CN115479952A - 3D-based visual detection method and equipment - Google Patents
3D-based visual detection method and equipment Download PDFInfo
- Publication number
- CN115479952A CN115479952A CN202211268115.XA CN202211268115A CN115479952A CN 115479952 A CN115479952 A CN 115479952A CN 202211268115 A CN202211268115 A CN 202211268115A CN 115479952 A CN115479952 A CN 115479952A
- Authority
- CN
- China
- Prior art keywords
- image
- height
- layer
- camera
- detection
- 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
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan 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
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/01—Arrangements or apparatus for facilitating the optical investigation
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan 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
- G01N2021/8887—Scan 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
- Chemical & Material Sciences (AREA)
- Biochemistry (AREA)
- Pathology (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Immunology (AREA)
- Analytical Chemistry (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Signal Processing (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention relates to a method and equipment based on 3D vision detection, wherein the method comprises the following steps: acquiring an image of a detected object at a first position through a high-frame-frequency camera, and preprocessing the image to obtain a first-layer photo; adjusting the height of the camera shooting equipment, acquiring the image of the detection object at a height different from the first position, preprocessing the image to obtain a second layer of pictures, repeatedly adjusting the height of the camera shooting equipment, obtaining a group of different layers of pictures at different positions, stacking the pictures obtained by shooting, extracting the clearest outline of each layer of pictures, arranging each layer of pictures according to the height to obtain a preliminary layer cloud picture, and processing and restoring the 3D effect picture through a software smoothing algorithm. The detection method has the advantages that the imaging speed is high by about 300 milliseconds, and lines and wafers can be completely seen.
Description
Technical Field
The invention relates to the field of chip detection, in particular to a method and equipment based on 3D vision detection.
Background
At present, in order to improve the quality after wire bonding in the chip industry, the appearance of each wire bonding on a chip board of a chip needs to be manually detected, and finished products with defects of re-bonding, damaged ball necks, wire collapse (wire bonding), pressed wire arcs, electrode pulling, bias welding of two welding points, bias welding, wire arc types, overlong wire tails, missing welding, sliding balls, metal mixed wires, dislocation of B points, line warping, wire breakage, two welding positions, radian deformation, stains, less glue, glue overflow, more glue, damage, scratches, contamination, wafer tilting, stele and the like are removed.
At present, compound eye detection equipment is adopted to realize large imaging speed and high imaging speed, but welding wires cannot be seen completely, the imaging effect is as shown in figure 5, spectrum co-shooting equipment is adopted to realize large imaging speed and low imaging speed, and complete welding wires and welding points cannot be seen, the imaging effect is as shown in figure 6, light source detection equipment is adopted, the speed is high, linear imaging is poor, the heights of the welding points and the wafer cannot be imaged normally, and the imaging effect is as shown in figure 7.
Disclosure of Invention
In order to solve the above problems, the present invention provides a 3D vision inspection method based on bonding wires.
In order to achieve the purpose, the invention adopts the following technical scheme that the 3D visual detection method based on the bonding wire comprises the following steps:
s1, setting the lowest point and the highest point of a detection object, and dividing the height into a plurality of layers with equal height;
s2, taking each picture on each layer through a high frame frequency camera;
acquiring an image of a detected object at a first position through a high-frame-frequency camera, and preprocessing the image to obtain a first-layer photo; adjusting the height of the camera shooting equipment, acquiring an image of the detected object at a height different from the first position, preprocessing the image to obtain a second layer of photos, repeatedly adjusting the height of the camera shooting equipment, and obtaining a group of different layer of photos at different positions;
s3, extracting the clearest outline of each layer of graphics, performing secondary fuzzy definition calculation on each picture, solving the theoretically clearest area of each picture, removing the fuzzy area, and converting the clear position of each picture into a depth map with a depth value;
s4, dividing the depth map into a plurality of times for averaging to enable the data to be closer to the real situation, removing noise points through median filtering, and filtering unreasonable depth through filtering with too small depth values;
and S5, extracting corresponding pixels of each layer of picture, and synthesizing a fully focused 2D image or 3D.
Preferably, the preprocessing the image specifically includes extracting an ROI image from the acquired image;
carrying out binarization processing on the ROI image;
performing edge detection on the ROI image area, and extracting the outline of a target area image; the target area image comprises bonding wires and a wafer;
performing morphological processing on the outline image to enable the edge of the wafer and the bonding wire to be more prominent;
and extracting the contour of the welding line.
A bonding wire 3D detection camera for realizing a bonding wire-based 3D visual detection method comprises a camera device;
the camera shooting height adjusting mechanism is connected with the camera shooting equipment so as to adjust the shooting height of the camera shooting equipment;
the image processing part is connected with the camera equipment, extracts ROI images from the received images, detects edges of the ROI images, extracts the outlines of the target area images and morphologically processes the outline images;
the image analysis part is connected with the image processing part and is used for detecting the defect state of the image processed by the image processing part and detecting the matching state of the defect state of the images at different shooting heights after the images are processed by the image processing part to obtain a detection result;
and the output part is connected with the defect state detection part, displays the NG image obtained by the image analysis part through a display, and outputs a full-depth-of-field 2D image, a 3D depth image and a point cloud image.
Preferably, the image pickup apparatus includes a high frame rate camera and a large-field-of-view shallow depth lens connected to the high frame rate camera.
Preferably, the target area image includes bonding wires and a wafer.
The invention has the beneficial effects that: the invention detects the defects of the chip IC element after welding wires, such as re-welding, damaged ball neck, wire collapse (welding wire), pressed wire arc, electrode pull, secondary welding bias, bias welding, wire arc type, overlong wire tail, missing welding, sliding ball, metal wire, B point dislocation, wire warping, wire breakage, secondary welding position, radian deformation, stain, less glue, glue overflow, more glue, damage, scratch, stain, wafer warping, tombstoning and the like, and outputs the judgment parameters of OK and NG for the IC element according to the sorting rule set by the equipment, and can simultaneously output a full depth-of-field 2D picture, a 3D picture fully fused with 3D and 2D, a 3D depth picture, a dot cloud picture and other 3D format files.
Drawings
FIG. 1 is a flow chart of the detection method of the present invention.
Fig. 2 is a structural diagram of a 3D inspection camera for bonding wires required by the inspection method of the present invention.
Fig. 3 is a graph of 2D effect output by the detection of the present invention.
Fig. 4 is a 3D effect diagram of the detection output of the present invention.
Fig. 5 is a diagram of the effect of imaging with a compound eye detection device (no weld line visible).
Fig. 6 is a graph of the effect of imaging using a spectral co-illumination device (the bond wires and pads are not visible in their entirety).
FIG. 7 is a diagram showing the effect of imaging with the light source inspection device (poor line-type imaging, abnormal imaging of the solder joint and wafer height).
Description of the labeling: 11. a high frame rate camera; 12. a large-field-of-view and shallow-depth-of-field lens; 21. a servo lift motor; 22. a slide rail; 23. a Z-axis moving platform; 24. lens mounting bracket.
Detailed Description
Referring to fig. 1-4, the present invention relates to a method for 3D vision inspection based on bonding wires, the method comprising:
setting the lowest point and the highest point of a detection object, wherein the detection object is a chip with bonding wires, dividing the height of the detection object into a plurality of layers with the same height, acquiring an image of the detection object at a first position through a camera device, and preprocessing the image to obtain a first layer picture;
adjusting the height of the camera equipment, acquiring an image of the detected object at a height different from the first position, preprocessing the image to obtain a second layer of photos, repeatedly adjusting the height of the camera equipment, and obtaining a group of different layers of photos at different positions, namely taking one picture at each layer by using a high frame frequency camera 11;
stacking the shot pictures, extracting the clearest outline of each layer of graph, specifically, performing secondary fuzzy definition calculation on each picture, solving the theoretically clearest area of each picture, and removing the fuzzy area;
arranging each layer of image according to height to obtain a preliminary layer cloud image, namely converting the clear position of each image into a depth image with a depth value;
the depth map is divided into a plurality of times for averaging, so that the data is closer to the real situation;
removing noise points through median filtering;
filtering unreasonable depth by filtering the depth value too small;
and finally, extracting corresponding pixels of each layer of picture, and synthesizing a fully focused 2D image or 3D image according to the depth data.
Compared with an image shot generally, the image can restore details with different heights in an object, the image is clearer, the color is richer, and compared with single-lens fixed shooting image, the image can show defects more suitably for defect detection.
Preferably, the preprocessing the image specifically includes extracting an ROI image from the acquired image;
carrying out binarization processing on the ROI image;
performing edge detection on the ROI image area, and extracting the outline of a target area image; wherein the target area image comprises bonding wires and a wafer;
performing morphological processing on the outline image to enable the edge of the wafer and the bonding wire to be more prominent;
and extracting the contour of the welding line.
A bonding wire 3D detection camera for realizing a bonding wire based 3D visual detection method comprises,
an image pickup apparatus; the image pickup device is characterized in that a large-view shallow-depth-of-field lens 12 is arranged on a high-frame-frequency camera 11, images are shot and transmitted at high speed, the frame rate of the high-frame-frequency camera 11 is 200 frames/second, 100 pictures need to be shot every millimeter, if the frame rate is lower than the frame rate, shooting is time-consuming, the device only needs 1 second at present, the large-view shallow-depth-of-field lens 12 comprises an amplifying lens and an ultra-shallow-depth-of-field lens, the high-frame-frequency camera, the amplifying lens and the ultra-shallow-depth-of-field lens are sequentially connected from top to bottom, a detected object is located below the ultra-shallow-depth-of-field lens, the larger aperture is, the smaller the depth of field is, the larger the same magnification is, the larger aperture is used for realizing the ultra-shallow depth of field, the shallow depth of field is provided, the large view field can reach 7.5mmX7.5mm, and the small view field has a small viewing range.
The camera shooting height adjusting mechanism is connected with the camera shooting equipment so as to adjust the shooting height of the camera shooting equipment; in the embodiment, the camera shooting height adjusting mechanism adopts the servo lifting module to drive the Z-axis moving platform, so that the camera shooting equipment can move up and down at high speed and accurately, the moving speed of the Z-axis platform influences imaging, and the imaging can be distorted and deformed;
the camera shooting height adjusting mechanism comprises a servo lifting module 21, a slide rail 22, a Z-axis moving platform 23 and a lens mounting frame 24, camera shooting equipment is mounted on the lens mounting frame 24, the lens mounting frame 24 is arranged on the Z-axis moving platform 23, the Z-axis moving platform 23 is slidably arranged on the slide rail 22, and the servo lifting module 21 drives the Z-axis moving platform 23 to move along the slide rail 22, so that the height of the camera shooting equipment is adjusted.
An image processing unit connected to the image pickup device, for performing ROI image extraction processing on the received image, performing edge detection on the ROI image, extracting a contour of a target region image, and performing morphological processing on the contour image;
the image analysis part is connected with the image processing part and is used for detecting the defect state of the image processed by the image processing part and detecting the matching state of the defect state of the images at different shooting heights after the images are processed by the image processing part to obtain a detection result;
and the output part is connected with the defect state detection part, displays the NG image obtained by the image analysis part through a display, and outputs a full-depth-of-field 2D image, a 3D depth image and a point cloud image.
Preferably, the target area image includes bonding wires and a wafer.
The above embodiments are merely illustrative of the preferred embodiments of the present invention, and not restrictive, and various changes and modifications to the technical solutions of the present invention may be made by those skilled in the art without departing from the spirit of the present invention, and the technical solutions of the present invention are intended to fall within the scope of the present invention defined by the appended claims.
Claims (5)
1. A method for 3D visual inspection based on bonding wires, the method comprising:
s1, setting the lowest point and the highest point of a detection object, and dividing the height of the detection object into a plurality of layers with equal height;
s2, taking each picture on each layer through a high frame frequency camera;
acquiring an image of a detected object at a first position through a high-frame-rate camera, and preprocessing the image to obtain a first-layer photo; adjusting the height of the camera shooting equipment, acquiring an image of the detected object at a height different from the first position, preprocessing the image to obtain a second layer of photos, repeatedly adjusting the height of the camera shooting equipment, and obtaining a group of different layer of photos at different positions;
s3, extracting the clearest outline of each layer of graph, performing secondary fuzzy definition calculation on each picture, solving the theoretically clearest area of each picture, removing the fuzzy area, and converting the clear position of each picture into a depth map with a depth value;
s4, dividing the depth map into a plurality of times for averaging to enable the data to be closer to the real situation, removing noise points through median filtering, and filtering unreasonable depth through filtering with too small depth values;
and S5, extracting corresponding pixels of each layer of picture, and synthesizing a fully focused 2D image or 3D.
2. The detection method according to claim 1, wherein the preprocessing of the image comprises, in particular, extracting an ROI image of the acquired image;
performing binarization processing on the ROI image;
performing edge detection on the ROI image area, and extracting the outline of a target area image; the target area image comprises bonding wires and a wafer;
performing morphological processing on the outline image to enable the edge of the wafer and the bonding wire to be more prominent;
and extracting the contour of the welding line.
3. A detection apparatus for implementing the detection method according to any one of claims 1 to 2, comprising, an image pickup apparatus;
the camera shooting height adjusting mechanism is connected with the camera shooting equipment so as to adjust the shooting height of the camera shooting equipment;
the image processing part is connected with the camera equipment, extracts ROI images from the received images, detects edges of the ROI images, extracts the outlines of the target area images and morphologically processes the outline images;
the image analysis part is connected with the image processing part and is used for detecting the defect state of the image processed by the image processing part and detecting the matching condition of the defect state of the images at different shooting heights after the images are processed by the image processing part to obtain a detection result;
and the output part is connected with the defect state detection part, displays the NG image obtained by the image analysis part through a display, and outputs a full-depth-of-field 2D image, a 3D depth image and a point cloud image.
4. The inspection apparatus of claim 3, wherein the imaging apparatus comprises a high frame rate camera and a large field of view shallow depth lens coupled to the high frame rate camera.
5. The inspection apparatus of any of claims 3 to 5, wherein the target area image includes bond wires and wafers.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211268115.XA CN115479952A (en) | 2022-10-17 | 2022-10-17 | 3D-based visual detection method and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211268115.XA CN115479952A (en) | 2022-10-17 | 2022-10-17 | 3D-based visual detection method and equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115479952A true CN115479952A (en) | 2022-12-16 |
Family
ID=84396586
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211268115.XA Pending CN115479952A (en) | 2022-10-17 | 2022-10-17 | 3D-based visual detection method and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115479952A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117152165A (en) * | 2023-11-01 | 2023-12-01 | 深圳中科精工科技有限公司 | Photosensitive chip defect detection method and device, storage medium and electronic equipment |
-
2022
- 2022-10-17 CN CN202211268115.XA patent/CN115479952A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117152165A (en) * | 2023-11-01 | 2023-12-01 | 深圳中科精工科技有限公司 | Photosensitive chip defect detection method and device, storage medium and electronic equipment |
| CN117152165B (en) * | 2023-11-01 | 2024-03-22 | 深圳中科精工科技有限公司 | Photosensitive chip defect detection method and device, storage medium and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109060827B (en) | Intelligent visual detection and identification equipment | |
| US7149344B2 (en) | Systems and methods for detecting defects in printed solder paste | |
| JP3092892B2 (en) | Semiconductor chip appearance inspection method and apparatus | |
| CN109142381B (en) | High-speed visual detection and identification equipment | |
| KR20180103701A (en) | Die bonding device and method of manufacturing semiconductor device | |
| CN115479952A (en) | 3D-based visual detection method and equipment | |
| JP2021193744A (en) | Semiconductor manufacturing equipment and method for manufacturing semiconductor device | |
| CN111693535A (en) | Touch screen defect detection equipment and method based on machine vision analysis | |
| CN113289911B (en) | Method and system for detecting defects of polyhedral material | |
| JP5079678B2 (en) | Device mounted appearance inspection apparatus and inspection method | |
| US11378520B2 (en) | Auto focus function for vision inspection system | |
| KR101732467B1 (en) | Bonding device, and method for detecting breakage in semiconductor die by bonding device | |
| JP2009198494A5 (en) | ||
| JP2009216647A (en) | Defect inspection method and defect inspection device | |
| CN114897821A (en) | Height detection method for probe card probe relative test platform | |
| JPH07260445A (en) | Method and apparatus for inspecting lead frame automatically | |
| CN117745723B (en) | Chip wire bonding quality detection method, system and storage medium | |
| JP2933109B2 (en) | Component mounting method and device | |
| Affolder et al. | Automated visual inspection and defect detection of large-scale silicon strip sensors | |
| JPH04239797A (en) | Screen printing device | |
| CN114034706B (en) | Device and method for detecting surface defects of carbon fiber part | |
| TWI752558B (en) | Die bonding device and manufacturing method of semiconductor device | |
| CN106601655B (en) | Chip monitoring device and method for chip flip process | |
| CN115060648A (en) | Metal surface defect detection device and method based on stereo camera machine vision | |
| CN117745627A (en) | Visual detection method and device for mirror image edge breakage of silicon wafer |
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 |