the first Mark point, and calculates position information of all units to be detected on the jointed board in combination with pre-stored jointed board layout parameters, so that only a single Mark point and the diagonally distributed Mark points of the whole board need to be identified by remarkably reducing the use number of Mark identification points, and the detection efficiency of the jointed board is improved. Therefore, the identification process is effectively simplified. And the time consumed by the equipment in the Mark identification link is greatly reduced, so that the overall production efficiency is remarkably improved. And the production cycle is further shortened, and the method is particularly suitable for application scenes of high-density jointed boards.

Description

Technical Field

[0001] This invention relates to the field of circuit board processing technology, specifically to an efficient Mark point identification method and system for circuit board testing. Background Technology

[0002] In the automated production and testing of circuit boards (such as PCBs) (including flying probe testing, surface mount technology, optical inspection, etc.), a vision system is typically needed to identify reference points (mark points) on the board surface for precise positioning. The current mainstream method is as follows: For a multi-panel array composed of multiple identical panels, the vision system needs to photograph and locate each of the two diagonal mark points (e.g., Mark1 and Mark2) of each panel. The specific process is as follows: the camera moves to above Mark1 of the first panel, takes a picture, and calculates its coordinates; then it moves to above Mark2 of the same panel, takes a picture, and calculates its coordinates; then it moves sequentially to the second panel, repeating this process until all N panels have been traversed. After obtaining the coordinates of all 2N mark points, the control unit calculates the offset and rotation angle of each panel based on the coordinates of each pair of mark points, thereby deriving the precise position of all test points or components on that panel.

[0003] While the above methods can guarantee positioning accuracy, their efficiency bottlenecks are very prominent, especially in large-scale panel applications (e.g., N>9), where the drawbacks are particularly significant: 1. Redundant recognition path: The camera needs to move long distances back and forth between Mark points in different units, with overlapping motion trajectories and significant wasted space. 2. High image processing load: Image acquisition and processing of 2N Mark points are required, and the computational load increases linearly with the number of units, resulting in significant time consumption. 3. Long mechanical waiting time: Each recognition point involves the precise stopping, focusing, and taking pictures of the camera, and the accumulated mechanical delay becomes a major obstacle to improving overall efficiency. These shortcomings limit the equipment's throughput and cannot meet the demands of modern electronic manufacturing for high efficiency and low cost.

[0004] Therefore, the present invention provides an efficient Mark point identification method and system for circuit board testing. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides an efficient Mark point identification method and system for circuit board testing, aiming to solve the aforementioned problems.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a high-efficiency Mark point recognition system for circuit board testing, comprising:

[0007] A high-precision carrier is used to fix high-density circuit board panels and ensures that the relative positional accuracy between each unit in the panel is within ±0.1mm through a positioning mechanism;

[0008] The visual acquisition module, including a camera, a light source, and an image processing module, is used to acquire image data of the circuit board panel.

[0009] The data processing module is communicatively connected to the vision acquisition module and is used to process the image data to identify Mark points, calculate coordinates, and generate control commands.

[0010] A motion control module, connected to the data processing module, is used to drive the test equipment to perform circuit board testing according to the control commands;

[0011] The data processing module is configured to: control the visual acquisition module to identify only a first Mark point and a second Mark point that are diagonally distributed with the first Mark point in a selected unit of the panel, and calculate the position information of all units to be tested on the panel based on the coordinates (x1, y1) of the first Mark point and the coordinates (x2, y2) of the second Mark point, combined with the pre-stored panel layout parameters.

[0012] Preferably, the panel layout parameters include the number of units M in the X direction and the number of units N in the Y direction; the data processing module calculates the reference position coordinates (X, Y, and Y) of the unit in the i-th row and j-th column using linear interpolation. ij ,Y ij The calculation formula is:

[0013]

[0014] Where i = 1, 2, ..., N; j = 1, 2, ..., M, the reference position coordinates are used to represent the positioning origin of the corresponding unit.

[0015] Preferably, the data processing module is further configured with a precision verification unit. This unit randomly selects at least one verification unit from other units besides the selected unit, and controls the visual acquisition module to identify a Mark point of the verification unit to obtain its actual coordinates (x, y). a ,y a ), calculate the actual coordinates and the reference position coordinates (X). a ,Y a The Euclidean distance deviation δ between them:

[0016]

[0017] If δ is greater than the preset error threshold Δ, the value of Δ is determined based on the panel positioning accuracy and the diameter of the test point. Usually, Δ≥0.1mm, then the current recognition is deemed invalid and the re-recognition process is started.

[0018] Preferably, the visual acquisition module includes an image processing unit, which performs the following operations sequentially on the acquired Mark point images:

[0019] a) Image preprocessing: Gaussian filtering function is used. The image is smoothed and denoised, where σ is the filtering scale parameter;

[0020] b) Feature extraction: The Canny operator is used, with the formula: gradient magnitude G = √(Gx^2 + Gy^2), direction θ = arctan(Gy / Gx), where Gx and Gy are the gradients of the image in the x and y directions. After Gaussian filtering and non-maximum suppression, the edges of the Mark points are extracted, and the center coordinates are finally located.

[0021] c) Coordinate positioning: Calculate the centroid coordinates of the extracted contour or circle center, and output them as the final coordinates (x, y) of the Mark point.

[0022] Preferably, the control command generated by the data processing module is to generate a test path sequence based on the calculated test point coordinates, so that the working end of the test equipment moves sequentially according to the test point coordinates, and the movement trajectory is optimized based on a greedy algorithm to minimize the total movement distance.

[0023] Preferably, the camera is a CMOS sensor with a resolution higher than 5 million pixels and a frame rate of 30fps, used to capture images of the circuit board panel; the light source is an adjustable brightness ring LED with a wavelength range of 400-700nm to ensure uniform illumination; the image processing module embeds a DSP chip and runs an image recognition algorithm to extract the coordinates of the Mark points; the data processing module PLC includes a processor, a memory, and a communication interface, the memory stores the panel parameters and calculation program, the processor executes a geometric transformation algorithm to calculate the coordinates of the test points, and controls the test equipment through the communication interface.

[0024] Preferably, the system is integrated into a circuit board processing device, which is one of a flying probe tester, a chip mounter, an automatic optical inspection device, and an automatic visual inspection device.

[0025] The motion control module is communicatively connected to the main controller of the circuit board processing equipment to drive the actuator of the circuit board processing equipment to perform operations, including:

[0026] When the device is a flying probe tester, the test probe is driven to move to the test point corresponding to the reference position to perform electrical testing;

[0027] When the device is a pick and place machine, the placement head is driven to move to the placement point corresponding to the reference position to place the component;

[0028] When the device is an automatic optical inspection or automatic visual inspection device, the inspection camera is driven to move to the inspection point corresponding to the reference position to perform image acquisition and analysis.

[0029] An efficient Mark point identification method for circuit board testing, applied to the above system, the method includes the following steps:

[0030] Step S1: Panel Fixing and Coordinate System Establishment. The circuit board panel is fixed on a high-precision carrier using positioning pins and high-precision clamps, and the base plate coordinate system O is established. base -XY;

[0031] Step S2: Baseline coordinate acquisition. Control the movement of the vision acquisition module to identify and acquire only the coordinates (x1, y1) of a first Mark point and the coordinates (x2, y2) of a second Mark point that is diagonally distributed with the first Mark point in the puzzle.

[0032] Step S3: Test point coordinate calculation. The data processing module calculates the test point coordinates of all units in the panel based on the coordinates of the first Mark point (x1, y1) and the second Mark point (x2, y2), combined with the number of rows M and columns N of the panel.

[0033] Step S4: Circuit board testing. The motion control module controls the testing equipment to perform circuit board testing according to the calculated test point coordinates.

[0034] Preferably, the image processing module identifies the first Mark point using a template matching algorithm. The template matching algorithm performs a convolution operation between a pre-stored Mark point template image and the captured image, with the formula: similarity score S(u,v)=ΣΣ[I(x,y)*T(xu,yv)], where I(x,y) is the pixel value of the captured image, T(x,y) is the pixel value of the template image, and (u,v) is the displacement. When S(u,v) exceeds the threshold, it is determined that the recognition is successful, and the center point coordinates are extracted.

[0035] Preferably, in step S2, the vision system controls the camera to move along the optimized path to the diagonal position. The optimized path is calculated based on the shortest path algorithm, and the path length L = √[(x2-x1)^2+(y2-y1)^2]. The moving speed is adjusted according to the camera's acceleration and deceleration to reduce the moving time.

[0036] Beneficial effects

[0037] Compared with the prior art, the present invention has the following advantages:

[0038] This invention significantly reduces the number of Mark recognition points required, requiring only the recognition of a single Mark point and Mark points diagonally across the board, thus effectively simplifying the recognition process and breaking through the traditional dual-Mark point recognition mode. This drastically reduces the time spent by the equipment in the Mark recognition stage, thereby significantly improving overall production efficiency. Simultaneously, this solution fully utilizes the control of panel precision and test point size to ensure that testing accuracy is not affected, further shortening the production cycle, making it particularly suitable for high-density panel applications. The entire system has a simple structural design, making it easy to integrate into existing equipment and effectively reducing the cost of modification and upgrades. Attached Figure Description

[0039] Figure 1 This is a block diagram of the high-efficiency Mark point recognition system of the present invention;

[0040] Figure 2 This is a flowchart of the efficient Mark point recognition method of the present invention;

[0041] Figure 3 This is a comparison diagram between the method of this invention and existing methods. Detailed Implementation

[0042] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0043] Please see Figure 1-3 A high-efficiency Mark point recognition system for circuit board testing, comprising:

[0044] A high-precision carrier is used to fix high-density circuit board panels and ensures that the relative positional accuracy between the units in the panel is within ±0.1mm through a positioning mechanism.

[0045] The visual acquisition module, including a camera, a light source, and an image processing module, is used to acquire image data of the circuit board panel.

[0046] The data processing module is communicatively connected to the vision acquisition module and is used to process the image data to identify Mark points, calculate coordinates, and generate control commands.

[0047] A motion control module, connected to the data processing module, is used to drive the test equipment to perform circuit board testing according to the control commands;

[0048] The data processing module is configured to: control the visual acquisition module to identify only a first Mark point and a second Mark point that are diagonally distributed with the first Mark point in a selected unit of the panel, and calculate the position information of all units to be tested on the panel based on the coordinates (x1, y1) of the first Mark point and the coordinates (x2, y2) of the second Mark point, combined with the pre-stored panel layout parameters.

[0049] The high-precision carrier first precisely fixes the high-density panel of circuit boards, using a positioning mechanism to ensure that the relative positional accuracy between the units in the panel is within an extremely small error range. The camera and light source of the vision acquisition module then work together. Under stable lighting conditions provided by the light source, the camera acquires image data of the circuit board panel, which is then transmitted to the image processing module for preliminary processing. Upon receiving the processed image data, the data processing module, according to its preset program, controls the vision acquisition module to focus its attention on the selected unit in the panel, accurately identifying the first Mark point and the second Mark point diagonally distributed to it. Next, based on the obtained coordinates of the first and second Mark points, and combined with pre-stored panel layout parameters, a complex and precise mathematical algorithm is used to calculate the positional information of all units to be tested on the panel. After calculating the positional information, the data processing module quickly generates control commands. Upon receiving these commands, the motion control module precisely drives the testing equipment, ensuring that the testing equipment accurately reaches the position of each unit to be tested, enabling comprehensive and efficient testing of the circuit board. During the testing process, all modules maintained close collaboration, continuously processing and analyzing the data to ensure that each unit under test could be accurately detected, greatly improving the efficiency and accuracy of circuit board testing.

[0050] Figure 3 The dashed arrows represent the recognition paths of existing mark recognition technologies, while the solid arrows represent the recognition paths of the efficient mark recognition technology proposed in this invention. The proposed solution significantly reduces the number of mark recognition points required, needing only to recognize a single mark point and the mark points diagonally across the entire board, thus effectively simplifying the recognition process and breaking through the traditional dual-mark point recognition mode. This significantly reduces the time spent by the equipment in the mark recognition stage, thereby significantly improving overall production efficiency. Simultaneously, this solution fully utilizes the control of panelization accuracy and test point size to ensure that testing accuracy is not affected, further shortening the production cycle, making it particularly suitable for high-density panelization applications. The entire system structure is simple in design and easy to integrate into existing equipment, effectively reducing the cost of modification and upgrades.

[0051] Specifically, the panel layout parameters include the number of units M in the X direction and the number of units N in the Y direction; the data processing module calculates the reference position coordinates (X, Y, and Y) of the unit in the i-th row and j-th column using linear interpolation. ij ,Y ij The calculation formula is:

[0052]

[0053] Where i = 1, 2, ..., N; j = 1, 2, ..., M, the reference position coordinates are used to represent the positioning origin of the corresponding unit.

[0054] By using linear interpolation, the reference positioning origin of each unit in the panel can be accurately determined based on the known boundary positions (x1, y1) and (x2, y2), as well as the number of units M and N in the X and Y directions. This provides an accurate positional data basis for subsequent panel processing and positioning operations. By uniformly distributing the total length in each direction according to the number of units, the accuracy and uniformity of the positional relationship between units are ensured, which meets the requirements of panel layout for positional accuracy.

[0055] Specifically, the data processing module is further configured with a precision verification unit. This unit randomly selects at least one verification unit from among the units other than the selected unit, and controls the visual acquisition module to identify a Mark point of the verification unit to obtain its actual coordinates (x, y). a ,y a ), calculate the actual coordinates and the reference position coordinates (X). a ,Y a The Euclidean distance deviation δ between them:

[0056]

[0057] If δ is greater than the preset error threshold Δ, the value of Δ is determined based on the panel positioning accuracy and the diameter of the test point. Usually, Δ≥0.1mm, then the current recognition is deemed invalid and the re-recognition process is started.

[0058] The accuracy verification unit, by introducing a mechanism of randomly selecting verification units and detecting coordinate deviations, can effectively identify single-positioning errors exceeding the standard caused by occasional factors such as temporary interference from the visual acquisition module, sudden changes in light, lens stains, or local defects in Mark points. This prevents erroneous positioning results from being directly used in subsequent panel processing or inspection processes, thus significantly improving the reliability and stability of the overall system positioning. By setting an error threshold Δ (≥0.1mm) that matches the panel positioning accuracy and test point diameter, the scientific nature and practical applicability of the verification standard are ensured. This avoids unnecessary repeated identification due to an excessively low threshold, which would affect efficiency, and also prevents overlooking positioning deviations that could impact product quality due to an excessively high threshold. When a deviation δ greater than Δ is detected, a re-identification process is immediately initiated. This utilizes multiple identification opportunities to eliminate accidental interference, increasing the probability of successful single-positioning and reducing the risk of equipment downtime or product scrap due to single-identification failures. This, in turn, ensures the continuity and accuracy of the production or inspection process, reduces the defect rate, and improves overall work efficiency and economic benefits.

[0059] Specifically, the visual acquisition module includes an image processing unit, which performs the following operations sequentially on the acquired Mark point image:

[0060] a) Image preprocessing: Gaussian filtering function is used. Image smoothing and denoising are performed, where σ is the filtering scale parameter. Using a Gaussian filter function to smooth and denoise the image effectively eliminates random interference introduced during image acquisition due to factors such as illumination variations and sensor noise. By appropriately setting the filtering scale parameter σ, the impact of high-frequency noise on subsequent feature extraction can be significantly reduced while preserving the key feature information of Mark points. This lays a clear and stable image foundation for accurate feature extraction and avoids false edges or false feature point detections caused by noise.

[0061] b) Feature Extraction: The Canny operator is used, with the formula: gradient magnitude G = √(Gx^2 + Gy^2), direction θ = arctan(Gy / Gx), where Gx and Gy are the gradients of the image in the x and y directions, respectively. After Gaussian filtering and non-maximum suppression, the edges of the Mark points are extracted, and the center coordinates are finally located. Using the Canny operator for edge detection accurately captures areas of dramatic gray-level changes in the Mark point image, i.e., edge information. The non-maximum suppression operation after Gaussian filtering further refines the edges, retaining only pixels with the largest gray-level changes in the gradient direction, effectively eliminating non-edge pixels. This makes the extracted Mark point edges more continuous, clear, and accurately located, greatly improving the accuracy and reliability of edge detection and ensuring the complete extraction of the Mark point contours.

[0062] c) Coordinate Positioning: Calculate the centroid coordinates of the extracted contour or circle center, outputting them as the final (x, y) coordinates of the Mark point. By using the centroid coordinates as the final (x, y) coordinates of the extracted contour or circle center, and leveraging the good stability and anti-interference properties of centroid coordinates for the contour shape, the center position of the Mark point can be accurately determined even when there is some local deformation or partial occlusion in the Mark point image. This centroid-based positioning method, combined with the accurate edge extraction results from the initial stage, results in Mark point coordinates with extremely high positioning accuracy and repeatability. This provides a precise coordinate reference for subsequent Mark point-based measurement, positioning, and stitching applications, effectively ensuring the working accuracy and stability of the entire system.

[0063] Specifically, the control instructions generated by the data processing module are to generate a test path sequence based on the calculated test point coordinates, so that the working end of the test equipment moves sequentially according to the test point coordinates. The movement trajectory is optimized based on a greedy algorithm to minimize the total movement distance.

[0064] The specific operation process of optimizing the test path sequence using the greedy algorithm is as follows:

[0065] 1. Initialize the path and set of unvisited points

[0066] Set an initial test point (such as the first test point or the current position of the device), and denote it as the starting point (P). start ), add it to the path sequence (Path = [P start Construct an unvisited test point set (U) containing the coordinates of all test points (P1, P2, ..., P) except the starting point. n (n) is the total number of test points minus 1.

[0067] 2. Iteratively select the shortest distance point

[0068] From the last point (P) of the current path current (Initially (P) start Starting from point (U), traverse the unvisited point set (P) and calculate (P). current ) and each unvisited point (P i Euclidean distance (d(P)) ∈U) current ,P i Choose the point with the smallest distance. Remove it from (U) and add it to the path sequence (\text{Path}).

[0069] Euclidean distance calculation, let the current point coordinates be (P) current =(x c ,y c The coordinates of the unvisited point are (P) i =(xi ,y i The formula for the distance between two points is:

[0070]

[0071] 3. Termination Conditions

[0072] Repeat step 2 until the set of unvisited points (U) is empty. At this point, the path sequence (\text{Path}) contains all test points, forming the optimized test path.

[0073] The optimized path sequence is (Path=[P0,P1,...,P...). n (P0 is the starting point, P1 to (P) n (where ) is the test point), and the total distance traveled is the sum of the distances between all adjacent points:

[0074]

[0075] Assume the coordinates of the test points are (A(0,0)), (B(1,3)), (C(4,0)), (D(2,2)), and the starting point is (A):

[0076] Initial Path = [A], U = B, C, D;

[0077] Calculate the distances from (A) to each point: d(A,C)=4, Select the nearest point (D), update the path to ([A,D]), U = B,C;

[0078] Calculate the distances from (D) to each point: Select the nearest point (B), and update the path to ([A,D,B]), (U={C});

[0079] Finally, add (C), the path is ([A,D,B,C]), the total distance is...

[0080] Through the above process and formula, the greedy algorithm can quickly generate an approximate shortest path, thereby minimizing the distance the test equipment moves.

[0081] Specifically, the camera is a CMOS sensor with a resolution higher than 5 million pixels and a frame rate of 30fps, used to capture images of the circuit board panel; the light source is an adjustable brightness ring LED with a wavelength range of 400-700nm to ensure uniform illumination; the image processing module embeds a DSP chip and runs an image recognition algorithm to extract the coordinates of the Mark points; the data processing module PLC includes a processor, a memory, and a communication interface. The memory stores the panel parameters and calculation program, the processor executes a geometric transformation algorithm to calculate the coordinates of the test points, and controls the test equipment through the communication interface.

[0082] Employing a CMOS sensor with a resolution exceeding 5 megapixels and a frame rate of 30fps, the camera accurately captures subtle features of the PCB panel image, ensuring clear and unblurred images and providing high-quality raw data for subsequent Mark point recognition. An adjustable brightness ring LED light source covering a wavelength range of 400-700nm allows for flexible brightness adjustment based on the panel material and surface characteristics, achieving uniform illumination, effectively eliminating shadows and reflections, improving image contrast, and ensuring accurate Mark point extraction. An image processing module with an embedded DSP chip runs a high-efficiency image recognition algorithm, enabling rapid and accurate Mark point coordinate extraction, significantly improving image processing speed and meeting the real-time requirements of the production line. A PLC-based data processing module, with its built-in high-performance processor, executes complex geometric transformation algorithms, accurately calculating the coordinates of each test point based on the Mark point coordinates. Simultaneously, it transmits the coordinate data to the testing equipment in real time via a communication interface, enabling precise control of the testing equipment and ensuring that the test probes accurately contact the test points. This significantly improves the automation and accuracy of PCB panel testing, reduces human error, and enhances production efficiency and product yield.

[0083] Specifically, the system is integrated into a circuit board processing device, which is one of a flying probe tester, a chip mounter, an automatic optical inspection device, and an automatic visual inspection device.

[0084] The motion control module is communicatively connected to the main controller of the circuit board processing equipment to drive the actuator of the circuit board processing equipment to perform operations, including:

[0085] When the device is a flying probe tester, the test probe is driven to move to the test point corresponding to the reference position to perform electrical testing;

[0086] When the device is a pick and place machine, the placement head is driven to move to the placement point corresponding to the reference position to place the component;

[0087] When the device is an automatic optical inspection or automatic visual inspection device, the inspection camera is driven to move to the inspection point corresponding to the reference position to perform image acquisition and analysis.

[0088] Taking a flying probe tester as an example, in actual production, circuit board panels enter the testing phase. The system accurately captures images of the circuit board panel using the aforementioned camera, with uniform illumination to eliminate interference. The image processing module quickly extracts the coordinates of the Mark points, and the data processing module calculates the coordinates of the test points and transmits them to the flying probe tester. Upon receiving the command, the motion control module drives the test probes to move quickly and accurately to the corresponding test points to perform electrical tests on the circuit board. Due to the system's efficient identification and precise control, the tedious process of repeatedly adjusting the test probe positions manually is greatly simplified, significantly shortening the testing time. It also avoids errors that may occur due to manual operation, making the test results more reliable and improving product quality and production efficiency.

[0089] For pick-and-place machines, the system also plays a crucial role when the circuit board panel enters the placement process. It accurately calculates the coordinates of the placement points, and the motion control module drives the placement head to quickly move to the designated position for component placement. During this process, the system's high precision and high speed ensure the accuracy and efficiency of component placement, enabling the pick-and-place machine to complete a large number of placement tasks in a short time, reducing production cycles and improving production efficiency.

[0090] In applications of automated optical inspection or automated visual inspection equipment, the system drives the inspection camera to accurately move to the inspection point for image acquisition and analysis. Through clear, high-quality image acquisition, the inspection equipment can quickly detect potential defects on circuit boards, such as short circuits, open circuits, and missing components. The system's efficient operation makes the inspection process faster and more accurate, enabling timely detection and handling of problems in the production process, further improving product yield.

[0091] An efficient Mark point identification method for circuit board testing, applied to the above system, the method includes the following steps:

[0092] Step S1: Panel Fixing and Coordinate System Establishment. The circuit board panel is fixed on a high-precision carrier using positioning pins and high-precision clamps, and the base plate coordinate system O is established. base -XY;

[0093] Step S2: Baseline coordinate acquisition. Control the movement of the vision acquisition module to identify and acquire only the coordinates (x1, y1) of a first Mark point and the coordinates (x2, y2) of a second Mark point that is diagonally distributed with the first Mark point in the puzzle.

[0094] Step S3: Test point coordinate calculation. The data processing module calculates the test point coordinates of all units in the panel based on the coordinates of the first Mark point (x1, y1) and the second Mark point (x2, y2), combined with the number of rows M and columns N of the panel.

[0095] Step S4: Circuit board testing. The motion control module controls the testing equipment to perform circuit board testing according to the calculated test point coordinates.

[0096] To more clearly demonstrate the practical application effect of this efficient Mark point recognition method, a specific example is given below. Assume we have a circuit board panel with 5 rows (M) and 6 columns (N). In step S1, we securely fix this circuit board panel onto a high-precision carrier using positioning pins and high-precision clamps, successfully establishing the Obase-XY coordinate system of the substrate.

[0097] Next, in step S2, the vision acquisition module is controlled to start moving. After careful identification, the coordinates of the first Mark point are obtained as (20, 30). Since the second Mark point is diagonally distributed with the first Mark point, its coordinates are (150, 120).

[0098] Then, in step S3, the data processing module comes into play. Based on the obtained coordinates of the first Mark point (20, 30) and the second Mark point (150, 120), and combined with the number of rows (5) and columns (6) of the panel, it uses a specific algorithm to perform calculations. The calculations show that the test point coordinates of each unit in the panel are distributed reasonably and accurately. For example, the test point coordinates of the unit in the first row and first column are calculated to be (25, 35), and the test point coordinates of the unit in the first row and second column are (40, 35), and so on.

[0099] Finally, in step S4, the motion control module precisely controls the testing equipment to test the circuit board based on the calculated coordinates of these test points. During the test, the testing equipment sequentially checks the test points of each unit according to a predetermined order, enabling quick and accurate determination of whether the circuit board has a fault. If the test result of a certain test point does not meet the preset standard, the system will promptly issue an alarm, prompting the operator to conduct further inspection and handling. This example demonstrates that this efficient Mark point recognition method can effectively improve the efficiency and accuracy of circuit board testing, significantly reducing testing time and labor costs.

[0100] Specifically, the image processing module identifies the first Mark point using a template matching algorithm. The template matching algorithm performs a convolution operation between the pre-stored Mark point template image and the captured image, with the formula: similarity score S(u,v)=ΣΣ[I(x,y)*T(xu,yv)], where I(x,y) is the pixel value of the captured image, T(x,y) is the pixel value of the template image, and (u,v) is the displacement. When S(u,v) exceeds the threshold, it is determined that the recognition is successful, and the center point coordinates are extracted.

[0101] Using the template matching algorithm described above, the image processing module can accurately locate the first Mark point in the captured image and obtain its center coordinates. To further ensure the accuracy and efficiency of recognition, the captured image and template image can be converted to grayscale before performing convolution operations to reduce the interference of color information on the matching process. Simultaneously, image filtering operations can be added to remove noise points from the image.

[0102] When identifying the second Mark point, the method used for identifying the first Mark point is the same, employing a template matching algorithm. However, based on the distribution pattern of Mark points on the circuit board, the identification range can be narrowed down. For example, knowing the coordinates of the first Mark point, and combining the typical layout of the circuit board design and the panel specifications, the possible area where the second Mark point might appear can be predicted. Template matching calculations can then be performed within this narrowed area, significantly improving the identification speed.

[0103] Specifically, in step S2, the vision system controls the camera to move along the optimized path to the diagonal position. The optimized path is calculated based on the shortest path algorithm, and the path length L = √[(x2-x1)^2+(y2-y1)^2]. The moving speed is adjusted according to the camera's acceleration and deceleration to reduce the moving time.

[0104] Once the camera moves to a diagonal position, the vision system quickly activates its image capture function to acquire an image of the circuit board area at that location. The acquired image is immediately transmitted to the image processing module, which first performs grayscale conversion and image filtering operations as described earlier to improve the accuracy of subsequent Mark point recognition.

[0105] Furthermore, any content not described in detail in this specification is existing technology known to those skilled in the art.

[0106] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or efficient Markpoint identification method and system for circuit board testing that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or efficient Markpoint identification method and system for circuit board testing.

[0107] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A high-efficiency Mark point recognition system for circuit board testing, characterized in that, include: A high-precision carrier is used to fix high-density circuit board panels and ensures that the relative positional accuracy between each unit in the panel is within ±0.1mm through a positioning mechanism; The visual acquisition module, including a camera, a light source, and an image processing module, is used to acquire image data of the circuit board panel. The data processing module is communicatively connected to the vision acquisition module and is used to process the image data to identify Mark points, calculate coordinates, and generate control commands. A motion control module, connected to the data processing module, is used to drive the test equipment to perform circuit board testing according to the control commands; The data processing module is configured to: control the visual acquisition module to identify only a first Mark point and a second Mark point that are diagonally distributed with the first Mark point in a selected unit of the panel, and calculate the position information of all units to be tested on the panel based on the coordinates (x1, y1) of the first Mark point and the coordinates (x2, y2) of the second Mark point, combined with the pre-stored panel layout parameters.

2. The high-efficiency Mark point recognition system for circuit board testing according to claim 1, characterized in that, The panel layout parameters include the number of units M in the X direction and the number of units N in the Y direction; the data processing module calculates the reference position coordinates (X, Y, and Y) of the unit in the i-th row and j-th column using linear interpolation. ij ,Y ij The calculation formula is: Where i = 1, 2, ..., N; j = 1, 2, ..., M, the reference position coordinates are used to represent the positioning origin of the corresponding unit.

3. The high-efficiency Mark point recognition system for circuit board testing according to claim 2, characterized in that, The data processing module is also equipped with a precision verification unit. This unit randomly selects at least one verification unit from other units besides the selected unit, and controls the visual acquisition module to identify a Mark point of the verification unit to obtain its actual coordinates (x, y). a ,y a ), calculate the actual coordinates and the reference position coordinates (X). a ,Y a The Euclidean distance deviation δ between them: If δ is greater than the preset error threshold Δ, the value of Δ is determined based on the panel positioning accuracy and the diameter of the test point. Usually, Δ≥0.1mm, then the current recognition is deemed invalid and the re-recognition process is started.

4. The high-efficiency Mark point recognition system for circuit board testing according to claim 1, characterized in that, The visual acquisition module includes an image processing unit, which performs the following operations sequentially on the acquired Mark point images: a) Image preprocessing: Gaussian filtering function is used. The image is smoothed and denoised, where σ is the filtering scale parameter; b) Feature extraction: The Canny operator is used, with the formula: gradient magnitude G = √(Gx^2 + Gy^2), direction θ = arctan(Gy / Gx), where Gx and Gy are the gradients of the image in the x and y directions. After Gaussian filtering and non-maximum suppression, the edges of the Mark points are extracted, and the center coordinates are finally located. c) Coordinate positioning: Calculate the centroid coordinates of the extracted contour or circle center, and output them as the final coordinates (x, y) of the Mark point.

5. The high-efficiency Mark point recognition system for circuit board testing according to claim 1, characterized in that, The control commands generated by the data processing module are to generate a test path sequence based on the calculated test point coordinates, so that the test equipment working end moves sequentially according to the test point coordinates. The movement trajectory is optimized based on a greedy algorithm to minimize the total movement distance.

6. The high-efficiency Mark point recognition system for circuit board testing according to claim 1, characterized in that, The camera is a CMOS sensor with a resolution higher than 5 megapixels and a frame rate of 30fps, used to capture images of the circuit board panel. The light source is an adjustable brightness ring LED with a wavelength range of 400-700nm to ensure uniform illumination. The image processing module embeds a DSP chip and runs an image recognition algorithm to extract the coordinates of the Mark points. The data processing module PLC includes a processor, a memory, and a communication interface. The memory stores the panel parameters and calculation program, the processor executes a geometric transformation algorithm to calculate the coordinates of the test points, and controls the test equipment through the communication interface.

7. The high-efficiency Mark point recognition system for circuit board testing according to claim 1, characterized in that, The system is integrated into a circuit board processing device, which is one of a flying probe tester, a chip mounter, an automatic optical inspection device, and an automatic visual inspection device. The motion control module is communicatively connected to the main controller of the circuit board processing equipment to drive the actuator of the circuit board processing equipment to perform operations, including: When the device is a flying probe tester, the test probe is driven to move to the test point corresponding to the reference position to perform electrical testing; When the device is a pick and place machine, the placement head is driven to move to the placement point corresponding to the reference position to place the component; When the device is an automatic optical inspection or automatic visual inspection device, the inspection camera is driven to move to the inspection point corresponding to the reference position to perform image acquisition and analysis.

8. A highly efficient Mark point identification method for circuit board testing, characterized in that, Applied to the system as described in any one of claims 1 to 7, the method comprises the following steps: Step S1: Panel Fixing and Coordinate System Establishment. The circuit board panel is fixed on a high-precision carrier using positioning pins and high-precision clamps, and the base plate coordinate system O is established. base -XY; Step S2: Baseline coordinate acquisition. Control the movement of the vision acquisition module to identify and acquire only the coordinates (x1, y1) of a first Mark point and the coordinates (x2, y2) of a second Mark point that is diagonally distributed with the first Mark point in the puzzle. Step S3: Test point coordinate calculation. The data processing module calculates the test point coordinates of all units in the panel based on the coordinates of the first Mark point (x1, y1) and the second Mark point (x2, y2), combined with the number of rows M and columns N of the panel. Step S4: Circuit board testing. The motion control module controls the testing equipment to perform circuit board testing according to the calculated test point coordinates.

9. The efficient Mark point identification method for circuit board testing according to claim 8, characterized in that, In step S2, the image processing module identifies the first Mark point using a template matching algorithm. The template matching algorithm performs a convolution operation between the pre-stored Mark point template image and the captured image, with the formula: similarity score S(u,v)=ΣΣ[I(x,y)*T(xu,yv)], where I(x,y) is the pixel value of the captured image, T(x,y) is the pixel value of the template image, and (u,v) is the displacement. When S(u,v) exceeds the threshold, it is determined that the recognition is successful, and the center point coordinates are extracted.

10. The efficient Mark point identification method for circuit board testing according to claim 8, characterized in that, In step S2, the vision system controls the camera to move along the optimized path to the diagonal position. The optimized path is calculated based on the shortest path algorithm, and the path length L = √[(x2-x1)^2+(y2-y1)^2]. The moving speed is adjusted according to the camera's acceleration and deceleration to reduce the moving time.

Images