A plane interface measurement method and device based on machine vision

Abstract

The present invention provides a plane interface measurement method and device based on machine vision, including: determining calibration parameters of a machine vision measurement system; the calibration parameters include: CCD camera calibration parameters and image processing parameters of a processor; the processing parameters include: global image adjustment parameters, local image adjustment parameters, and a comparison table of each pixel point on the image and coordinates; the calibration parameters are determined by measuring a standard printed circuit board PCB board, the standard PCB board includes a plurality of circular pads and a plurality of square pads, the circular pads and the square pads have a variety of sizes, are distributed in different areas of the PCB board, and the position and size of each pad are known; the calibration parameters are used to measure the plane interface using a machine vision measurement system to identify the plane signal interface and the GND interface therein, and the position and size of each identification interface are determined based on the comparison table of pixel points and coordinates. The present invention has high measurement accuracy for the plane interface.

Description

Technical Field

[0001] The present invention belongs to the field of integrated circuit testing, and more specifically, relates to a planar interface measurement method and device based on machine vision. Background Art

[0002] In the field of circuit board testing, especially in the field of integrated circuit test system measurement, the interface of the measured electrical signal mostly appears in the form of a plane interface, generally a copper pad (pad) is pressed into contact with the plane interface by a metal probe to derive the electrical signal from the plane interface. The plane signal interface specifications of different types of circuit boards or different types of integrated circuit test systems are also different, and the main differences are inconsistencies in spacing, position distribution, size, etc. In addition, there are a large number of plane interfaces, and currently it is necessary to rely on machine vision systems and mechanical devices to automatically identify, measure and locate the interfaces in order to achieve test docking of different plane interfaces and meet the needs of multi-interface testing. In this process, relying on the accurate measurement of the machine vision system is the key link. The accuracy of the measurement results of physical quantities such as the position and size of the interface is related to whether the mechanical device can accurately locate the test interface and achieve test docking. Therefore, it is necessary to design a high-precision machine vision system to achieve accurate measurement of the interface and solve the problem of accurate measurement and positioning of multi-plane interfaces. Summary of the invention

[0003] In view of the defects of the prior art, the purpose of the present invention is to provide a planar interface measurement method and device based on machine vision, aiming to solve the problem of inaccurate planar interface measurement results in existing integrated circuit test systems.

[0004] To achieve the above object, in a first aspect, the present invention provides a plane interface measurement method based on machine vision, comprising the following steps:

[0005] A machine vision measurement system is determined, comprising: a charge coupled device (CCD) camera and a processor; the CCD camera is used to obtain image information of a plane signal interface to be measured and a ground terminal (GND) interface in the plane interface, and the processor is used to determine the position and size of the plane signal interface and the GND interface based on the image information; the pad of the plane signal interface is circular, and the pad of the GND interface is square;

[0006] Determine calibration parameters of a machine vision measurement system; the calibration parameters include: CCD camera calibration parameters and image processing parameters of a processor; the processing parameters include: global image adjustment parameters, local image adjustment parameters, and a comparison table of each pixel point on the image and coordinates; the calibration parameters are determined by measuring a standard printed circuit board (PCB) using a machine vision measurement system, the standard PCB board includes a plurality of circular pads and a plurality of square pads, the circular pads and the square pads have a variety of sizes, and are distributed in different areas of the PCB board, and the position and size of each pad are known; when the CCD camera calibration parameters are used, the image captured by the CCD camera is optimal, when the global image adjustment parameters are used, the processor has the highest recognition rate for all interfaces in the captured image, and when the local image adjustment parameters are used, the processor has the highest recognition rate for all interfaces in the corresponding local area;

[0007] The calibration parameters are adopted to measure the plane interface using a machine vision measurement system to identify the plane signal interface and the GND interface therein, and the position and size of each identified interface are determined based on the comparison table of the pixel points and the coordinates.

[0008] In an optional example, the circular pads and the square pads in the standard PCB board each include at least three sizes, which are distributed in different areas of the standard PCB board;

[0009] Circular pads of different sizes are evenly spaced on two mutually perpendicular central axes of the standard PCB board, so that the machine vision measurement system can identify the center point of the standard PCB board;

[0010] Square pads of different sizes are evenly spaced around the standard PCB board to facilitate the machine vision measurement system to identify the outer frame boundary of the standard PCB board;

[0011] The standard PCB board has circular pads of different sizes and square pads of different sizes spaced apart in four quadrants with two mutually perpendicular central axes as coordinate axes, so as to distinguish and identify planar signal interfaces and GND interfaces of different sizes and positions, so as to improve the detection and recognition rate of different interfaces by the machine vision system; the spacing distribution pattern of the circular pads and the square pads in each quadrant is different; wherein the CCD image of the circular pad corresponds to the CCD image of the planar signal interface, and the CCD image of the square pad corresponds to the CCD image of the GND interface.

[0012] In an optional example, the CCD camera parameters include: focal length, aperture size and exposure time of the CCD camera;

[0013] The CCD camera calibration parameters are determined by the following steps:

[0014] Use a CCD camera to take a picture of the standard PCB board to obtain an image of the standard PCB board;

[0015] Determine whether the best definition image is obtained, if not, adjust the parameters of the CCD camera until the best definition image is obtained;

[0016] The CCD camera parameters when obtaining the best definition image are used as the calibration parameters of the CCD camera.

[0017] In an optional example, the global image adjustment parameters and the local image adjustment parameters both include: image contrast, brightness, and grayscale;

[0018] The global image adjustment parameters and the local image adjustment parameters are determined by the following steps:

[0019] The processor sets the contrast, brightness and grayscale of the global image of the standard PCB board acquired by the CCD camera, then identifies the interface in the adjusted global image, and compares it with the interface distribution of the standard PCB board to determine the corresponding interface recognition rate as the global recognition rate;

[0020] The processor performs different settings on the contrast, brightness and grayscale of the global image to determine the global recognition rates corresponding to the different settings, and uses the parameters set when the global recognition rate is the highest as the global image adjustment parameters;

[0021] After obtaining the highest global recognition rate, the processor cuts the global image according to a preset format and outputs a plurality of local images;

[0022] The processor sets different contrast, brightness and grayscale of each local image to determine the local recognition rate corresponding to the different settings of each local image, and uses the parameters set when the local recognition rate of each local image is the highest as the adjustment parameters of each local image; the local recognition rate is determined by identifying the interfaces in the local image, and comparing the number of identified interfaces with the distribution of interfaces in the local image corresponding to the standard PCB board.

[0023] In an optional example, the comparison table between each pixel point and coordinate on the image is determined by the following steps:

[0024] After adjusting each local image to obtain the highest local recognition rate, each local image is binarized to identify the clear outline of each interface, determine the pixel points of each outline, and determine the comparison table of each pixel point and coordinate according to the position and size of each interface of the standard PCB board.

[0025] In an optional example, the interval distribution rules of the circular pads and the square pads in each quadrant area are different, and the distribution rules of the four quadrants are as follows:

[0026] The first rule: pads of the same size and type are placed in each row, circular pads and square pads are distributed adjacent to each other, the pads of the same type are distributed in different rows from large to small in size, and circular pads are placed in the row closest to the horizontal central axis;

[0027] The second rule: Place pads of the same size and type in each row. Circular pads and square pads are adjacent to each other. The sizes of pads of the same type vary from large to small in different rows. Square pads are placed in the row closest to the horizontal center axis.

[0028] The third rule: Place pads of the same size and type in each row. Circular pads and square pads are adjacent to each other. The sizes of pads of the same type vary from small to large in different rows. Circular pads are placed in the row closest to the horizontal center axis.

[0029] The fourth rule: Place pads of the same size and type in each row. Circular pads and square pads are adjacent to each other. The sizes of pads of the same type vary from small to large in different rows. Square pads are placed in the row closest to the horizontal center axis.

[0030] The above four distribution patterns appear randomly in four quadrants.

[0031] In a second aspect, the present invention provides a plane interface measurement device based on machine vision, comprising:

[0032] A parameter calibration unit is used to determine a machine vision measurement system, including: a charge coupled device (CCD) camera and a processor; the CCD camera is used to obtain image information of a plane signal interface to be measured and a ground terminal (GND) interface in a plane interface, and the processor is used to determine the position and size of the plane signal interface and the GND interface based on the image information; the pad of the plane signal interface is circular, and the pad of the GND interface is square; and determine the calibration parameters of the machine vision measurement system; the calibration parameters include: CCD camera calibration parameters and image processing parameters of the processor; the processing parameters include: global image adjustment parameters, local image adjustment parameters , a comparison table of each pixel point on the image and the coordinates; the calibration parameters are determined by measuring a standard printed circuit board PCB board using a machine vision measurement system, the standard PCB board includes a plurality of circular pads and a plurality of square pads, the circular pads and the square pads have a variety of sizes, distributed in different areas of the PCB board, and the position and size of each pad are known; when the CCD camera calibration parameters are used, the image taken by the CCD camera is the best, when the global image adjustment parameters are used, the processor has the highest recognition rate for all interfaces in the captured image, and when the local image adjustment parameters are used, the processor has the highest recognition rate for all interfaces in the corresponding local area;

[0033] The interface measurement unit is used to adopt the calibration parameters and use a machine vision measurement system to measure the planar interface to identify the planar signal interface and the GND interface therein, and determine the position and size of each identified interface based on the comparison table of pixel points and coordinates.

[0034] In an optional example, the circular pads and square pads in the standard PCB board used by the parameter calibration unit both include at least three sizes, which are distributed in different areas of the standard PCB board; circular pads of different sizes are evenly spaced on two mutually perpendicular central axes of the standard PCB board, so that the machine vision measurement system can identify the center point of the standard PCB board; square pads of different sizes are evenly spaced around the standard PCB board, so that the machine vision measurement system can identify the outer frame boundary of the standard PCB board; circular pads of different sizes and square pads of different sizes are spaced in four quadrants of the standard PCB board with two mutually perpendicular central axes as coordinate axes, so as to distinguish and identify planar signal interfaces and GND interfaces of different sizes and positions, so as to improve the detection and recognition rate of different interfaces by the machine vision system; the spacing distribution rules of the circular pads and square pads in each quadrant are different; wherein, the CCD image of the circular pad corresponds to the CCD image of the planar signal interface, and the CCD image of the square pad corresponds to the CCD image of the GND interface.

[0035] In an optional example, the CCD camera parameters include: focal length, aperture size and exposure time of the CCD camera;

[0036] The parameter calibration unit uses a CCD camera to take a picture of a standard PCB board to obtain an image of the standard PCB board; determines whether an image with the best definition is obtained, and if not, adjusts the parameters of the CCD camera until the image with the best definition is obtained; and uses the CCD camera parameters when the image with the best definition is obtained as the calibration parameters of the CCD camera.

[0037] In an optional example, the global image adjustment parameters and the local image adjustment parameters both include: image contrast, brightness, and grayscale;

[0038] The parameter calibration unit uses a processor to set the contrast, brightness and grayscale of the global image of the standard PCB board acquired by the CCD camera, then identifies the interface in the adjusted global image, and compares it with the interface distribution of the standard PCB board to determine the corresponding interface recognition rate as the global recognition rate; uses the processor to set the contrast, brightness and grayscale of the global image differently to determine the global recognition rates corresponding to different settings, and uses the parameters set when the global recognition rate is the highest as the global image adjustment parameters; after obtaining the highest global recognition rate, uses the processor to cut the global image according to a preset format and output multiple local images; and uses the processor to set the contrast, brightness and grayscale of each local image differently to determine the local recognition rates corresponding to different settings of each local image, and uses the parameters set when the local recognition rate of each local image is the highest as the adjustment parameters of each local image; the local recognition rate is determined by identifying the interfaces in the local image and comparing the number of identified interfaces with the distribution of interfaces in the local image corresponding to the standard PCB board.

[0039] In general, the above technical solution conceived by the present invention has the following beneficial effects compared with the prior art:

[0040] The present invention provides a method and device for measuring a plane interface based on machine vision. The standard PCB board used in the present invention is different from the traditional chessboard. Its material, light sensitivity, structural dimensions and other characteristics are closer to the actual plane interface to be measured, which can effectively reduce the lighting conditions required when using traditional measurement methods, improve calibration accuracy, and enhance operational convenience. In addition, the parameter calibration method described in the present invention can provide standard reference information for the optimization and extraction of image parameters in the interface recognition process, further improving the accuracy of the present invention in measuring the plane interface. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 A flow chart of a plane interface measurement method based on machine vision provided by an embodiment of the present invention;

[0042] Figure 2 The overall deployment and application scenario diagram of the measurement and calibration of plane interface dimensions based on machine vision provided in an embodiment of the present invention;

[0043] Figure 3 A specific flow chart of plane interface recognition and dimension measurement improvement based on machine vision provided by an embodiment of the present invention;

[0044] Figure 4 A standard PCB board design diagram provided by an embodiment of the present invention;

[0045] Figure 5 This is an architecture diagram of a plane interface measurement system based on machine vision provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. 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.

[0047] The present invention discloses a method for plane interface recognition optimization and measurement calibration based on machine vision, the method comprising a high-definition CCD, a standard PCB board and its design drawing; the high-definition CCD is used to identify the plane interface and measure its size and orientation information; the standard PCB contains plane interfaces of different shapes and sizes, which are distributed in different areas of the PCB board, and the accurate information of the size and orientation of the interface can be obtained through the design drawing of the standard PCB board, which is used to calibrate the CCD to ensure that the measurement result of the CCD is accurate. The standard PCB board used in the method is different from the traditional chessboard, and is closer to the real application scenario of the plane interface to be measured, which can improve the calibration accuracy, and at the same time can provide standard reference information for the optimization extraction of image parameters in the interface recognition process.

[0048] The purpose of the present invention is to overcome the above technical deficiencies and propose a method for improving the accuracy of plane interface recognition and ruler measurement based on machine vision, so as to solve the problem of accurate measurement of physical quantities such as multi-interface positions and sizes.

[0049] Figure 1 A flow chart of a plane interface measurement method based on machine vision provided by an embodiment of the present invention; Figure 1 As shown, the following steps are included:

[0050] A machine vision measurement system is determined, comprising: a charge coupled device (CCD) camera and a processor; the CCD camera is used to obtain image information of a plane signal interface to be measured and a ground terminal (GND) interface in the plane interface, and the processor is used to determine the position and size of the plane signal interface and the GND interface based on the image information; the pad of the plane signal interface is circular, and the pad of the GND interface is square;

[0051] Determine calibration parameters of a machine vision measurement system; the calibration parameters include: CCD camera calibration parameters and image processing parameters of a processor; the processing parameters include: global image adjustment parameters, local image adjustment parameters, and a comparison table of each pixel point on the image and coordinates; the calibration parameters are determined by measuring a standard printed circuit board (PCB) using a machine vision measurement system, the standard PCB board includes a plurality of circular pads and a plurality of square pads, the circular pads and the square pads have a variety of sizes, and are distributed in different areas of the PCB board, and the position and size of each pad are known; when the CCD camera calibration parameters are used, the image captured by the CCD camera is optimal, when the global image adjustment parameters are used, the processor has the highest recognition rate for all interfaces in the captured image, and when the local image adjustment parameters are used, the processor has the highest recognition rate for all interfaces in the corresponding local area;

[0052] The calibration parameters are adopted to measure the plane interface using a machine vision measurement system to identify the plane signal interface and the GND interface therein, and the position and size of each identified interface are determined based on the comparison table of the pixel points and the coordinates.

[0053] In an optional example, the circular pads and the square pads in the standard PCB board each include at least three sizes, which are distributed in different areas of the standard PCB board;

[0054] Circular pads of different sizes are evenly spaced on two mutually perpendicular central axes of the standard PCB board, so that the machine vision measurement system can identify the center point of the standard PCB board;

[0055] Square pads of different sizes are evenly spaced around the standard PCB board to facilitate the machine vision measurement system to identify the outer frame boundary of the standard PCB board;

[0056] The standard PCB board has circular pads of different sizes and square pads of different sizes spaced apart in four quadrants with two mutually perpendicular central axes as coordinate axes, so as to distinguish and identify planar signal interfaces and GND interfaces of different sizes and positions, so as to improve the detection and recognition rate of different interfaces by the machine vision system; the spacing distribution pattern of the circular pads and the square pads in each quadrant is different; wherein the CCD image of the circular pad corresponds to the CCD image of the planar signal interface, and the CCD image of the square pad corresponds to the CCD image of the GND interface.

[0057] In an optional example, the CCD camera parameters include: focal length, aperture size and exposure time of the CCD camera;

[0058] The CCD camera calibration parameters are determined by the following steps:

[0059] Use a CCD camera to take a picture of the standard PCB board to obtain an image of the standard PCB board;

[0060] Determine whether the best definition image is obtained, if not, adjust the parameters of the CCD camera until the best definition image is obtained;

[0061] The CCD camera parameters when obtaining the best definition image are used as the calibration parameters of the CCD camera.

[0062] In an optional example, the global image adjustment parameters and the local image adjustment parameters both include: image contrast, brightness, and grayscale;

[0063] The global image adjustment parameters and the local image adjustment parameters are determined by the following steps:

[0064] The processor sets the contrast, brightness and grayscale of the global image of the standard PCB board acquired by the CCD camera, then identifies the interface in the adjusted global image, and compares it with the interface distribution of the standard PCB board to determine the corresponding interface recognition rate as the global recognition rate;

[0065] The processor performs different settings on the contrast, brightness and grayscale of the global image to determine the global recognition rates corresponding to the different settings, and uses the parameters set when the global recognition rate is the highest as the global image adjustment parameters;

[0066] After obtaining the highest global recognition rate, the processor cuts the global image according to a preset format and outputs a plurality of local images;

[0067] The processor sets different contrast, brightness and grayscale of each local image to determine the local recognition rate corresponding to the different settings of each local image, and uses the parameters set when the local recognition rate of each local image is the highest as the adjustment parameters of each local image; the local recognition rate is determined by identifying the interfaces in the local image, and comparing the number of identified interfaces with the distribution of interfaces in the local image corresponding to the standard PCB board.

[0068] In an optional example, the comparison table between each pixel point and coordinate on the image is determined by the following steps:

[0069] After adjusting each local image to obtain the highest local recognition rate, each local image is binarized to identify the clear outline of each interface, determine the pixel points of each outline, and determine the comparison table of each pixel point and coordinate according to the position and size of each interface of the standard PCB board.

[0070] In an optional example, the interval distribution rules of the circular pads and the square pads in each quadrant area are different, and the distribution rules of the four quadrants are as follows:

[0071] The first rule: pads of the same size and type are placed in each row, circular pads and square pads are distributed adjacent to each other, the pads of the same type are distributed in different rows from large to small in size, and circular pads are placed in the row closest to the horizontal central axis;

[0072] The second rule: Place pads of the same size and type in each row. Circular pads and square pads are adjacent to each other. The sizes of pads of the same type vary from large to small in different rows. Square pads are placed in the row closest to the horizontal center axis.

[0073] The third rule: Place pads of the same size and type in each row. Circular pads and square pads are adjacent to each other. The sizes of pads of the same type vary from small to large in different rows. Circular pads are placed in the row closest to the horizontal center axis.

[0074] The fourth rule: Place pads of the same size and type in each row. Circular pads and square pads are adjacent to each other. The sizes of pads of the same type vary from small to large in different rows. Square pads are placed in the row closest to the horizontal center axis.

[0075] The above four distribution patterns appear randomly in four quadrants.

[0076] Figure 2 The overall deployment and application scenario diagram of the measurement and calibration of the plane interface size based on machine vision provided by the embodiment of the present invention; Figure 2 As shown, the measuring device provided by the present invention includes: a CCD, a bracket, a standard PCB board and its design drawing, and a computer.

[0077] CCD is used to take pictures of the signal (electrical signal) and GND interface on the plane interface (the interface of the measured signal), and identify and calculate the position and size of the test interface.

[0078] The computer identifies the test interface of the captured measurement image and calculates the position and size of the interface.

[0079] The standard PCB board provides standard information, assists in optimizing the setting of CCD shooting parameters, assists the computer in setting image processing parameters, optimizes shooting recognition accuracy, and improves measurement accuracy.

[0080] A standard PCB board contains pads with various obvious shape features, such as round and square. Each type of pad has a variety of sizes and is distributed according to certain rules to form a central axis, center point and outer frame boundary with obvious characteristics that can be identified by the naked eye, as well as local partitions that can be distinguished by the naked eye.

[0081] Figure 3 As shown, the present invention provides a specific method for plane interface recognition, measurement and calibration based on machine vision, and the method is:

[0082] 1. First, calibrate the machine vision measurement system and place the standard PCB board within the CCD measurement range. Figure 3 As shown, plane interface identification and measurement of parameters such as spacing, position, and size are performed. The specific steps include:

[0083] Under actual working conditions, the center point of the standard PCB board is placed at the center point of the CCD imaging and photographed. By continuously adjusting the CCD focal length, aperture size, exposure time and other parameters, the captured image has better clarity and contrast. The CCD setting parameters at this time are recorded as the preset standard parameters for the subsequent actual measurement process.

[0084] The captured image is globally parameterized, including but not limited to contrast, brightness, grayscale, etc., to obtain the highest one-time global recognition rate of the interface. At the same time, image segmentation is performed on the areas with low recognition rates in the measured image, and parameters of the segmented local images are adjusted separately to improve their local recognition rates. The relevant global adjustment coefficients and local adjustment coefficients are recorded as preset empirical parameter values.

[0085] By comparing the standard PCB board and its design drawing, a one-to-one mapping is performed with the identified interface to establish a pixel-level coordinate association table.

[0086] 2. Based on the preset parameters and coordinate association table obtained by calibration, improve the recognition accuracy and measurement accuracy of the subsequent CCD for the measured interface. The specific steps are as follows:

[0087] The high-definition camera CCD takes pictures of the plane interface board under test, and the computer assists in setting the parameters of the CCD according to the calibration results of the standard PCB board, including focal length, aperture size, exposure time and other parameters to obtain the best clarity.

[0088] The computer performs contrast optimization, brightness adjustment, binarization and other operations on the captured image, identifies the clear outline of the target interface, converts the center point of the outline into pixels, and calculates the interface's orientation, size, spacing and other parameters.

[0089] The computer processes the captured images in a global balanced and local optimized manner, and the parameter setting criteria during the processing are based on the calibration of the standard PCB board. The global parameters of the measured image to be processed are adjusted through the empirical parameter values ​​obtained through the calibration of the standard PCB board, and the specific parameters include but are not limited to contrast, brightness, grayscale, etc., to obtain the highest one-time global recognition rate; at the same time, the image is segmented for the areas with low recognition rates in the measured image, and the parameters of the segmented local images are adjusted separately to improve their local recognition rates, thereby improving the overall recognition rate.

[0090] If necessary, the CCD exposure parameters can be adjusted to shoot the plane interface board multiple times, and the processing can be performed separately according to the above method. The processed recognition results can be superimposed with the previous recognition to improve the overall recognition rate.

[0091] like Figure 4 As shown, a standard PCB board for calibration proposed by the present invention is different from the traditional checkerboard used for calibration. The standard PCB board includes pads of various shapes such as round and square, and each type of pad includes at least 3 sizes, which are distributed in different areas of the PCB.

[0092] Circular pads of different sizes are evenly spaced on the central axis to identify the center point, and square pads are distributed around to identify the outer frame boundary of the PCB board. Circular pads and square pads are distributed at intervals in the 4-quadrant area to distinguish the signal end and GND end of the plane interface.

[0093] Figure 5 The plane interface measurement system architecture diagram based on machine vision provided by the embodiment of the present invention is as follows: Figure 5 As shown, including:

[0094] The parameter calibration unit 510 is used to determine the machine vision measurement system, including: a charge coupled device CCD camera and a processor; the CCD camera is used to obtain image information of a plane signal interface to be measured and a ground terminal GND interface in the plane interface, and the processor is used to determine the position and size of the plane signal interface and the GND interface based on the image information; the pad of the plane signal interface is circular, and the pad of the GND interface is square; and determine the calibration parameters of the machine vision measurement system; the calibration parameters include: CCD camera calibration parameters and image processing parameters of the processor; the processing parameters include: global image adjustment parameters, local image adjustment parameters The calibration parameters are determined by measuring a standard printed circuit board (PCB) using a machine vision measurement system. The standard PCB includes a plurality of circular pads and a plurality of square pads. The circular pads and the square pads have a variety of sizes and are distributed in different areas of the PCB. The position and size of each pad are known. When the CCD camera calibration parameters are used, the image taken by the CCD camera is optimal. When the global image adjustment parameters are used, the processor has the highest recognition rate for all interfaces in the captured image. When the local image adjustment parameters are used, the processor has the highest recognition rate for all interfaces in the corresponding local area.

[0095] The interface measurement unit 520 is used to adopt the calibration parameters and use a machine vision measurement system to measure the planar interface to identify the planar signal interface and the GND interface therein, and determine the position and size of each identified interface based on the comparison table of pixel points and coordinates.

[0096] Understandably, Figure 5 The detailed functional implementation of each unit in the method can be found in the introduction of the aforementioned method embodiment, which will not be repeated here.

[0097] It will be easily understood by those skilled in the art that 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 and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A plane interface measurement method based on machine vision, characterized in that: The steps include: A machine vision measurement system is determined, comprising: a charge coupled device (CCD) camera and a processor; the CCD camera is used to obtain image information of a plane signal interface to be measured and a ground terminal (GND) interface in the plane interface, and the processor is used to determine the position and size of the plane signal interface and the GND interface based on the image information; the pad of the plane signal interface is circular, and the pad of the GND interface is square; Determine the calibration parameters of the machine vision measurement system; the calibration parameters include: CCD camera calibration parameters and processor processing parameters for images; the processing parameters include: global image adjustment parameters, local image adjustment parameters, and a comparison table of each pixel point on the image and the coordinates; the calibration parameters are determined by measuring a standard printed circuit board (PCB) using a machine vision measurement system, the standard PCB board includes a plurality of circular pads and a plurality of square pads, the circular pads and the square pads have a variety of sizes, and are distributed in different areas of the PCB board, and the position and size of each pad are known; when the CCD camera calibration parameters are used, the image captured by the CCD camera is optimal, when the global image adjustment parameters are used, the processor has the highest recognition rate for all interfaces in the captured image, and when the local image adjustment parameters are used, the processor has the highest recognition rate for all interfaces in the corresponding local area; the global image adjustment parameters and the local image adjustment parameters both include: image contrast, brightness and grayscale; the global image adjustment parameters and the local image adjustment parameters are determined by the following steps: processing The processor sets the contrast, brightness and grayscale of the global image of the standard PCB board acquired by the CCD camera, then identifies the interface in the adjusted global image, and compares it with the interface distribution of the standard PCB board to determine the corresponding interface recognition rate as the global recognition rate; the processor sets the contrast, brightness and grayscale of the global image differently to determine the global recognition rates corresponding to the different settings, and uses the parameters set when the global recognition rate is the highest as the global image adjustment parameters; after obtaining the highest global recognition rate, the processor cuts the global image according to a preset format and outputs a plurality of local images; the processor sets the contrast, brightness and grayscale of each local image differently to determine the local recognition rates corresponding to the different settings of each local image, and uses the parameters set when the local recognition rate of each local image is the highest as the adjustment parameters of each local image; the local recognition rate is determined by identifying the interfaces in the local image, and comparing the number of identified interfaces with the distribution of interfaces in the local image corresponding to the standard PCB board; The calibration parameters are adopted to measure the plane interface using a machine vision measurement system to identify the plane signal interface and the GND interface therein, and the position and size of each identified interface are determined based on a comparison table of pixel points and coordinates.

2. The method according to claim 1, characterized in that The circular pads and the square pads in the standard PCB board both include at least three sizes, which are distributed in different areas of the standard PCB board; Circular pads of different sizes are evenly spaced on two mutually perpendicular central axes of the standard PCB board, so that the machine vision measurement system can identify the center point of the standard PCB board; Square pads of different sizes are evenly spaced around the standard PCB board to facilitate the machine vision measurement system to identify the outer frame boundary of the standard PCB board; The standard PCB board has circular pads of different sizes and square pads of different sizes spaced apart in four quadrants with two mutually perpendicular central axes as coordinate axes, so as to distinguish and identify planar signal interfaces and GND interfaces of different sizes and positions, so as to improve the detection and recognition rate of different interfaces by the machine vision system; the spacing distribution pattern of the circular pads and the square pads in each quadrant is different; wherein the CCD image of the circular pad corresponds to the CCD image of the planar signal interface, and the CCD image of the square pad corresponds to the CCD image of the GND interface.

3. The method according to claim 1 or 2, characterized in that: The CCD camera parameters include: focal length, aperture size and exposure time of the CCD camera; The CCD camera calibration parameters are determined by the following steps: Use a CCD camera to take a picture of the standard PCB board to obtain an image of the standard PCB board; Determine whether the best definition image is obtained, if not, adjust the parameters of the CCD camera until the best definition image is obtained; The CCD camera parameters when obtaining the best definition image are used as the calibration parameters of the CCD camera.

4. The method according to claim 1 or 2, characterized in that: The comparison table between each pixel point and coordinate on the image is determined by the following steps: After adjusting each local image to obtain the highest local recognition rate, each local image is binarized to identify the clear outline of each interface, determine the pixel points of each outline, and determine the comparison table of each pixel point and coordinate according to the position and size of each interface of the standard PCB board.

5. The method according to claim 2, characterized in that: The interval distribution rules of the circular pads and the square pads in each quadrant are different, and the distribution rules of the four quadrants are as follows: The first rule: pads of the same size and type are placed in each row, circular pads and square pads are distributed adjacent to each other, the pads of the same type are distributed in different rows from large to small in size, and circular pads are placed in the row closest to the horizontal central axis; The second rule: Place pads of the same size and type in each row. Circular pads and square pads are adjacent to each other. The sizes of pads of the same type vary from large to small in different rows. Square pads are placed in the row closest to the horizontal center axis. The third rule: Place pads of the same size and type in each row. Circular pads and square pads are adjacent to each other. The sizes of pads of the same type vary from small to large in different rows. Circular pads are placed in the row closest to the horizontal center axis. The fourth rule: Place pads of the same size and type in each row. Circular pads and square pads are adjacent to each other. The sizes of pads of the same type vary from small to large in different rows. Square pads are placed in the row closest to the horizontal center axis. The above four distribution patterns appear randomly in four quadrants.

6. A plane interface measurement device based on machine vision, characterized in that: include: A parameter calibration unit is used to determine a machine vision measurement system, including: a charge coupled device (CCD) camera and a processor; the CCD camera is used to obtain image information of a plane signal interface to be measured and a ground terminal (GND) interface in a plane interface, and the processor is used to determine the position and size of the plane signal interface and the GND interface based on the image information; the pad of the plane signal interface is circular, and the pad of the GND interface is square; and determine the calibration parameters of the machine vision measurement system; the calibration parameters include: CCD camera calibration parameters and image processing parameters of the processor; the processing parameters include: global image adjustment parameters, local image adjustment parameters , a comparison table of each pixel point on the image and the coordinates; the calibration parameters are determined by measuring a standard printed circuit board PCB board using a machine vision measurement system, the standard PCB board includes a plurality of circular pads and a plurality of square pads, the circular pads and the square pads have a variety of sizes, distributed in different areas of the PCB board, and the position and size of each pad are known; when the CCD camera calibration parameters are used, the image taken by the CCD camera is the best, when the global image adjustment parameters are used, the processor has the highest recognition rate for all interfaces in the captured image, and when the local image adjustment parameters are used, the processor has the highest recognition rate for all interfaces in the corresponding local area; The global image adjustment parameters and the local image adjustment parameters both include: image contrast, brightness and grayscale; The parameter calibration unit uses a processor to set the contrast, brightness and grayscale of the global image of the standard PCB board acquired by the CCD camera, then identifies the interface in the adjusted global image, and compares it with the interface distribution of the standard PCB board to determine the corresponding interface recognition rate as the global recognition rate; uses the processor to set the contrast, brightness and grayscale of the global image differently to determine the global recognition rates corresponding to different settings, and uses the parameters set when the global recognition rate is the highest as the global image adjustment parameters; after obtaining the highest global recognition rate, uses the processor to cut the global image according to a preset format and output multiple local images; and uses the processor to set the contrast, brightness and grayscale of each local image differently to determine the local recognition rates corresponding to different settings of each local image, and uses the parameters set when the local recognition rate of each local image is the highest as the adjustment parameters of each local image; the local recognition rate is determined by identifying the interfaces in the local image and comparing the number of identified interfaces with the distribution of interfaces in the local image corresponding to the standard PCB board; The interface measurement unit is used to adopt the calibration parameters and use a machine vision measurement system to measure the planar interface to identify the planar signal interface and the GND interface therein, and determine the position and size of each identified interface based on the comparison table of pixel points and coordinates.

7. The device according to claim 6, characterized in that The circular pads and square pads in the standard PCB board used by the parameter calibration unit both include at least three sizes and are distributed in different areas of the standard PCB board; circular pads of different sizes are evenly spaced on two mutually perpendicular central axes of the standard PCB board, so that the machine vision measurement system can identify the center point of the standard PCB board; square pads of different sizes are evenly spaced around the standard PCB board, so that the machine vision measurement system can identify the outer frame boundary of the standard PCB board; circular pads of different sizes and square pads of different sizes are spaced in four quadrants of the standard PCB board with two mutually perpendicular central axes as coordinate axes, so as to distinguish and identify planar signal interfaces and GND interfaces of different sizes and positions, so as to improve the detection and recognition rate of different interfaces by the machine vision system; the spacing distribution rules of the circular pads and the square pads in each quadrant are different; wherein the CCD image of the circular pad corresponds to the CCD image of the planar signal interface, and the CCD image of the square pad corresponds to the CCD image of the GND interface.

8. The device according to claim 6 or 7, characterized in that The CCD camera parameters include: focal length, aperture size and exposure time of the CCD camera; The parameter calibration unit uses a CCD camera to take a picture of a standard PCB board to obtain an image of the standard PCB board; determines whether an image with the best definition is obtained, and if not, adjusts the parameters of the CCD camera until the image with the best definition is obtained; and uses the CCD camera parameters when the image with the best definition is obtained as the calibration parameters of the CCD camera.

Images