CN112669272B - AOI rapid detection method and rapid detection system - Google Patents
AOI rapid detection method and rapid detection system Download PDFInfo
- Publication number
- CN112669272B CN112669272B CN202011529747.8A CN202011529747A CN112669272B CN 112669272 B CN112669272 B CN 112669272B CN 202011529747 A CN202011529747 A CN 202011529747A CN 112669272 B CN112669272 B CN 112669272B
- Authority
- CN
- China
- Prior art keywords
- defect
- suspected
- defects
- gray
- aoi
- 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.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 68
- 230000007547 defect Effects 0.000 claims abstract description 242
- 238000012545 processing Methods 0.000 claims abstract description 24
- 239000000463 material Substances 0.000 claims abstract description 23
- 238000001914 filtration Methods 0.000 claims abstract description 17
- 239000002184 metal Substances 0.000 claims abstract description 15
- 229910052751 metal Inorganic materials 0.000 claims abstract description 15
- 230000035945 sensitivity Effects 0.000 claims abstract description 15
- RYGMFSIKBFXOCR-UHFFFAOYSA-N Copper Chemical group [Cu] RYGMFSIKBFXOCR-UHFFFAOYSA-N 0.000 claims description 59
- 229910052802 copper Inorganic materials 0.000 claims description 46
- 239000010949 copper Substances 0.000 claims description 46
- 238000000034 method Methods 0.000 claims description 10
- 239000000758 substrate Substances 0.000 claims description 9
- 238000007254 oxidation reaction Methods 0.000 claims description 8
- 230000003647 oxidation Effects 0.000 claims description 7
- 238000010586 diagram Methods 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 6
- 239000011889 copper foil Substances 0.000 claims description 5
- 238000007689 inspection Methods 0.000 claims description 4
- 230000008859 change Effects 0.000 claims description 3
- 150000002739 metals Chemical class 0.000 claims description 3
- 238000000746 purification Methods 0.000 claims description 3
- 239000013598 vector Substances 0.000 claims description 3
- 238000004519 manufacturing process Methods 0.000 abstract description 16
- 239000000047 product Substances 0.000 description 11
- 238000012216 screening Methods 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 5
- 230000003287 optical effect Effects 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000011049 filling Methods 0.000 description 3
- 238000013102 re-test Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 239000002893 slag Substances 0.000 description 2
- 208000034656 Contusions Diseases 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 208000034526 bruise Diseases 0.000 description 1
- 230000001680 brushing effect Effects 0.000 description 1
- 239000006227 byproduct Substances 0.000 description 1
- 239000011248 coating agent Substances 0.000 description 1
- 238000000576 coating method Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000005530 etching Methods 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 239000012535 impurity Substances 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012634 optical imaging Methods 0.000 description 1
- 238000007747 plating Methods 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 238000012552 review Methods 0.000 description 1
Landscapes
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention belongs to the technical field of AOI detection, and discloses an AOI rapid detection method and a rapid detection system, wherein the AOI rapid detection method comprises the steps of utilizing AOI equipment to perform scanning detection to obtain a gray-scale physical image, performing binarization processing on the obtained image through adjusting a threshold value, and determining suspected defects of a PCB; filtering part of false defects by adopting off-line rechecking; and determining true and false defects of the filtered suspected defects by using a VT recheck machine. The invention reduces false detection, improves the production efficiency of expensive AOI equipment, and saves the production cost and the detection time. The invention reasonably adjusts the threshold value according to different reflective sensitivities of the base material and metal on the surfaces of different products, and the th demarcation of the gray value reaches an ideal value as far as possible, thereby effectively reducing false alarm and also reducing equipment leakage. The invention can filter some false defects by offline rechecking, saves a great amount of time for the VT rechecking machine, and can theoretically improve the efficiency by 6 times.
Description
Technical Field
The invention belongs to the technical field of AOI detection, and particularly relates to an AOI rapid detection method and an AOI rapid detection system.
Background
At present, with the high-speed development of technology, the circuit density of the PCB is higher and higher, the layer number is also developed to be higher and higher, and the detection of the product quality is also a serious challenge while the production difficulty is improved. The traditional PCB defect needs manual inspection and cannot meet the quality and production requirements of the PCB industry, and the AOI technology which automatically detects based on the optical principle by using the digital image processing technology has high detection speed and high detection precision and is widely popularized and applied in the PCB industry.
AOI (Automated Optical Inspection) is called automatic optical detection, which obtains a digital image of a PCB board in an optical imaging mode, processes the image by using a high-speed image acquisition processing card, detects various defects on the PCB board, avoids the problem board from being sent to a post-process to continue production, increases the scrapping cost, and meets the detection requirement of a production line.
However, because AOI is based on image processing of a PCB and gives a result by comparing a statistical modeling or gray scale with a CAM file, it is not an accurate measuring instrument, the PCB is of a wide variety, and the AOI is characterized by image acquisition, different base materials, copper foils, and different defect types have different sensitivity to light, so that after comparing the acquired image with the CAM file, false detection is inevitably generated, so that the detection efficiency of AOI is greatly affected, and these efficiency differences are mainly represented by a large number of false defects generated during scanning.
Therefore, one or more rechecking machines are arranged beside the AOI host machine to manually detect whether defects reported by the AOI are true or not, and if the defects are too many, more rechecking machines are required, which clearly increases a large amount of production cost. And how to reduce the false defects reported by AOI to improve the production efficiency and the cost is very critical.
Through the above analysis, the problems and defects existing in the prior art are as follows: the existing AOI detection method has low efficiency, high cost and inaccurate detection result.
Disclosure of Invention
Aiming at the problems existing in the prior art, the invention provides an AOI rapid detection method and an AOI rapid detection system.
The invention is realized in such a way that an AOI rapid detection method comprises the following steps:
firstly, scanning and detecting by utilizing an AOI device to obtain a gray-scale physical image, and performing binarization processing on the obtained image by adjusting a threshold value to determine suspected defects of a PCB;
Step two, adopting off-line rechecking to filter part of false defects; and determining true and false defects of the filtered suspected defects by using a VT recheck machine.
In the first step, the scanning detection by the AOI device to obtain the gray scale physical image includes the following steps:
(1) Intermittently shooting the PCB images for a plurality of times by using a camera, wherein the intermittent time ensures that the adjacent shot images have an overlapping area;
(2) Performing feature detection on the shot image in a scale space, and determining the positions of key points and the scale of the key points;
(3) Extracting feature vectors irrelevant to scale scaling rotation brightness change from a plurality of images to be matched by using the main direction of the neighborhood gradient of the key points as the direction features of the key points, and matching;
(4) Performing matching point pair purification by adopting a RANSAC algorithm, and optimally calculating a perspective matrix by adopting an L-M algorithm;
(5) Exchanging corresponding images according to the perspective matrix between the images to determine an overlapping area between the images, and registering the images to be fused into a new blank image to form a mosaic;
(6) Performing vertical projection on the spliced image to eliminate left and right shadows and adhesion of the element; performing horizontal projection on the spliced image to eliminate front and rear shadows and adhesion of the element; and converting the spliced image into a gray-scale physical image by adopting a gray-scale critical value.
Further, the scanning detection by the AOI device is performed to obtain the gray-scale physical image:
1) Receiving the obtained gray scale physical image, and counting gray scale brightness values of each pixel point of the gray scale physical image to obtain a histogram of the gray scale brightness values;
2) Dividing the gray-scale brightness value into a plurality of gray-scale brightness value intervals according to the distribution characteristics of the gray-scale brightness values in the histogram;
3) Carrying out gray-scale brightness compensation on pixel points with gray-scale brightness values falling into the first type brightness interval by adopting a compensation curve corresponding to the gain value;
4) And respectively carrying out gray-scale brightness compensation on the pixel points with gray-scale brightness values falling into the second-type brightness interval by adopting gain values corresponding to two compensation curves corresponding to the adjacent first-type brightness interval.
Further, the gray-scale luminance value interval includes a first type luminance interval and a second type luminance interval located between the first type luminance intervals, and a common endpoint exists between the first type luminance interval and the second type luminance interval.
Further, in the first step, the binarizing the acquired image by adjusting the threshold includes:
And (3) carrying out threshold adjustment based on different reflection sensitivity between the surface base materials of different products and metals, and carrying out binarization processing on the obtained image.
Further, in the first step, the threshold adjustment based on the difference of the reflection sensitivity between the surface substrate and the metal of different products includes: the threshold of the substrate is lowered and the threshold of the metal is raised.
Further, in the second step, the filtering the partial false defect by using the offline recheck includes:
and acquiring the determined suspected defects, judging true and false defects based on preset defect filtering rules, and deleting the false defects.
Further, the defect filtering rule is as follows:
When the suspected defect is a copper surface or a line which is in mild and uniform subsidence or in regular descending of faults or edges, judging the defect as a true defect;
when the suspected defect is copper with small area oxidation, judging the suspected defect as a false defect;
When the suspected defect is exposed copper caused by damage to the copper surface or the circuit caused by mechanical external force, judging the suspected defect as a true defect;
when the suspected defects are copper residues, namely between wires, copper sheets and PAD, between the copper sheets and between the PAD and the copper between the PAD is not etched cleanly, judging the suspected defects as true defects;
when the suspected defect is a clamped film, judging the suspected defect to be a true defect;
when the suspected defect is sundries in the hole, judging the suspected defect as a true defect;
when the suspected defect is copper foil wrinkling, judging the suspected defect to be a true defect;
When the suspected defect is the line damage dislocation, judging the suspected defect as a true defect;
When the suspected defect is a line child, judging the suspected defect as a true defect;
when the suspected defect is a copper surface hole, judging that the suspected defect is a false defect;
when the suspected defect is incomplete, insufficient or a part of holes are plugged, judging the suspected defect as a true defect;
and when the suspected defect is a copper wire in the hole, judging the suspected defect as a true defect.
Further, in the second step, the determining the true and false defects of the filtered suspected defects by using the VT recheck machine includes:
The VT recheck machine locates the suspected defect position based on the mechanical movement of the CCD sensor in the XY axis, compares the suspected defect with the CAM file, and determines the suspected defect as a true defect or a false defect according to the product requirement.
Another object of the present invention is to provide an AOI rapid detection system implementing the AOI rapid detection method, the AOI rapid detection system including:
the scanning detection module is used for carrying out scanning detection by utilizing the AOI equipment to obtain a gray-scale object diagram;
the image processing module is used for carrying out binarization processing on the acquired image by adjusting the threshold value;
The suspected defect determining module is used for determining the suspected defects of the PCB based on the images obtained through binarization processing;
The defect filtering module is used for filtering part of false defects by adopting an improved off-line rechecking method;
and the defect determining module is used for determining true and false defects of the filtered suspected defects by utilizing a VT rechecking machine.
By combining all the technical schemes, the invention has the advantages and positive effects that:
The invention reduces false detection, improves the production efficiency of expensive AOI equipment, and saves the production cost and the detection time.
The invention reasonably adjusts the threshold value according to different reflective sensitivities of the base material and metal on the surfaces of different products, and the th demarcation of the gray value reaches an ideal value as far as possible, thereby effectively reducing false alarm and also reducing equipment leakage.
The invention reduces the threshold value of the base material, can improve the sensitivity of the equipment to the defects (such as micro short, copper slag, copper protruding and the like) on the base material, and enhances the capability of the equipment to search the short circuit defects; the sensitivity of the equipment to defects (gaps, pinholes, micro-opens and the like) on the circuit can be improved by improving the threshold value of the metal, and the capability of the equipment to search open defects is enhanced.
The invention can filter some false defects by offline rechecking, saves a great amount of time for the VT rechecking machine, and can theoretically improve the efficiency by 6 times.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments of the present application will be briefly described below, and it is obvious that the drawings described below are only some embodiments of the present application, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic diagram of an AOI rapid detection method according to an embodiment of the present invention.
Fig. 2 is a flowchart of an AOI rapid detection method according to an embodiment of the present invention.
FIG. 3 is a schematic diagram of an AOI rapid detection system according to an embodiment of the present invention;
In the figure: 1. a scanning detection module; 2. an image processing module; 3. a suspected defect determination module; 4. a defect filtering module; 5. and a defect determination module.
Fig. 4 is a schematic diagram of threshold adjustment according to an embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the following 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.
Aiming at the problems existing in the prior art, the invention provides an AOI rapid detection method and an AOI rapid detection system, and the invention is described in detail below with reference to the accompanying drawings.
As shown in fig. 1-2, the AOI rapid detection method provided by the embodiment of the present invention includes the following steps:
s101, scanning and detecting by utilizing an AOI device to obtain a gray-scale physical image, and performing binarization processing on the obtained image by adjusting a threshold value to determine suspected defects of the PCB;
S102, filtering part of false defects by adopting off-line rechecking; and determining true and false defects of the filtered suspected defects by using a VT recheck machine.
In step S101, the scanning detection by using the AOI device provided by the embodiment of the present invention to obtain a gray scale physical image includes the following steps:
Intermittently shooting the PCB images for a plurality of times by using a camera, wherein the intermittent time ensures that the adjacent shot images have an overlapping area; performing feature detection on the shot image in a scale space, and determining the positions of key points and the scale of the key points; extracting feature vectors irrelevant to scale scaling rotation brightness change from a plurality of images to be matched by using the main direction of the neighborhood gradient of the key points as the direction features of the key points, and matching; performing matching point pair purification by adopting a RANSAC algorithm, and optimally calculating a perspective matrix by adopting an L-M algorithm; exchanging corresponding images according to the perspective matrix between the images to determine an overlapping area between the images, and registering the images to be fused into a new blank image to form a mosaic; performing vertical projection on the spliced image to eliminate left and right shadows and adhesion of the element; performing horizontal projection on the spliced image to eliminate front and rear shadows and adhesion of the element; and converting the spliced image into a gray-scale physical image by adopting a gray-scale critical value.
The scanning detection by using the AOI equipment provided by the embodiment of the invention is also needed to be performed after the gray-scale physical image is obtained:
Receiving the obtained gray scale physical image, and counting gray scale brightness values of each pixel point of the gray scale physical image to obtain a histogram of the gray scale brightness values; dividing the gray-scale brightness value into a plurality of gray-scale brightness value intervals according to the distribution characteristics of the gray-scale brightness values in the histogram; carrying out gray-scale brightness compensation on pixel points with gray-scale brightness values falling into the first type brightness interval by adopting a compensation curve corresponding to the gain value; and respectively carrying out gray-scale brightness compensation on the pixel points with gray-scale brightness values falling into the second-type brightness interval by adopting gain values corresponding to two compensation curves corresponding to the adjacent first-type brightness interval.
The gray-scale brightness value interval provided by the embodiment of the invention comprises a first type brightness interval and a second type brightness interval positioned between the first type brightness intervals, wherein a common endpoint exists between the first type brightness interval and the second type brightness interval.
In step S101, the binarizing processing of the acquired image by adjusting the threshold according to the embodiment of the present invention includes:
And (3) carrying out threshold adjustment based on different reflection sensitivity between the surface base materials of different products and metals, and carrying out binarization processing on the obtained image.
In step S101, the threshold adjustment based on the difference in reflection sensitivity between the surface substrate and the metal of different products according to the embodiment of the present invention includes: the threshold of the substrate is lowered and the threshold of the metal is raised.
In step S102, the filtering of a part of false defects by offline rechecking provided by the embodiment of the present invention includes:
and acquiring the determined suspected defects, judging true and false defects based on preset defect filtering rules, and deleting the false defects.
The defect filtering rule provided by the embodiment of the invention is as follows:
When the suspected defect is a copper surface or a line which is in mild and uniform subsidence or in regular descending of faults or edges, judging the defect as a true defect;
when the suspected defect is copper with small area oxidation, judging the suspected defect as a false defect;
When the suspected defect is exposed copper caused by damage to the copper surface or the circuit caused by mechanical external force, judging the suspected defect as a true defect;
when the suspected defects are copper residues, namely between wires, copper sheets and PAD, between the copper sheets and between the PAD and the copper between the PAD is not etched cleanly, judging the suspected defects as true defects;
when the suspected defect is a clamped film, judging the suspected defect to be a true defect;
when the suspected defect is sundries in the hole, judging the suspected defect as a true defect;
when the suspected defect is copper foil wrinkling, judging the suspected defect to be a true defect;
When the suspected defect is the line damage dislocation, judging the suspected defect as a true defect;
When the suspected defect is a line child, judging the suspected defect as a true defect;
when the suspected defect is a copper surface hole, judging that the suspected defect is a false defect;
when the suspected defect is incomplete, insufficient or a part of holes are plugged, judging the suspected defect as a true defect;
and when the suspected defect is a copper wire in the hole, judging the suspected defect as a true defect.
In step S102, determining true and false defects of the filtered suspected defects by using the VT recheck machine according to the embodiment of the present invention includes:
The VT recheck machine locates the suspected defect position based on the mechanical movement of the CCD sensor in the XY axis, compares the suspected defect with the CAM file, and determines the suspected defect as a true defect or a false defect according to the product requirement.
As shown in fig. 3, in step S101, the AOI rapid detection system provided in the embodiment of the present invention includes:
The scanning detection module 1 is used for carrying out scanning detection by utilizing the AOI equipment to obtain a gray-scale object diagram;
an image processing module 2, configured to perform binarization processing on the acquired image by adjusting a threshold value;
A suspected defect determining module 3, configured to determine a suspected defect of the PCB based on the image obtained by the binarization processing;
A defect filtering module 4, configured to filter a part of the false defects by using an improved offline rechecking method;
and the defect determining module 5 is used for determining true and false defects of the filtered suspected defects by utilizing a VT rechecking machine.
The technical effects of the present invention will be further described with reference to specific examples.
Example 1
Although AOI has higher efficiency than manual detection, after all, the result is obtained through image acquisition and analysis processing, and the more production process variables are, the more complex the defect detection is. The related software technology of image analysis processing does not reach the level of human brain at present, and the principle of 'error-relieving and missing-preventing' in the design of an AOI system requires that all suspected defects must be alarmed, so that in most cases in actual use, misjudgment and missing judgment of the AOI are unavoidable. From the principle analysis of the machine, the misjudgment rate and the defect detection sensitivity can be influenced by the inspection parameters, and the system errors can be reduced to a certain extent by reasonably adjusting the system parameters. From the defect analysis of the PCB, a plurality of defects can not influence the quality of the PCB, the offline rechecking operation is simple, the cost is low, the theoretical efficiency of the offline rechecking can reach six times of that of VT rechecking, and a good platform is provided for the accurate and efficient detection of the defects of the PCB, so that the method has important practical value. In the embodiment of the invention, a black-and-white lens is adopted, if a color CCD image is used, more information can be obtained, more oxidized dirty points which do not affect the quality of a PCB can be removed, and along with the rapid development of a computer technology and a digital image processing technology, an off-line rechecking computer is matched with an AI machine learning algorithm, image data filtered by people are collected through AI, analysis and learning are carried out, a large database is established, finally, automatic screening of false defects is realized, the false defects can be reduced more efficiently, and the detection efficiency of AOI is improved.
1. AOI efficiency improving method
Through analyzing the equipment principle and the defect types, the AOI detection efficiency is improved. Firstly, based on the software adjustment in the equipment, false detection is reduced under the condition of no leakage side through reasonable parameter setting. And secondly, comparing the equipment with different points of CAM data based on manual off-line rechecking outside the equipment, and manually screening out the passing defects.
1.1 Software tuning
After the optical part of the AOI equipment acquires the 256-level gray-scale physical image, the image is required to be binarized by adjusting the threshold value, and different types of defects are emphasized by different threshold values. The threshold value is reduced, the white pixels of the image are increased, the black pixels are reduced, namely the line width is widened, and the line distance is narrowed; the threshold increases, the white pixels of the image decrease, and the black pixels increase, i.e., the line width becomes thinner and the line spacing becomes wider.
The threshold value of the base material is reduced, the sensitivity of the equipment to defects (micro short, copper slag, copper protruding and the like) on the base material can be improved, and the capability of the equipment for searching short circuit defects is enhanced; the sensitivity of the equipment to defects (gaps, pinholes, micro-opens and the like) on the circuit can be improved by improving the threshold value of the metal, and the capability of the equipment to search open defects is enhanced. While the excessively high or excessively low adjustment threshold value improves the capability of searching for defects, a plurality of defects which are slightly not influenced by products are misreported accordingly, in actual production, PCB materials of different models are different, different materials have different reflection characteristics, and the reflection characteristics directly influence the scanning result. Therefore, aiming at different reflection sensitivity of the base material and metal on the surfaces of different products, the threshold value is reasonably adjusted, and the th demarcation of the gray value reaches an ideal value as far as possible, so that false alarm can be effectively reduced, and equipment leakage can be reduced.
If the substrate threshold is too high, or the substrate has strong or weak light reflecting capability, short circuits and copper residues cannot be detected.
If the metal threshold is too low, or the substrate has strong or weak light reflecting capacity, open circuits and depressions cannot be detected.
1.2 Manual off-line rechecking
1.2.1 Manual off-line review summary
In the AOI procedure, after the PCB is subjected to AOI scanning detection, the AOI equipment firstly selects all defects or suspected objects which are possibly defects, the defect data are transmitted to a VT rechecking machine, each suspected defect is rechecked manually, and the authenticity of the defects is determined according to the requirement of a product. The software parameters can only be adjusted to ensure that the defects are ideal as far as possible and the defects of products are truly influenced, but the phenomenon of false detection can be caused by the AOI equipment inevitably due to the limitation of the principle, so that after the AOI equipment detects the defects, a screen is manually added (off-line recheck), and the false defects can be further reduced. The production flow is shown in the figure.
The advantage of adding an off-line rechecking in the two links of AOI and VT is that the VT rechecking machine needs to check the defect position through the mechanical movement of the CCD sensor on the XY axis, then manually comparing the suspected defect with the CAM file, and screening true and false defects according to the production requirement. If the false defects are too many (such as oxidization), the time generated by mechanical movement is not significant, the offline reinspection does not need mechanical movement, more defect graphs can be displayed in a user interface, more than 8 defects can be reinspected at the same time, then some false defects can be filtered according to the requirement of production quality, a great amount of time is saved for the VT reinspector, and the efficiency can be improved by 6 times theoretically.
1.2.2 Manual off-line rechecking operation flow
1. Software start-up
2. Selection of material number
A. hooking the front of the selected machine, and selecting an information source machine;
b. the triangle is pulled down and the start date and end date are selected. Determining a time range;
c. clicking to update batch numbers, waiting for 1-3 seconds, and brushing discharge numbers;
d. the material number is selected, and the material number can be selected by direct double-click. When the material number is more, the material number can be input for searching and then selecting;
e. the setting is modified, and whether the setting is modified or not can be selected according to actual conditions.
3. Modifying settings
A. The upper limit of the defect is set according to the actual situation. The part exceeding the upper limit is automatically screened out without rechecking;
b. setting minimum residence time of each picture according to the user capability;
4. defect screening
A. the interface displays eight defects once, clicks on the defect graph, and clicks twice as pass once for NG. Clicking no default pass;
b. when clicking the green pass, all are pass. When the NG is selected, all are NG;
c. after the screening of the defects of the current interface is finished, the upper left corner can be clicked to switch the page number, and the direction key can be pressed to switch the page number.
5. Ending recheck of current material number
A. When the last sheet of the current material number is rechecked, LAST PICE ARRIVED is jumped out, and the determination is pressed down;
b. clicking the right lower corner to finish the rechecking of the current material number;
6. completion status checking
After the recheck is completed, the completion state of the selected material number is confirmed. Blank indicates no retest, V indicates complete retest, and "partial" indicates partial retest.
1.2.3 False point elimination determination guideline
| Sequence number | Defect name | Defect description | Determination result | Remarks |
| 1 | Pit | The depression of the copper surface or the circuit is called dent, and the fault or the edge is orderly reduced to be called dent. | NG | Dent pits appear on the circuit, judge NG, dent appears on the copper surface, under the condition of not affecting the circuit, judge OK |
| 2 | Oxidation | Copper oxidation caused by prolonged contact with air | OK | When large area oxidation occurs, the oxidation needs to be selected and washed out before scanning |
| 3 | Wiping flower | Copper exposure caused by damage to copper surface or circuit due to mechanical external force | NG | The scratch on the circuit is judged by NG, and on the large copper surface, the scratch area is extremely small and judged by OK |
| 4 | Residual copper | The copper between the wires, the copper sheets and the PAD is not etched completely. | NG | |
| 5 | Sandwich film | The thickness of the electroplated coating is higher than that of the dry film, so that the dry film is clamped by copper plating, and a residual copper short circuit is formed after the film is difficult to remove | NG | |
| 6 | Sundries in holes | Impurities adhere to the holes after etching | NG | |
| 7 | Copper foil wrinkling | The surface copper sheet or line is corrugated | NG | |
| 8 | Line for bruise | Line damage dislocation caused by external force | NG | |
| 9 | Young wire | The actual line width is less than one third of the predetermined line width | NG | |
| 10 | Copper surface hole | Copper dew-lack base material on copper surface | OK | Holes on the large copper surface do not affect the copper surface effect, and the OK judgment is performed in a smaller range. When the hole is so large as to affect the appearance or effect, the judgment is made by NG |
| 11 | Leakage and filling of holes, incomplete filling of holes | Incomplete plugging, insufficient filling, or missing a portion of the holes | NG | |
| 12 | Copper wire in hole | Copper wires are arranged in the holes | NG |
The screening of false defects can be modified according to the quality requirements of the production site, and is not repeated here.
The foregoing is merely illustrative of specific embodiments of the present invention, and the scope of the invention is not limited thereto, but any modifications, equivalents, improvements and alternatives falling within the spirit and principles of the present invention will be apparent to those skilled in the art within the scope of the present invention.
Claims (4)
1. The AOI rapid detection method is characterized by comprising the following steps of:
firstly, scanning and detecting by utilizing an AOI device to obtain a gray-scale physical image, and performing binarization processing on the obtained image by adjusting a threshold value to determine suspected defects of a PCB;
Step two, adopting off-line rechecking to filter part of false defects; determining true and false defects of the filtered suspected defects by using a VT recheck machine;
the scanning detection by the AOI equipment is performed to obtain the gray-scale physical image, and then the following steps are performed:
1) Receiving the obtained gray scale physical image, and counting gray scale brightness values of each pixel point of the gray scale physical image to obtain a histogram of the gray scale brightness values;
2) Dividing the gray-scale brightness value into a plurality of gray-scale brightness value intervals according to the distribution characteristics of the gray-scale brightness values in the histogram;
3) Carrying out gray-scale brightness compensation on pixel points with gray-scale brightness values falling into a first type brightness interval by adopting a compensation curve corresponding to the gain value;
4) Respectively carrying out gray-scale brightness compensation on pixel points with gray-scale brightness values falling into a second-type brightness interval by adopting gain values corresponding to two compensation curves corresponding to adjacent first-type brightness intervals;
The gray-scale brightness value interval comprises a first type brightness interval and a second type brightness interval positioned between the first type brightness intervals, and common endpoints exist between the first type brightness interval and the second type brightness interval;
In the first step, the scanning detection by the AOI device to obtain the gray scale physical image includes the following steps:
(1) Intermittently shooting the PCB images for a plurality of times by using a camera, wherein the intermittent time ensures that the adjacent shot images have an overlapping area;
(2) Performing feature detection on the shot image in a scale space, and determining the positions of key points and the scale of the key points;
(3) Using the main direction of the key point neighborhood gradient as the direction characteristic of the key point, from a plurality of pieces of information to be matched
Extracting and matching feature vectors irrelevant to scale scaling rotation brightness change from the image;
(4) Performing matching point pair purification by adopting a RANSAC algorithm, and optimally calculating a perspective matrix by adopting an L-M algorithm;
(5) Exchanging corresponding images according to the perspective matrix between the images to determine an overlapping area between the images, and registering the images to be fused into a new blank image to form a mosaic;
(6) Performing vertical projection on the spliced image to eliminate left and right shadows and adhesion of the element; performing horizontal projection on the spliced image to eliminate front and rear shadows and adhesion of the element; converting the spliced image into a gray-scale physical image by adopting a gray-scale critical value;
In the first step, the threshold adjustment based on the difference of reflection sensitivity between the surface substrate and the metal of different products comprises: lowering the threshold of the substrate and raising the threshold of the metal;
in the second step, the filtering part of the false defects by adopting the off-line recheck method comprises the following steps:
Acquiring the determined suspected defects, judging true and false defects based on preset defect filtering rules, and deleting the false defects;
the defect filtering rule is as follows:
When the suspected defect is a copper surface or a line which is in mild and uniform subsidence or in regular descending of faults or edges, judging the defect as a true defect;
when the suspected defect is copper with small area oxidation, judging the suspected defect as a false defect;
When the suspected defect is exposed copper caused by damage to the copper surface or the circuit caused by mechanical external force, judging the suspected defect as a true defect;
when the suspected defects are copper residues, namely between wires, copper sheets and PAD, between the copper sheets and between the PAD and the copper between the PAD is not etched cleanly, judging the suspected defects as true defects;
when the suspected defect is a clamped film, judging the suspected defect to be a true defect;
when the suspected defect is sundries in the hole, judging the suspected defect as a true defect;
when the suspected defect is copper foil wrinkling, judging the suspected defect to be a true defect;
When the suspected defect is the line damage dislocation, judging the suspected defect as a true defect;
When the suspected defect is a line child, judging the suspected defect as a true defect;
when the suspected defect is a copper surface hole, judging that the suspected defect is a false defect;
when the suspected defect is incomplete, insufficient or a part of holes are plugged, judging the suspected defect as a true defect;
and when the suspected defect is a copper wire in the hole, judging the suspected defect as a true defect.
2. The AOI rapid detection method of claim 1, wherein in the first step, the binarizing the acquired image by adjusting the threshold value includes:
And (3) carrying out threshold adjustment based on different reflection sensitivity between the surface base materials of different products and metals, and carrying out binarization processing on the obtained image.
3. The AOI rapid inspection method of claim 1, wherein in step two, the determining the true and false defects of the filtered suspected defects using the VT recheck machine includes:
The VT recheck machine locates the suspected defect position based on the mechanical movement of the CCD sensor in the XY axis, compares the suspected defect with the CAM file, and determines the suspected defect as a true defect or a false defect according to the product requirement.
4. An AOI rapid detection system for implementing the AOI rapid detection method according to any one of claims 1 to 3, wherein the AOI rapid detection system includes:
the scanning detection module is used for carrying out scanning detection by utilizing the AOI equipment to obtain a gray-scale object diagram;
the image processing module is used for carrying out binarization processing on the acquired image by adjusting the threshold value;
The suspected defect determining module is used for determining the suspected defects of the PCB based on the images obtained through binarization processing;
The defect filtering module is used for filtering part of false defects by adopting an improved off-line rechecking method;
and the defect determining module is used for determining true and false defects of the filtered suspected defects by utilizing a VT rechecking machine.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011529747.8A CN112669272B (en) | 2020-12-22 | 2020-12-22 | AOI rapid detection method and rapid detection system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011529747.8A CN112669272B (en) | 2020-12-22 | 2020-12-22 | AOI rapid detection method and rapid detection system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112669272A CN112669272A (en) | 2021-04-16 |
| CN112669272B true CN112669272B (en) | 2024-09-06 |
Family
ID=75407683
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011529747.8A Active CN112669272B (en) | 2020-12-22 | 2020-12-22 | AOI rapid detection method and rapid detection system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112669272B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115908420A (en) * | 2023-01-10 | 2023-04-04 | 北京阿丘机器人科技有限公司 | Method, device and equipment for detecting defects of printed circuit board and storage medium |
| CN117214181A (en) * | 2023-09-18 | 2023-12-12 | 上海感图网络科技有限公司 | Material defect rechecking method, device and medium for automatic series connection of AVI |
| CN117647531B (en) * | 2023-12-27 | 2024-04-26 | 惠州学院 | Deep learning-based AOI detection method and system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105184739A (en) * | 2015-09-09 | 2015-12-23 | 苏州威盛视信息科技有限公司 | Printed circuit board AOI detection image stitching method |
| CN107610669A (en) * | 2017-10-30 | 2018-01-19 | 青岛海信电器股份有限公司 | Image gray-scale level luminance compensation method and device |
| CN110455822A (en) * | 2019-07-10 | 2019-11-15 | 苏州卓融新能源科技有限公司 | A kind of detection method of pcb board defect |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7203355B2 (en) * | 2002-12-24 | 2007-04-10 | Orbotech Ltd. | Automatic optical inspection system and method |
| CN102496161B (en) * | 2011-12-13 | 2013-10-16 | 浙江欧威科技有限公司 | Method for extracting contour of image of printed circuit board (PCB) |
| CN206821076U (en) * | 2017-05-11 | 2017-12-29 | 景旺电子科技(龙川)有限公司 | A kind of production line of reduction aluminium base AOI vacation point rates |
| CN107367682B (en) * | 2017-06-26 | 2019-11-19 | 王俊 | Printed circuit board repair method and system |
| CN109300102B (en) * | 2017-07-21 | 2021-05-04 | 凌云光技术股份有限公司 | Defect detection method for PCB |
| CN109451652A (en) * | 2018-10-25 | 2019-03-08 | 东莞泰山电子有限公司 | The test method of back-shaped route PCB line anomalies |
| CN110678003B (en) * | 2019-10-18 | 2020-07-21 | 深圳劲鑫科技有限公司 | PCB solder mask detection and repair integrated machine and process method |
| CN111586976B (en) * | 2020-04-23 | 2023-07-21 | 黄石广合精密电路有限公司 | Overhauling equipment for repairing PCB on line and overhauling method thereof |
| CN111754456B (en) * | 2020-05-15 | 2022-10-14 | 清华大学 | Two-dimensional PCB appearance defect real-time automatic detection technology based on deep learning |
-
2020
- 2020-12-22 CN CN202011529747.8A patent/CN112669272B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105184739A (en) * | 2015-09-09 | 2015-12-23 | 苏州威盛视信息科技有限公司 | Printed circuit board AOI detection image stitching method |
| CN107610669A (en) * | 2017-10-30 | 2018-01-19 | 青岛海信电器股份有限公司 | Image gray-scale level luminance compensation method and device |
| CN110455822A (en) * | 2019-07-10 | 2019-11-15 | 苏州卓融新能源科技有限公司 | A kind of detection method of pcb board defect |
Non-Patent Citations (1)
| Title |
|---|
| AOI在PCB应用中的漏检有效预防方案;黄荣琼;;印制电路信息(第06期);第60-65页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112669272A (en) | 2021-04-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112669272B (en) | AOI rapid detection method and rapid detection system | |
| CN113688807B (en) | Self-adaptive defect detection method, device, recognition system and storage medium | |
| CN112577969B (en) | Defect detection method and defect detection system based on machine vision | |
| CN115205290B (en) | Online detection method and system for PCB production process | |
| TWI669519B (en) | Board defect filtering method and device thereof and computer-readabel recording medium | |
| CN108665458A (en) | Transparent body surface defect is extracted and recognition methods | |
| CN111223078B (en) | Method for determining flaw level and storage medium | |
| CN115561247A (en) | Automatic visual inspection system for appearance of electronic component | |
| CN108918093B (en) | Optical filter mirror surface defect detection method and device and terminal equipment | |
| CN113034488A (en) | Visual detection method of ink-jet printed matter | |
| CN112669295A (en) | Lithium battery pole piece defect detection method based on secondary threshold segmentation theory | |
| CN111724375B (en) | Screen detection method and system | |
| JPH11337498A (en) | Apparatus and method for inspecting printed circuit board | |
| CN112561875A (en) | Photovoltaic cell panel coarse grid detection method based on artificial intelligence | |
| KR100827906B1 (en) | Substrate inspection device | |
| CN114998217A (en) | Method for determining defect grade of glass substrate, computer device and storage medium | |
| CN114720473A (en) | Defect detection method of vehicle-mounted screen backboard based on machine vision | |
| CN117455917B (en) | Establishment of false alarm library of etched lead frame and false alarm on-line judging and screening method | |
| JP2005164565A (en) | Defect detection method for flat panel light- related plate element in low and high resolution images | |
| CN116993654B (en) | Camera module defect detection method, device, equipment, storage medium and product | |
| CN113888539B (en) | Defect classification method, device, equipment and storage medium | |
| CN111563869A (en) | Stain testing method for quality inspection of camera module | |
| JP4247993B2 (en) | Image inspection apparatus, image inspection method, control program, and readable storage medium | |
| CN114782436B (en) | PCB defect detection method, circuit board metering measurement method and manufacturing method | |
| JP3753234B2 (en) | Defect detection method |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |