CN116087209A - Appearance detection method and system for gold wire bonding process - Google Patents

Appearance detection method and system for gold wire bonding process Download PDF

Info

Publication number
CN116087209A
CN116087209A CN202310086795.1A CN202310086795A CN116087209A CN 116087209 A CN116087209 A CN 116087209A CN 202310086795 A CN202310086795 A CN 202310086795A CN 116087209 A CN116087209 A CN 116087209A
Authority
CN
China
Prior art keywords
gold wire
gold
gray
area
point
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202310086795.1A
Other languages
Chinese (zh)
Inventor
梁师国
何俊
梁飞
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Accelink Technologies Co Ltd
Original Assignee
Accelink Technologies Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Accelink Technologies Co Ltd filed Critical Accelink Technologies Co Ltd
Priority to CN202310086795.1A priority Critical patent/CN116087209A/en
Publication of CN116087209A publication Critical patent/CN116087209A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84Systems specially adapted for particular applications
    • G01N21/88Investigating the presence of flaws or contamination
    • G01N21/8806Specially adapted optical and illumination features
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/01Arrangements or apparatus for facilitating the optical investigation
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84Systems specially adapted for particular applications
    • G01N21/88Investigating the presence of flaws or contamination
    • G01N21/8851Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84Systems specially adapted for particular applications
    • G01N21/88Investigating the presence of flaws or contamination
    • G01N21/95Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N35/00Automatic analysis not limited to methods or materials provided for in any single one of groups G01N1/00 - G01N33/00; Handling materials therefor
    • G01N35/00584Control arrangements for automatic analysers
    • G01N35/0092Scheduling
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/01Arrangements or apparatus for facilitating the optical investigation
    • G01N2021/0106General arrangement of respective parts
    • G01N2021/0112Apparatus in one mechanical, optical or electronic block
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84Systems specially adapted for particular applications
    • G01N21/88Investigating the presence of flaws or contamination
    • G01N21/8851Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • G01N2021/8887Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques

Landscapes

  • Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • General Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Signal Processing (AREA)
  • Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)

Abstract

The invention relates to the technical field of machine vision detection, in particular to an appearance detection method and system of a gold wire bonding process, wherein a reference picture and a plurality of background pictures of gold wires are obtained through different light source modes; and detecting defects such as spacing, bending, crossing, wire breakage and the like of the gold wires to be detected according to the set detection requirements. The invention realizes full-automatic machine vision detection of the optical communication product through flexible light source configuration and detection method, has the characteristics of high detection adaptability and strong anti-interference, has good detection stability and product detection rate, and realizes automatic inspection of gold wire bonding technology.

Description

Appearance detection method and system for gold wire bonding process
Technical Field
The invention relates to the technical field of machine vision, in particular to an appearance detection method and system for a gold wire bonding process.
Background
The photoelectric detector serving as a core component of the optical communication receiving end product has the annual shipment quantity of tens of millions. In order to cope with large-scale production, the main flow process realizes full-automatic production. However, in auxiliary processes such as inspection or manual visual inspection, not only is the efficiency low, but also there is a problem that missed inspection occurs due to personnel difference, visual fatigue and the like. Machine vision is well-developed for defect detection in the semiconductor industry and the PCB (Printed Circuit Board ) industry, however, the technologies are directly applied to optical communication products, and have the characteristics of high false alarm rate and poor adaptability of product detection programs. The reason is that the optical communication product has larger process and material variation, and the element height difference of the same product is large, a plurality of alternative elements exist, and the material chromatic aberration tolerance is inconsistent; the texture of the product background is complex and has large interference and the contrast of gold wires is poor; the variety of wire loops is large and the bending inflection point is highly fluctuating, and these variations are where the machine vision detection apparatus and method are sensitive, so the detection system and method need to be tailored to the characteristics of the optical communication product.
In view of this, overcoming the drawbacks of the prior art is a problem to be solved in the art.
Disclosure of Invention
The invention aims to solve the technical problems of providing an appearance detection method and an appearance detection system for a gold wire bonding process, and solves the problems of high defect false alarm rate and false missing alarm rate when a gold wire is detected due to the fact that the background texture of a product is complex and greatly disturbed and the obtained gold wire contrast is poor when the existing machine vision detection method is directly applied to defect detection of an optical communication product.
In a first aspect, the present invention provides a method for detecting appearance of a gold wire bonding process, including:
acquiring a reference picture of the gold wire through a first light source mode, and switching the light source mode of the composite light source 2 according to the type of the background object to acquire at least one background picture, wherein the reference picture and the background picture both comprise the gold wire and the background object;
setting a cutting line according to the contrast ratio of the gold thread and the background object in the reference picture;
dividing gold wires in the reference picture into a plurality of gold wire segments according to the dividing lines, selecting a gold wire segment with highest contrast from the background picture and the reference picture for each gold wire segment, and splicing the selected gold wire segments to obtain a gold wire to be detected which meets the contrast requirement;
and performing defect detection on the gold wire to be detected.
Further, the background object comprises a lead frame 3, a chip 4 and a capacitor 5; the composite light source 2 comprises coaxial light 24, upper annular light 23, middle annular light 22 and lower annular light 21, wherein the upper annular light 23, the middle annular light 22 and the lower annular light 21 are sequentially stacked;
the obtaining the reference picture of the gold thread through the first light source mode, and switching the light source mode of the composite light source 2 according to the type of the background object to obtain at least one background picture includes:
a reference picture of the gold wire is obtained by adopting a mode of combining and polishing the middle annular light 22 and the lower annular light 21 so as to improve the contrast ratio among the gold wire, the welding spot and the background object;
when the background object is the chip 4, the coaxial light 24 is adopted to acquire a background picture so as to improve the contrast between the gold thread and the chip 4;
when the background object is the capacitor 5, a mode of lighting by combining the coaxial light 24 and the upper annular light 23 is adopted to obtain a background picture so as to improve the contrast between the gold thread and the capacitor 5;
when the background object is the lead frame 3, the coaxial light 24 and the lower annular light 21 are combined to light to obtain a background picture, so that the contrast between the gold wire and the lead frame 3 is improved.
Further, in the reference picture, setting the slicing line according to the contrast ratio of the gold thread and the background object includes:
Analyzing the reference picture to obtain the contrast ratio of the gold thread and the background object;
and taking the position with the contrast smaller than the set contrast threshold value as a cutting line.
Further, performing defect detection on the gold wire to be detected includes:
performing gold wire rough positioning on the gold wire to be tested according to the cutting line to obtain an initial gold wire area;
accurately positioning the initial gold thread region through a gray threshold and a contrast threshold to obtain a target gold thread region, and extracting a skeleton of the target gold thread region to generate a gold thread center line;
and judging the defects of the gold wires through the geometric characteristics and gray information of the central lines of the gold wires.
Further, the performing gold wire rough positioning on the gold wire to be measured according to the cutting line, and obtaining the initial gold wire region includes:
acquiring a head welding spot and a tail welding spot of the gold wire to be tested;
sequentially acquiring a target point before each parting line, and connecting the target point with a tail end welding point to obtain a transition line; for the parting line closest to the head end welding point, the target point before the parting line is the head end welding point, and the target points corresponding to other parting lines are the positioning points before the parting line;
acquiring an intersection point of the transition line and the segmentation line, taking the intersection point as a normal line of the transition line, and taking at least one transition point on the normal line;
Connecting the transition point with a target point and a tail end welding point respectively to establish a first image area and a second image area respectively, calculating the gray scale g (A1) of the first image area and the gray scale g (A2) of the second image area respectively, and calculating the average gray scale gm of the transition point according to the gray scale g (A1) and the gray scale g (A2);
taking a transition point with the average gray gm meeting a preset condition as a positioning point of the gold wire to be detected;
and sequentially connecting the head end welding spot, all positioning points and the tail end welding spot to form an initial gold wire area.
Further, the connecting the transition point with the target point and the tail end welding point respectively to establish a first image area and a second image area respectively, calculating the gray scale g (A1) of the first image area and the gray scale g (A2) of the second image area respectively, and calculating the average gray scale gm of the transition point according to the gray scale g (A1) and the gray scale g (A2) includes:
taking the connecting line distance between the transition point and the positioning point as a first long axis and the connecting line distance between the transition point and the tail welding point as a second long axis;
establishing a first image area by taking a first long axis as the length of the rectangular area and taking the width of the gold wire as the width of the rectangular area; establishing a second image area by taking a second long axis as the length of the rectangular area and taking the width of the gold wire as the width of the rectangular area;
The gray g (A1) of the first image area and the gray g (A2) of the second image area are calculated respectively, the area A1 of the first image area and the area A2 of the second image area are calculated respectively, the gray sum of the gray g (A1) and the gray g (A2) is obtained, the area sum of the area A1 and the area A2 is obtained, and the ratio of the gray sum and the area sum is used as the average gray gm of the transition point.
Further, the step of taking the transition point with the average gray gm meeting the preset condition as the positioning point of the gold wire to be measured includes:
when the average gray value gm of the transition point is larger than or equal to the gray threshold value gt, the average gray value gm-1 of the previous transition point and the average gray value gm+1 of the next transition point are obtained;
and if the difference value between the average gray value gm and the average gray value gm-1 and the difference value between the average gray value gm and the average gray value gm+1 are larger than the set contrast threshold gc, taking the transition point as the positioning point of the gold wire to be tested.
Further, the precisely positioning the initial gold thread region through the gray threshold and the contrast threshold to obtain a target gold thread region, and extracting the skeleton of the target gold thread region to generate a gold thread center line comprises:
generating a search area along the initial gold wire area, selecting a bright area or a dark area as a target area according to the gold wire images to be detected of different segments, determining a skeleton of the gold wire to be detected in the target area, and processing the skeleton to obtain a precisely positioned gold wire central line. Further, the determining the defect of the gold wire according to the geometric feature and gray information of the gold wire center line includes:
Grouping the gold wires to be detected, sequentially solving intersection of the gold wire central lines of the groups, and judging that the gold wires to be detected are abnormal in intersection if the intersection state does not meet the preset requirement;
according to the grouping of the gold wires to be detected, sequentially calculating the distances between the gold wire central lines of each group, and judging that the spacing between the gold wires to be detected is abnormal if the distances do not meet a preset value;
generating an image area of the gold wire to be measured along the central line of the gold wire, generating two tolerance lines parallel to a straight line between a head end welding point and a tail end welding point of the gold wire to be measured, and judging that the gold wire to be measured is abnormal in bending if the image area is intersected with the tolerance lines;
calculating the length of the central line of the gold wire, and judging that the gold wire to be tested is broken if the length does not meet the preset requirement;
calculating the lengths of the central lines of the gold wires in the head-end welding spots and the tail-end welding spots of the gold wires to be tested, and judging that the gold wires to be tested are broken if the lengths do not meet the preset requirements;
calculating intersection of the annular region near the head end and the tail end and the gold wire region, extracting the gold wire outline in the region, judging the gray level and the characteristics of the outline, and judging that the gold wire is broken abnormally if the preset requirement is not met.
In a second aspect, the present invention further provides an appearance inspection system for performing the appearance inspection method of the first aspect, the appearance inspection system comprising a visual inspection module and a controller, the visual inspection module comprising a camera 1 and a composite light source 2, the composite light source 2 comprising a coaxial light 24, an upper annular light 23, an intermediate annular light 22 and a lower annular light 21, the upper annular light 23, the intermediate annular light 22 and the lower annular light 21 being sequentially stacked;
The composite light source 2 is used for providing different light source modes;
the controller is used for switching the light source mode of the composite light source 2 according to the type of the detected object;
the camera 1 is used for acquiring a reference picture of a gold wire in a first light source mode, and acquiring at least one background picture in a light source mode corresponding to a background object;
the controller is used for setting a cutting line according to the contrast ratio of the gold thread and the background object in the reference picture; and the gold thread segments are spliced to obtain gold threads to be detected which meet the contrast requirement, so that the gold threads to be detected are subjected to defect detection.
According to the embodiment of the invention, a reference picture and a plurality of background pictures of gold wires are obtained through different light source modes, in the reference picture, a cutting line is set according to the contrast ratio of the gold wires to a background object, the gold wires in the reference picture are divided into a plurality of gold wire segments according to the cutting line, for each gold wire segment, a gold wire segment with the highest contrast ratio is selected from the background picture and the reference picture, and the selected gold wire segments are spliced to obtain the gold wires to be tested which meet the contrast ratio requirement; and detecting defects such as spacing, bending, crossing, wire breakage and the like of the gold wires to be detected according to the set detection requirements. The invention realizes full-automatic machine vision detection of the optical communication product through flexible light source configuration and detection method, has the characteristics of high detection adaptability and strong anti-interference, has good detection stability and product detection rate, and realizes automatic inspection of gold wire bonding technology.
Drawings
In order to more clearly illustrate the technical solution of the embodiments of the present invention, the drawings that are required to be used in the embodiments of the present invention will be briefly described below. It is evident that the drawings described below are only some embodiments of the present invention and that other drawings may be obtained from these drawings without inventive effort for a person of ordinary skill in the art.
Fig. 1 is a flow chart of an appearance detection method of a gold wire bonding process according to an embodiment of the present invention;
fig. 2 is a schematic diagram of a structure of a product to be tested according to an appearance detection method of a gold wire bonding process according to an embodiment of the present invention; fig. 3 is a schematic structural diagram of an appearance detecting system of a gold wire bonding process according to an embodiment of the present invention;
fig. 4 is a schematic flowchart of step 101 in an appearance inspection method of a gold wire bonding process according to an embodiment of the present invention;
fig. 5 is a schematic diagram of gold thread cutting in an appearance detection method of a gold thread bonding process according to an embodiment of the present invention;
fig. 6 is a schematic flowchart of step 104 in the appearance inspection method of the gold wire bonding process according to the embodiment of the present invention;
Fig. 7 is a schematic flowchart of step 1041 in an appearance detection method of a gold wire bonding process according to an embodiment of the present invention;
FIG. 8 is a schematic diagram of rough positioning of gold wires in an appearance inspection method of a gold wire bonding process according to an embodiment of the present invention;
fig. 9 is a specific flowchart of step 10414 in the appearance detection method according to the embodiment of the present invention.
Wherein the reference numerals are as follows: a camera 1; a composite light source 2; lower annular light 21; middle annular light 22; upper annular light 23; coaxial light 24; a lead frame 3; a chip 4; and a capacitor 5.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
In the description of the present invention, the terms "inner", "outer", "longitudinal", "transverse", "upper", "lower", "top", "bottom", etc. refer to an orientation or positional relationship based on that shown in the drawings, merely for convenience of describing the present invention and do not require that the present invention must be constructed and operated in a specific orientation, and thus should not be construed as limiting the present invention.
In addition, the technical features of the embodiments of the present invention described below may be combined with each other as long as they do not collide with each other.
Example 1:
referring to fig. 1, embodiment 1 of the present invention provides an appearance detection method of a gold wire bonding process:
step 101: and acquiring a reference picture of the gold wire through the first light source mode, and switching the light source mode of the composite light source 2 according to the type of the background object to acquire at least one background picture, wherein the reference picture and the background picture both comprise the gold wire and the background object.
Step 102: in the reference picture, a cutting line is set according to the contrast ratio of the gold wire and the background object.
Step 103: and dividing the gold thread in the reference picture into a plurality of gold thread segments according to the dividing lines, selecting a gold thread segment with highest contrast from the background picture and the reference picture for each gold thread segment, and splicing the selected gold thread segments to obtain the gold thread to be detected meeting the contrast requirement.
Step 104: and performing defect detection on the gold wire to be detected.
In the embodiment, a reference picture and a plurality of background pictures of gold wires are obtained through different light source modes, in the reference picture, a cutting line is set according to the contrast ratio of the gold wires to a background object, the gold wires in the reference picture are divided into a plurality of gold wire segments according to the cutting line, for each gold wire segment, a gold wire segment with the highest contrast ratio is selected from the background picture and the reference picture, and the selected gold wire segments are spliced to obtain the gold wires to be tested which meet the contrast ratio requirement; and detecting defects such as spacing, bending, crossing, wire breakage and the like of the gold wires to be detected according to the set detection requirements. The invention realizes full-automatic machine vision detection of the optical communication product through flexible light source configuration and detection method, has the characteristics of high detection adaptability and strong anti-interference, has good detection stability and product detection rate, and realizes automatic inspection of gold wire bonding technology.
In order to obtain a reference picture and a background picture with better contrast, targeted lighting and photographing are required according to different background objects, as shown in fig. 2, the background objects comprise a lead frame 3, a chip 4 and a capacitor 5; referring to fig. 3, the composite light source 2 includes a coaxial light 24, an upper annular light 23, an intermediate annular light 22, and a lower annular light 21, and the upper annular light 23, the intermediate annular light 22, and the lower annular light 21 are sequentially stacked.
In consideration of factors such as more product element types, changeable curvature of gold wire arc, and complex texture of background object, referring to fig. 4, step 101 specifically includes:
step 1011: and a reference picture of the gold wire is acquired by adopting a mode of combining and polishing the middle annular light 22 and the lower annular light 21 so as to improve the contrast ratio among the gold wire, the welding spot and the background object.
Step 1012: when the background object is the chip 4, the on-axis light 24 is used to obtain a background picture, so as to improve the contrast between the gold wire and the chip 4.
Step 1013: when the background object is the capacitor 5, the background picture is obtained by adopting a mode of lighting by combining the coaxial light 24 and the upper annular light 23 so as to improve the contrast between the gold thread and the capacitor 5.
Step 1014: when the background object is the lead frame 3, the coaxial light 24 and the lower annular light 21 are combined to light to obtain a background picture, so that the contrast between the gold wire and the lead frame 3 is improved.
In an alternative embodiment, the lower annular light 21 is red light, the lower annular light 21 is arranged at the lowest layer of the composite light source 2, the height of the lower annular light from the product to be measured is 1mm, the incident angle is 20 degrees, and the lower annular red light can project the reflected light of the gold wire into the lens so as to facilitate imaging of the camera 1; the middle annular light 22 is blue light, the middle annular light 22 is arranged above the lower annular light 21, and the incident angle is 45 degrees; the upper annular light 23 is upper annular white light, the upper annular light 23 is arranged above the middle annular light 22, and the incident angle is 60 degrees; the coaxial light 24 is white light and is arranged above the upper annular light 23. The foregoing provides only the preferred values of the height and angle, and in other embodiments, the values may be adaptively adjusted according to the product to be tested.
In step 1011, the gold thread adopts a mode of polishing by combining the middle annular light 22 and the lower annular light 21, so that the gold thread can form a reference picture with good contrast; in step 1012, the background object is the chip 4, and since the chip 4 is a mirror surface and the inside of the chip 4 is generally textured, the texture of the chip 4 can be clearly seen by polishing with the coaxial light 24; in step 1013, the background object is the capacitor 5, and since the capacitor 5 has a mirror surface and a rough surface, but no texture, a mode of polishing by combining the coaxial light 24 and the upper annular light 23 is adopted, a background picture with clear imaging of the boundary of the capacitor 5 and obvious contrast with other backgrounds can be formed; in step 1014, the background object is the lead frame 3, and since the lead frame 3 has a lot of textures, the textures of the lead frame 3 can be homogenized by polishing the combination of the coaxial light 24 and the lower ring light 21. Moreover, the brightness of the composite light source 2 is adjustable, and the brightness of the composite light source 2 can be adjusted according to actual conditions. By the means of shining and photographing for different background objects, images with better contrast can be obtained, and later machine vision detection is facilitated.
It should be noted that, under the condition that the coaxial light 24 participates in lighting, gold wires in the photographed background picture will be displayed as black with lower gray value; when the coaxial light 24 does not participate in the lighting, that is, when various types of coaxial light 24 perform combined lighting, gold wires in the photographed reference picture are displayed as white with a higher gray value.
In order to obtain a gold thread image with good contrast, in the reference picture, setting a segmentation line according to the contrast between the gold thread and a background object includes:
analyzing the reference picture to obtain the contrast ratio of the gold thread and the background object; and taking the position with the contrast smaller than the set contrast threshold value as a cutting line.
Because the gold thread has certain crooked and span, the change of camber leads to light reflection not to the camera 1 the inside, and the unable whole section of gold thread is lighted, and the span of gold thread leads to the gold thread background to disturb a lot, these all lead to being difficult to obtain the good complete gold thread image of contrast in a picture, consequently need carry out multistage segmentation to the contrast difference of gold thread route in the benchmark picture, for every segmentation, select the good image of gold thread contrast in benchmark picture and the background picture and splice, and then obtain complete good gold thread image of contrast, the cutting line can be appointed by hand according to benchmark image and background image, also can select by the procedure. For example, as shown in fig. 5, the gold wire is divided into 4 segments (a segmentation segment 11, a segmentation segment 12, a segmentation segment 14, and a segmentation segment 15) by a segmentation line, the segmentation segment 11 and the segmentation segment 15 are closer to the chip 4 or the capacitor 5, and the background on the segmentation segment 12 and the segmentation segment 14 contains glue.
For example, when the contrast of the gold wire segment in the reference picture is good, the gold wire segment can select the reference picture; when the gold thread segment is above the chip 4, the chip 4 can lead the light to form mirror reflection, when the coaxial light 24 is adopted for illumination, the chip 4 is bright, and the gold thread is dark, so that in the background picture formed in the step 1012, the gold thread is darker in color and is black, the gray value is lower, the chip 4 is brighter in color and is white, the gray value is higher, the gold thread has good contrast, and therefore, the background picture formed in the step 1012 can be selected through the gold thread segment above the chip 4; when glue is present on the gold thread background, when the upper annular light 23 is combined with the coaxial light 24 to polish, the glue is lightened, the gold thread is dark, the gray value is low, the background color is bright, the color is white, the gray value is high, and therefore the gold thread segment with the glue on the background can select the background image formed in step 1013; the gold thread segments adopt the parts with better contrast in the background picture and the reference picture, and then are spliced to obtain gold thread images with better contrast, wherein part of gold threads are displayed in white and part of gold threads are displayed in black in the gold thread images.
In this embodiment, the appearance detection method further includes positioning and defect detection of a solder joint, and before performing defect detection on the gold wire to be detected, positioning and defect detection are performed on the solder joint to be detected. Specifically, firstly matching an element to be detected, then obtaining a welding point with a welding point reference as the welding point of the element to be detected, and searching the welding point to be detected according to the coordinate of the welding point relative to the element to be detected so as to obtain the welding point to be detected; after the welding spot to be measured is obtained, judging whether the welding spot to be measured has defects according to the diameter and the offset distance of the welding spot to be measured.
When the welding spots are searched, the welding spots are searched in a rectangular search frame mode, when the rectangular search frame searches for a region possibly provided with the welding spots, gray scale characteristic calculation, thresholding segmentation, morphological processing and processing of fitting into a circumcircle are carried out on the region, and then whether the welding spots are the welding spots or not is judged according to the fitted circumcircle area and roundness.
When the element to be tested is matched, the element to be tested is matched according to a preset matching mode, and the pattern element is matched in shape, such as a chip 4; the non-patterned elements require edge matching with scaling, such as capacitor 5, due to dimensional tolerance issues.
The element information, the welding spot information and the gold thread information are stored in a visual detection template, and the visual detection template is independently archived according to the serial numbers of products and can be called when the appearance detection method detects different products.
The element information stores the coordinates and matching modes of the elements for matching use of the elements to be tested; the welding spot information is stored with welding spot references and elements to which the welding spots belong, wherein the welding spot references comprise elements followed by the welding spots and coordinates of the welding spots relative to the elements, and the welding spot references are used for searching the welding spots to be detected according to the elements to be detected; the welding spot information is also stored with a welding spot diameter and a welding spot offset allowable range and is used for judging whether the welding spot to be detected has defects or not; the gold wire information comprises gold wire pairing information, wherein the gold wire pairing information comprises head end welding spot and tail end welding spot information of a gold wire; the gold thread information also comprises gold thread grouping information, so that the gold thread grouping is convenient to carry out defect judgment; the gold wire information also comprises gold wire spacing, gold wire bending range and gold wire crossing information, and is used for judging whether the gold wire has defects or not.
In this embodiment, referring to fig. 6, a specific process for detecting the defect of the gold wire to be detected includes:
step 1041: and carrying out rough gold wire positioning on the gold wire to be tested according to the cutting line to obtain an initial gold wire area.
Step 1042: and accurately positioning the initial gold thread region through a gray level threshold value and a contrast threshold value to obtain a target gold thread region, and extracting a skeleton of the target gold thread region to generate a gold thread center line.
Step 1043: and judging the defects of the gold wires through the geometric characteristics and gray information of the central lines of the gold wires.
Referring to fig. 7, step 1041 specifically includes the following steps:
step 0411: and acquiring a head welding spot and a tail welding spot of the gold wire to be tested.
Step 0412: sequentially acquiring a target point before each parting line, and connecting the target point with a tail end welding point to obtain a transition line; for the parting line closest to the head-end welding point, the target point before the parting line is the head-end welding point, and the target points corresponding to other parting lines are the positioning points before the parting line.
Step 0413: and acquiring an intersection point of the transition line and the segmentation line, taking the intersection point as a normal line of the transition line, and taking at least one transition point on the normal line.
Step 0414: and respectively connecting the transition point with the target point and the tail end welding point to respectively establish a first image area and a second image area, respectively calculating the gray scale g (A1) of the first image area and the gray scale g (A2) of the second image area, and calculating the average gray scale gm of the transition point according to the gray scale g (A1) and the gray scale g (A2).
Step 0415: and taking a transition point with the average gray gm meeting a preset condition as a positioning point of the gold wire to be detected.
Step 0416: and sequentially connecting the head end welding spot, all positioning points and the tail end welding spot to form an initial gold wire area.
Specifically, in connection with fig. 8, there are three parting lines A, B and C, where P1 is a head welding point and P2 is a tail welding point; firstly, connecting P1 and P2 to obtain a transition line P2P1; the parting line corresponding to the P1 is a parting line A, an intersection point Pm1 of the parting line P2P1 and the parting line A is taken, the normal line of the parting line P2P1 is taken as the normal line of the parting line P2P1, and Pm1 is taken as the transition point; connecting Pm1 with P1 and P2 respectively, and establishing a first image area and a second image area; respectively calculating the gray g (A1) of the first image area and the gray g (A2) of the second image area, and calculating the average gray gm of the transition point Pm1 according to the gray g (A1) and the gray g (A2); recording the average gray value and the transition point coordinates into a gray value set, judging whether the position of the current transition point Pm1 meets a preset condition, taking the transition point meeting the preset condition as a positioning point Pc1, otherwise, moving the transition point Pm1 along a normal line on one side of the transition line by one pixel unit at a time, simultaneously calculating the symmetry point Pm1' of the transition line by the transition point Pm1, and comparing the value with a larger average gray value in Pm1 or Pm1' each time until the position of the current transition point Pm1 or Pm1' meets the preset condition; if the positioning point is not found after the preset times of movement, taking a transition point corresponding to the maximum average gray value in the gray value set as the positioning point.
After Pc1 is found, connecting Pc1 with P2 to obtain a transition line P2Pc1; taking an intersection point Pm2 of a transition line P2Pc1 and a segmentation line B, taking Pc1 as a normal line of the transition line P2Pc1, and taking Pm2 as a transition point; connecting Pm2 with P1 and P2 respectively, and establishing a first image area and a second image area; pm2 moves along the normal line, and judges whether the average gray level of the gray level g (A1) of the first image area and the gray level g (A2) of the second image area meet the requirement or not until Pc2 is obtained; the method for obtaining Pc3 is the same as that for obtaining Pc1 and Pc2, and will not be described in detail.
Referring to fig. 9, step 0414 specifically includes the following steps:
step 4141: the connecting line distance between the transition point and the positioning point is used as a first long axis, and the connecting line distance between the transition point and the tail welding point is used as a second long axis.
Step 4142: establishing a first image area by taking a first long axis as the length of the rectangular area and taking the width of the gold wire as the width of the rectangular area; and taking the second long axis as the length of the rectangular area, and taking the width of the gold wire as the width of the rectangular area, so as to establish a second image area.
Step 4143: the gray g (A1) of the first image area and the gray g (A2) of the second image area are calculated respectively, the area A1 of the first image area and the area A2 of the second image area are calculated respectively, the gray sum of the gray g (A1) and the gray g (A2) is obtained, the area sum of the area A1 and the area A2 is obtained, and the ratio of the gray sum and the area sum is used as the average gray gm of the transition point.
Referring to fig. 8, after Pm1 is connected to P1 and P2, respectively, straight lines P1Pm1 and P2Pm1 are obtained; establishing a rectangular area taking P1Pm1 as a long side and taking the width as the width of the gold wire to be measured as a first image area, and establishing a rectangular area taking P1Pm2 as a long side and taking the width as the width of the gold wire to be measured as a second image area;
the average gray gm of the transition point is calculated by the gray g (A1) of the first image area and the gray g (A2) of the second image area, the area A1 of the first image area and the area A2 of the second image area, and the specific formula is expressed as:
gm=(g(A1)+g(A2))÷(A1+A2)
in step 0415, the process of using the transition point with the average gray gm satisfying the preset condition as the positioning point of the gold wire to be tested includes:
when the average gray value gm of the transition point is larger than or equal to the gray threshold value gt, the average gray value gm-1 of the previous transition point and the average gray value gm+1 of the next transition point are obtained; and if the difference value between the average gray value gm and the average gray value gm-1 and the difference value between the average gray value gm and the average gray value gm+1 are larger than the set contrast threshold gc, taking the transition point as the positioning point of the gold wire to be tested. Meanwhile, the gray threshold value gt is not a fixed value, is determined by the gold wire image to be detected, and can be set by a program or manually.
In step 1042, accurately positioning the initial gold thread region by using a gray threshold and a contrast threshold to obtain a target gold thread region, and extracting a skeleton of the target gold thread region to generate a gold thread center line comprises:
generating a search area along the initial gold wire area, selecting a bright area or a dark area as a target area according to the gold wire color change in the gold wire image to be detected according to different segmented gold wire images to be detected, determining the skeleton of the gold wire to be detected in the target area, and processing the skeleton to obtain a precisely positioned gold wire central line. Specifically, since the gold wire center line may be above or below the initial gold wire region, it is necessary to increase the width of the initial gold wire region to a certain range with the initial gold wire region as the center, generate an enlarged search region, and precisely position the gold wire center line in the enlarged search region.
Because the gold thread is displayed in black when the coaxial white light is polished, the gray value is lower, the background is displayed in white, and the gray value is higher; when the ring light combination is polished, the gold thread is displayed in white, the gray value is higher, the background is displayed in black, and the gray value is lower; the gold wire image to be measured is formed by splicing a plurality of images, a part of gold wires are displayed in black, and a part of gold wires are displayed in white, so that when the gold wires are precisely positioned, a bright area or a dark area is selected as a target area according to the change of the gold wire color in the gold wire image to be measured, the search of a gold wire skeleton is performed, the skeleton is the central line outline of the gold wires, specifically, when the gold wires are darker in color and are black, the dark area is selected as the target area, and when the gold wires are brighter in color and are white, the bright area is selected as the target area, so that the image outline of the central line of the gold wires can be obtained in the image of the gold wires to be measured.
After the target area is determined, the contrast difference between the gold thread image area and the background picture is larger, and meanwhile, the closer to the central line of the gold thread, the larger the contrast difference between the gold thread image area and the background picture is; the image contour of the gold thread center line in the target area can be obtained through the comparison of the contrast of the target area image and the background image; after the image contour of the gold thread center line is obtained, the image contour of the gold thread center line is screened and smoothed to obtain the precisely positioned gold thread center line.
In this embodiment, the defects of the gold wires are determined by the geometric features and gray information of the center line of the gold wires, wherein the defect types include poor intersection abnormality of the gold wires, abnormal spacing between the gold wires, abnormal bending of the gold wires, broken wire of the gold wires, few wires of the gold wires, and the like, and the detection modes of the defect types are as follows:
grouping the gold wires to be detected, sequentially solving intersection of the gold wire central lines of the groups, and judging that the gold wires to be detected are abnormal in crossing if the intersection state does not meet the preset requirement.
And according to the grouping of the gold wires to be detected, sequentially calculating the distances between the gold wire central lines of each group, and judging that the spacing between the gold wires to be detected is abnormal if the distances do not meet the preset value.
Generating an image area of the gold wire to be measured along the central line of the gold wire, generating two tolerance lines parallel to the straight line between the head end welding point and the tail end welding point of the gold wire to be measured, and judging that the gold wire to be measured is abnormal in bending if the image area is intersected with the tolerance lines.
Calculating the length of the central line of the gold wire, and judging that the gold wire to be tested is broken if the length does not meet the preset requirement.
Calculating the lengths of the central lines of the gold wires in the head-end welding spots and the tail-end welding spots of the gold wires to be tested, and judging that the gold wires to be tested are broken if the lengths do not meet the preset requirements. If the lengths of the central lines of the gold wires in the head-end welding spots and the tail-end welding spots meet the requirements, calculating the lengths of the middle sections of the gold wires to be detected, and if the lengths of the middle sections of the gold wires to be detected do not meet the preset requirements, judging that the gold wires to be detected are broken.
Calculating intersection of annular areas near the head and tail ends and gold thread areas, extracting the gold thread outline in the areas, judging the gray level and the characteristics of the outline, judging that the gold thread is broken if the preset requirements are not met.
And when the gold wire to be tested cannot extract the central line, judging that the gold wire to be tested is missing and the gold wire is few.
Specifically, the distance and the cross relation between the gold thread center lines can be calculated according to the gold thread center lines extracted from the gold thread to be detected, whether the distance and the cross relation between the gold threads to be detected meet the requirements or not is judged, and the gold thread to be detected is grouped.
And when judging whether the bending state of the gold wire to be measured is abnormal, generating an image contour of the gold wire to be measured according to the central line of the gold wire, setting two tolerance lines, and if the image contour of the gold wire to be measured touches the tolerance lines, considering that the gold wire to be measured is abnormal in bending.
Judging whether the gold wire to be tested is broken or not mainly through the length of the gold wire central line of the gold wire to be tested, and if the length of the gold wire central line does not meet the length requirement, considering that the gold wire to be tested is broken. The reason for the disconnection discrimination of the head end welding point, the tail end welding point and the middle part of the gold wire to be detected is that the gold wire is easy to be disconnected at the welding point end and the central end, and the disconnection discrimination of the gold wire is divided, so that the complexity of the disconnection discrimination of the gold wire is increased, but the accuracy of defect detection is improved.
When the gold wires are less, the gold wires to be detected can finish coarse positioning, but finish fine positioning of the gold wires is not finished, namely, the central line of the gold wires cannot be extracted between the head end welding point and the tail end welding point, and at the moment, the gold wires less can be directly judged.
Example 2:
this embodiment 2 provides an appearance inspection system for performing the appearance inspection method of embodiment 1, the appearance inspection system including a visual inspection module and a controller not shown in the drawing, as shown in fig. 3, the visual inspection module including a camera 1 and a composite light source 2, the composite light source 2 including a coaxial light 24, an upper annular light 23, an intermediate annular light 22, and a lower annular light 21, the upper annular light 23, the intermediate annular light 22, and the lower annular light 21 being sequentially stacked.
The composite light source 2 is used for providing different light source modes; the controller is used for switching the light source mode of the composite light source 2 according to the type of the detected object; the camera 1 is configured to obtain a reference picture of a gold thread in a first light source mode, and obtain at least one background picture in a light source mode corresponding to a background object.
The controller is used for setting a cutting line according to the contrast ratio of the gold thread and the background object in the reference picture; and the gold thread segments are spliced to obtain gold threads to be detected which meet the contrast requirement, so that the gold threads to be detected are subjected to defect detection.
Wherein, the camera 1 adopts black-white CCD, the minimum defect is required to be detected 1.5 mu m, the pixel of the camera 1 is 1500 ten thousand, the depth of field of the camera 1 lens is not less than 0.4mm, and the magnification is 1 time. The lower annular light 21 is red light, the lower annular light 21 is arranged at the lowest layer of the composite light source 2, the height of the composite light source from a product to be measured is 1mm, the incident angle is 20 degrees, and the lower annular red light can project the reflected light of gold wires into a lens so as to facilitate imaging of the camera 1; the middle annular light 22 is blue light, the middle annular light 22 is arranged above the lower annular light 21, and the incident angle is 45 degrees; the upper annular light 23 is upper annular white light, the upper annular light 23 is arranged above the middle annular light 22, and the incident angle is 60 degrees; the coaxial light 24 is white light and is arranged above the upper annular light 23. The on-axis light 24, the upper annular light 23, the middle annular light 22 and the lower annular light 21 in the composite light source 2 can be combined arbitrarily, the brightness and the incident angle can be adjusted, and the composite light source can be used for flexibly using a proper light source combination according to a detection object to obtain higher image contrast.
The gold wire can form a reference picture with good contrast by adopting a mode of polishing by combining the middle annular light 22 and the lower annular light 21; when the background object is the chip 4, the chip 4 is a mirror surface, and the inside of the chip 4 is generally provided with textures, and the textures of the chip 4 can be clearly seen by adopting a coaxial light 24 polishing mode; when the background object is the capacitor 5, as the capacitor 5 has a mirror surface and a rough surface, but no texture exists, a mode of polishing by combining the coaxial light 24 and the upper annular light 23 is adopted, and a background picture which has clear imaging of the boundary of the capacitor 5 and obvious contrast with other backgrounds can be formed; when the background object is the lead frame 3, the lead frame 3 has more textures, and at the moment, the textures of the lead frame 3 can be homogenized by adopting a mode of polishing by combining the coaxial light 24 and the lower annular light 21; and the brightness of the composite light source 2 is adjustable, and the brightness of the composite light source 2 can be adjusted according to actual conditions. By adopting different light source combinations to carry out lighting photographing aiming at different background objects, images with better contrast can be obtained, and the later machine vision detection is facilitated.
The appearance detection system further comprises an automatic feeding module, an automatic sorting module and an automatic discharging module, and the controller controls the moving mechanisms in the automatic feeding module, the automatic sorting module and the automatic discharging module to execute full-automatic feeding and discharging.
The specific implementation manner of the appearance detection method is described in embodiment 1, and is not described herein.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, and alternatives falling within the spirit and principles of the invention.

Claims (10)

1. The appearance detection method of the gold wire bonding process is characterized by comprising the following steps of:
acquiring a reference picture of the gold wire through a first light source mode, and switching the light source mode of the composite light source according to the type of the background object to acquire at least one background picture, wherein the reference picture and the background picture both comprise the gold wire and the background object;
setting a cutting line according to the contrast ratio of the gold thread and the background object in the reference picture;
dividing gold wires in the reference picture into a plurality of gold wire segments according to the dividing lines, selecting a gold wire segment with highest contrast from the background picture and the reference picture for each gold wire segment, and splicing the selected gold wire segments to obtain a gold wire to be detected which meets the contrast requirement;
and performing defect detection on the gold wire to be detected.
2. The appearance inspection method of claim 1, wherein the background object comprises a lead frame, a chip, and a capacitor; the composite light source comprises coaxial light, upper annular light, middle annular light and lower annular light, and the upper annular light, the middle annular light and the lower annular light are sequentially arranged in a stacked mode;
The obtaining the reference picture of the gold thread through the first light source mode, and switching the light source mode of the composite light source according to the type of the background object to obtain at least one background picture comprises:
obtaining a reference picture of the gold wire by adopting a mode of polishing the combination of the middle annular light and the lower annular light so as to improve the contrast ratio among the gold wire, the welding spot and the background object;
when the background object is a chip, acquiring a background picture by adopting coaxial light so as to improve the contrast between the gold thread and the chip;
when the background object is a capacitor, a mode of polishing by combining coaxial light and upper annular light is adopted to obtain a background picture so as to improve the contrast between the gold thread and the capacitor;
when the background object is a lead frame, a mode of polishing by combining coaxial light and lower annular light is adopted to obtain a background picture so as to improve the contrast between the gold thread and the lead frame.
3. The appearance detecting method according to claim 1, wherein the setting a cut line in the reference picture according to a contrast of a gold wire and a background object includes:
analyzing the reference picture to obtain the contrast ratio of the gold thread and the background object;
and taking the position with the contrast smaller than the set contrast threshold value as a cutting line.
4. The appearance inspection method of claim 1, wherein performing defect inspection on the gold wire under inspection comprises:
performing gold wire rough positioning on the gold wire to be tested according to the cutting line to obtain an initial gold wire area;
accurately positioning the initial gold thread region through a gray threshold and a contrast threshold to obtain a target gold thread region, and extracting a skeleton of the target gold thread region to generate a gold thread center line;
and judging the defects of the gold wires through the geometric characteristics and gray information of the central lines of the gold wires.
5. The method of claim 4, wherein the performing rough positioning on the gold wire according to the cutting line to obtain an initial gold wire region includes:
acquiring a head welding spot and a tail welding spot of the gold wire to be tested;
sequentially acquiring a target point before each parting line, and connecting the target point with a tail end welding point to obtain a transition line; for the parting line closest to the head end welding point, the target point before the parting line is the head end welding point, and the target points corresponding to other parting lines are the positioning points before the parting line;
acquiring an intersection point of the transition line and the segmentation line, taking the intersection point as a normal line of the transition line, and taking at least one transition point on the normal line;
Connecting the transition point with a target point and a tail end welding point respectively to establish a first image area and a second image area respectively, calculating the gray scale g (A1) of the first image area and the gray scale g (A2) of the second image area respectively, and calculating the average gray scale gm of the transition point according to the gray scale g (A1) and the gray scale g (A2);
taking a transition point with the average gray gm meeting a preset condition as a positioning point of the gold wire to be detected;
and sequentially connecting the head end welding spot, all positioning points and the tail end welding spot to form an initial gold wire area.
6. The appearance inspection method according to claim 5, wherein the connecting the transition point with the target point and the tail end welding point, respectively, to establish a first image area and a second image area, respectively, calculating a gray g (A1) of the first image area and a gray g (A2) of the second image area, respectively, and calculating an average gray gm of the transition point according to the gray g (A1) and the gray g (A2) includes:
taking the connecting line distance between the transition point and the positioning point as a first long axis and the connecting line distance between the transition point and the tail welding point as a second long axis;
establishing a first image area by taking a first long axis as the length of the rectangular area and taking the width of the gold wire as the width of the rectangular area; establishing a second image area by taking a second long axis as the length of the rectangular area and taking the width of the gold wire as the width of the rectangular area;
The gray g (A1) of the first image area and the gray g (A2) of the second image area are calculated respectively, the area A1 of the first image area and the area A2 of the second image area are calculated respectively, the gray sum of the gray g (A1) and the gray g (A2) is obtained, the area sum of the area A1 and the area A2 is obtained, and the ratio of the gray sum and the area sum is used as the average gray gm of the transition point.
7. The appearance inspection method according to claim 5, wherein the step of using the transition point of the average gray gm satisfying the preset condition as the positioning point of the gold wire to be inspected comprises:
when the average gray value gm of the transition point is larger than or equal to the gray threshold value gt, the average gray value gm-1 of the previous transition point and the average gray value gm+1 of the next transition point are obtained;
and if the difference value between the average gray value gm and the average gray value gm-1 and the difference value between the average gray value gm and the average gray value gm+1 are larger than the set contrast threshold gc, taking the transition point as the positioning point of the gold wire to be tested.
8. The method of claim 4, wherein the accurately positioning the initial gold wire region by the gray threshold and the contrast threshold to obtain a target gold wire region, and extracting the skeleton of the target gold wire region to generate a gold wire center line comprises:
Generating a search area along the initial gold wire area, selecting a bright area or a dark area as a target area according to the gold wire color change in the gold wire image to be detected according to different segmented gold wire images to be detected, determining the skeleton of the gold wire to be detected in the target area, and processing the skeleton to obtain a precisely positioned gold wire central line.
9. The method of claim 4, wherein the determining the defect of the gold wire by the geometric feature of the center line of the gold wire and the gray information comprises:
grouping the gold wires to be detected, sequentially solving intersection of the gold wire central lines of the groups, and judging that the gold wires to be detected are abnormal in intersection if the intersection state does not meet the preset requirement;
according to the grouping of the gold wires to be detected, sequentially calculating the distances between the gold wire central lines of each group, and judging that the spacing between the gold wires to be detected is abnormal if the distances do not meet a preset value;
generating an image area of the gold wire to be measured along the central line of the gold wire, generating two tolerance lines parallel to a straight line between a head end welding point and a tail end welding point of the gold wire to be measured, and judging that the gold wire to be measured is abnormal in bending if the image area is intersected with the tolerance lines;
calculating the length of the central line of the gold wire, and judging that the gold wire to be tested is broken if the length does not meet the preset requirement;
Calculating the lengths of the central lines of the gold wires in the head-end welding spots and the tail-end welding spots of the gold wires to be tested, and judging that the gold wires to be tested are broken if the lengths do not meet the preset requirements;
calculating intersection of the annular region near the head end and the tail end and the gold wire region, extracting the gold wire outline in the region, judging the gray level and the characteristics of the outline, and judging that the gold wire is broken abnormally if the preset requirement is not met.
10. An appearance inspection system of a wire bonding process for performing the appearance inspection method of any one of claims 1 to 9, characterized in that the appearance inspection system comprises a visual inspection module and a controller, the visual inspection module comprising a camera and a composite light source, the composite light source comprising a coaxial light, an upper annular light, a middle annular light, and a lower annular light, the upper annular light, the middle annular light, and the lower annular light being sequentially stacked;
the composite light source is used for providing different light source modes;
the controller is used for switching the light source mode of the composite light source according to the type of the detected object;
the camera is used for acquiring a reference picture of the gold wire in a first light source mode, and acquiring at least one background picture in a light source mode corresponding to the background object;
The controller is used for setting a cutting line according to the contrast ratio of the gold thread and the background object in the reference picture; and the gold thread segments are spliced to obtain gold threads to be detected which meet the contrast requirement, so that the gold threads to be detected are subjected to defect detection.
CN202310086795.1A 2023-01-17 2023-01-17 Appearance detection method and system for gold wire bonding process Pending CN116087209A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310086795.1A CN116087209A (en) 2023-01-17 2023-01-17 Appearance detection method and system for gold wire bonding process

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310086795.1A CN116087209A (en) 2023-01-17 2023-01-17 Appearance detection method and system for gold wire bonding process

Publications (1)

Publication Number Publication Date
CN116087209A true CN116087209A (en) 2023-05-09

Family

ID=86206156

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310086795.1A Pending CN116087209A (en) 2023-01-17 2023-01-17 Appearance detection method and system for gold wire bonding process

Country Status (1)

Country Link
CN (1) CN116087209A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117933175A (en) * 2024-03-25 2024-04-26 杭州广立微电子股份有限公司 Layout equivalent line width calculation method and device, electronic device and storage medium

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117933175A (en) * 2024-03-25 2024-04-26 杭州广立微电子股份有限公司 Layout equivalent line width calculation method and device, electronic device and storage medium

Similar Documents

Publication Publication Date Title
CN108765416B (en) PCB surface defect detection method and device based on rapid geometric alignment
JP6630545B2 (en) Positioning method, positioning device, program, and computer-readable recording medium
US9075026B2 (en) Defect inspection device and defect inspection method
JP5947169B2 (en) Appearance inspection apparatus, appearance inspection method and program
US9773304B2 (en) Inspection apparatus, inspection method, and program
CN111127571B (en) Small sample defect classification method and device
EP0898163B1 (en) Method and apparatus for automatic inspection of moving surfaces
CN101526484B (en) Bearing defect detecting technique based on embedded-type machine vision
EP1943502B1 (en) Apparatus and methods for inspecting a composite structure for defects
CN109523541A (en) A kind of metal surface fine defects detection method of view-based access control model
CN102439708B (en) Check method and the connected structure inspection machine of the connected structure of substrate
CN208207914U (en) PCB surface defect detecting device based on fast geometric alignment
CN102410811A (en) Method and system for measuring parameters of bent pipe
EP2387796A2 (en) System and method for inspecting a wafer
CN107610085A (en) A kind of welding point defect detecting system based on computer vision
CN116087209A (en) Appearance detection method and system for gold wire bonding process
CN114136975A (en) Intelligent detection system and method for surface defects of microwave bare chip
CN100538347C (en) The automatic optical detecting system of membrane type or belt printed circuit board and disposal route
CN117092113A (en) Camera mould welding quality detection device and system thereof
CN109701890A (en) Magnetic tile surface defect detection and method for sorting
CN116256366A (en) Chip defect detection method, detection system and storage medium
CN115775245A (en) Coil winding detection method and device, electronic equipment and storage medium
CN114392940B (en) Pin detection method and device for special-shaped component
CN114813761A (en) Double-light-stroboscopic-based film pinhole and bright spot defect identification system and method
CN114037682A (en) Two-dimensional automatic detection method for optical element surface defects

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination