CN115825104A - Soldering tin defect detection method and system based on integration volume calculation - Google Patents
Soldering tin defect detection method and system based on integration volume calculation Download PDFInfo
- Publication number
- CN115825104A CN115825104A CN202211324836.8A CN202211324836A CN115825104A CN 115825104 A CN115825104 A CN 115825104A CN 202211324836 A CN202211324836 A CN 202211324836A CN 115825104 A CN115825104 A CN 115825104A
- Authority
- CN
- China
- Prior art keywords
- point cloud
- data
- workpiece
- defect detection
- dimensional 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
Links
Images
Landscapes
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention discloses a soldering tin defect detection method based on integration volume calculation, which comprises the following steps: a PLC in the production line sends out a detection demand, controls a three-dimensional point cloud line scanning camera to scan the PCB, and obtains and stores three-dimensional point cloud source data; denoising and filtering the point cloud data; screening the interested region to obtain a soldering tin region to be calculated; carrying out fine positioning on the region of interest; carrying out integral conversion on the volume calculation of the final region, and converting into summation; and judging whether the PCB has defects or not according to whether the volume value is in the threshold value or not. The invention also discloses a defect detection system for realizing the method, which comprises a three-dimensional point cloud line scanning device, a computer and a production line PLC. Compared with the prior art, the method has the advantages of greatly shortening the production period, automatically assisting manual judgment, saving the capital investment, improving the economic benefit and the like.
Description
Technical Field
The invention belongs to the field of three-dimensional image processing and computer vision, and relates to a soldering tin defect detection method and system based on integral volume calculation.
Background
In the production of electronic information components such as automobile parts, the PCB is often required to be subjected to soldering tin processing, but in the actual production process, the condition of soldering tin error also can occur in the automatic soldering tin processing, namely, the tin quantity is incorrect or the interference of external factors causes the soldering tin to be in the lower part of the PCB and the upper part of the electrode plate, in an interlayer between the two layers, the soldering tin is excessive, the short circuit is caused, and the defect of the soldering tin brings new challenges to the high-quality production of small parts.
Since defects are small, it is often necessary to manually power-up each board for such defect detection, which causes significant delays in production cycle and product delivery, and at the same time, the cost of manual resources is significant.
However, since the defects are hidden in a gap between the PCB and the workpiece, the amount of light entering is small, and imaging cannot be performed, and even if imaging is difficult, data distortion is caused by too small amount of light entering, and thus the solder defects cannot be detected well by using the light source detection and image recognition method.
Disclosure of Invention
In order to solve the defects in the prior art, the invention aims to provide a soldering tin defect detection method for calculating volume based on integral and a system for realizing the defect detection method aiming at the current data, and the method can be used for producing and manufacturing parts in the automobile industry. The invention solves the problem that the defect can not be detected by scanning the three-dimensional point cloud, calculating the volume through the imageable data and reversely deducing the defect volume by using the known data.
The purpose of the invention can be realized by the following technical scheme:
the invention provides a soldering tin defect detection system for solving volume based on integral, which can be used for producing small parts in the automobile industry, and comprises a three-dimensional point cloud line scanning device, as shown in figure 2, a conventionally used computer and a corresponding production line PLC (not shown in the figure); the three-dimensional point cloud line scanning device comprises technological equipment, a three-dimensional point cloud line scanning camera, a Programmable Logic Controller (PLC) and a computer, wherein the PLC and the computer are used for controlling the whole soldering tin defect detection system; the workpiece to be detected is arranged on a workpiece table of the process equipment in the three-dimensional point cloud line scanning device.
The three-dimensional point cloud line scanning camera is a camera purchased by a third party and used for scanning a PCB (printed circuit board) in a line mode to acquire height data and brightness data of a workpiece.
The computer is used for deploying the algorithm, acquiring camera data and communicating with the three-dimensional point cloud line scanning camera and the production line PLC.
The PLC can send out the detection requirement for the defects of the PCB.
The invention also provides a soldering tin defect detection method based on integral volume calculation by applying the system, which is used as the final link in the whole set of production flow, can judge whether the PCB short circuit caused by excessive soldering tin can occur or not, flows in the whole production line, is used for recording the information of the production state and transmitting the information to the staff on the production line;
the method comprises the following steps:
firstly, a Programmable Logic Controller (PLC) in a defect detection production line sends a PCB defect detection requirement, a three-dimensional point cloud line scanning camera is controlled to scan a workpiece needing to detect whether defects exist or not and the defect degree to obtain three-dimensional point cloud data of the workpiece, and the data are stored on a computer;
step two, filtering and denoising the three-dimensional point cloud data of the workpiece to be detected obtained in the step one;
thirdly, performing primary coarse positioning on the region of interest, namely the approximate region with the single-PIN PIN, on the three-dimensional point cloud data;
accurately positioning the region of interest on the three-dimensional point cloud data to obtain a region only having a single PIN PIN, wherein the bottom surface is the junction of the PIN and the PCB base surface;
step five, integrating to obtain the volume of the region of interest obtained in the step four;
and step six, judging the threshold value of the obtained volume of the region of interest, and judging whether soldering tin of the PCB has defects.
In the first step, the controlling of the three-dimensional point cloud line scanning camera refers to running a camera control program for secondary development based on a Dynamic Link Library (DLL) for secondary development given by a third-party camera and document data on a Ubuntu operating system, and the camera control program has functions of controlling camera scanning, camera memory emptying and camera data storage in multiple formats.
The three-dimensional point cloud data of the workpiece obtained in the first step is npy file obtained by controlling a three-dimensional point cloud line scanning camera, and comprises three-dimensional point cloud data of front imaging of a PCB workpiece and three-dimensional point cloud data of a scanning side obtained by rotating the workpiece, wherein the three-dimensional point cloud data comprises data of X, Y, Z three coordinates.
And in the second step, the filtering and denoising are carried out through a numerical filtering invalid point algorithm and a discrete point removing algorithm based on a statistical radius, and the filtering and denoising are used for removing noise point cloud data generated by light or the camera in the data.
Specifically, the numerical filtering invalid point algorithm means that when a three-dimensional camera scans, in places where light cannot be reflected on a workpiece to be measured, a plurality of invalid data can be generated in the areas, the absolute value of the Z-direction coordinate values of the points is extremely large, compared with normal data, the Z-direction coordinate data are marked as infinity by the camera, the data are marked as 99.9999 or-99.9999, and the data are removed uniformly by judging the size; the statistical radius-based discrete point removal algorithm is that the algorithm draws a sphere with a radius of a preset value for each point in the point cloud, and if the number of other points in the sphere is smaller than a preset point cloud number threshold nb _ points, the algorithm judges the point as a special case and deletes the special case. The larger the radius is given, the larger the number nb _ points of the point clouds is, the fewer the points are removed, the radius is given to be 0.025m generally according to the empirical setting, and the threshold of the number of the point clouds is 20. The discrete point removal algorithm of radius is used to remove the points scattered at the periphery due to the farther distance from the body.
Through the processing, the data become more centralized and more accurate.
In the third step, the preliminary rough positioning is mainly to perform positioning through the horizontal, vertical and height coordinates of the point cloud data to obtain the position positioning of the horizontal and vertical surfaces, so as to obtain the approximate position of each soldering PIN PIN.
The method comprises the steps of performing preliminary rough positioning, determining an interested area, corresponding the coordinate of the three-dimensional point cloud and the coordinate of an actual workpiece design drawing, and positioning by judging the approximate position of the interested area in data.
In the fourth step, the fine positioning refers to finding the reference surface of the PCB by a cutting type jump surface query mode in the invention, and in the subsequent volume solving process, the height data of the reference surface is the bottom surface of the irregular PIN PIN, so that the roughly positioned region of interest is converted into a soldering tin region only containing the PIN PIN, and the fine positioning data is point cloud soldering tin data only containing the PIN PIN reference surface and being higher than the reference surface.
The cutting type jump surface query means that in the vertical direction, the point cloud number of the current height is continuously calculated in an interval set according to the actual situation, then the point cloud number of the current height is subtracted by the last calculated point cloud number to obtain a point cloud number difference, and when the point cloud number difference suddenly jumps and is 100 times different from the previous point cloud number difference, the jump plane is the reference plane of the PCB.
In the fifth step, the volume is obtained by integration, because the interval of the point cloud data is only 12 micrometers and is close to 0, as shown in fig. 10, the left point cloud data and the right point cloud data are respectively point cloud data, according to the idea of integration, the point cloud data above a PCB reference surface can be considered to be approximate to a cuboid, the center of the bottom surface of the cuboid is X, Y coordinates of the point cloud, and because the distance between the left side and the right side is 12 micrometers, the bottom surface of the cuboid is a square with the side length of 12 micrometers, the height of the point cloud is converted into the height of the cuboid, the final PCB is formed by the small cuboid, and therefore the process of obtaining the volume by integration is converted into the operation of summing the heights of the point clouds.
Step six, judging the threshold value according to a given experience threshold value and the total soldering tin amount, wherein the experience threshold value is the volume calculated by a large number of defect-free workpiece samples, and when the detected experience threshold value is exceeded, the detected defect workpiece is a defect workpiece, and when the detected defect workpiece is not exceeded, the detected defect workpiece is a normal workpiece, and judging is carried out; and after the judgment is finished, communicating with a production line PLC in a memory DB block writing mode.
Compared with the prior art, the invention has the following advantages:
1. the beat on the production line is accelerated, more workpieces can be produced in unit time, and more energy cost is saved under the condition of producing the same workpieces. The stored data is changed into a binary storage NPY format from the traditional three-dimensional data formats of CSV, PLY and PCD, the storage time is changed from 3 seconds to 1 second, the reading time is changed from 1 second to 0.4 second, and sixty percent of reduction is realized. The whole detection time is saved from 50.5 seconds to 45.3 seconds.
2. The machine vision based on the three-dimensional point cloud replaces the manual operation to detect the defects, and the labor cost is saved.
3. The software and hardware integrated solution is provided, the three-dimensional point cloud camera and the detection computer are arranged on the hardware, meanwhile, the control software and the detection software of the camera and the software for communicating with the PLC are designed on the software, and the software integrated solution can also be used for migrating to other related requirements of production lines. The hardware adopts a camera of a third party, but the software used by the invention is the related software which is originally developed and is adapted to the invention.
4. The soldering tin defect detection system for calculating the volume based on the integral has economic benefits and makes positive contribution to the national clean production and energy conservation and emission reduction strategies.
Drawings
Fig. 1 is a flow chart of a solder defect detection algorithm based on integration to determine volume according to the present invention.
FIG. 2 is a hardware diagram of a part of the system of the present invention.
Fig. 3 is a UML deployment diagram of a solder defect detection algorithm operating system based on integration volume. The main requests are sent through a front-end browser, and the algorithm is divided into three parts, namely memory reading and writing of a memory DB block, and is used for communication of a PLC and reading and writing of multiple types; the second part is a related program of camera control, which is used for controlling the number of the camera program and acquiring height data and brightness data; the third part is a detection algorithm, and the defect detection method of using integration to calculate the volume is adopted.
FIG. 4 is a three-dimensional rendering of a PCB scanned from the front of a camera, with a PIN circled in the box.
FIG. 5 is a three-dimensional rendering of a PCB scanned from the side of a camera, with a PIN circled in the box.
FIG. 6 is a coarsely positioned three-dimensional rendering of the PIN region.
FIG. 7 is a pinpoint three-dimensional rendering of a PIN region.
FIG. 8 is a representation of a three-dimensional view of a rotationally scanned workpiece, where P is 1 、P 2 Two adjacent point cloud data are respectively two point cloud data with the same horizontal coordinate when the camera scans back and forth, and X, Y and Z represent three-dimensional coordinates of the center of a rotating circle.
FIG. 9 is a coordinate demonstration of a rotational scan, P 1 、P 2 Same as FIG. 8, P 2 ' is P 1 P 2 The point cloud data surface R after the connecting line is converted into a curve is P 1 Distance to the center of the rotation center, distance dy between the front and rear scanning surfaces of the camera, and theta is P 1 P 2 The angle of the angle.
FIG. 10 is a schematic diagram of the calculation of the height of the point cloud during the integration and volume calculation process of the present invention.
Detailed Description
The invention is further described in detail with reference to the following specific examples and the accompanying drawings. The procedures, conditions, experimental methods and the like for carrying out the present invention are general knowledge and common general knowledge in the art except for the contents specifically mentioned below, and the present invention is not particularly limited.
The invention provides a soldering tin defect detection method based on integration volume calculation, which comprises the following steps: a PLC in the production line sends out a detection requirement, controls a three-dimensional point cloud line scanning camera to scan the PCB, and obtains and stores three-dimensional point cloud source data; denoising and filtering the point cloud data; screening the interested region to obtain a soldering tin interval which needs to be calculated; carrying out fine positioning on the region of interest; carrying out integral conversion on the volume calculation of the final region, and converting into summation; and judging whether the PCB has defects or not according to whether the volume value is in the threshold value or not. The invention also discloses a defect detection system for realizing the method, which comprises a three-dimensional point cloud line scanning device, a computer and a production line PLC. The system of the invention automatically positions the interested area and obtains the result of the volume by controlling the camera to scan the front and the side of the PCB, and the result is communicated with the PLC in a DB block modification mode. Compared with the prior art, the method has the advantages of greatly shortening the production period, automatically assisting manual judgment, saving the capital investment, improving the economic benefit and the like.
The invention designs a soldering tin defect detection system based on integral volume calculation according to the principles of the Internet of things and intelligent production. In the attached drawings, the rectangular frames in fig. 4 and 5 are marked with PIN PINs, the system of the invention carries out full-automatic detection on workpieces on a production line, and the workpieces are linked through three modules, namely a camera, a computer, a PLC and the like, wherein the read-write operation of the PLC is mainly used for information interaction, and through the series of processes, the defect detection of soldering tin is realized, the efficiency can be improved, and the production cost can be reduced.
The design scheme of the soldering tin defect detection system based on the integration volume calculation is as follows:
a special soldering tin defect detection system based on integration solving volume is designed, and the state of a workpiece in the production process is automatically detected by software.
Firstly, when a workpiece is transmitted to a module in charge of system software on a production line, an activation signal is triggered to a PLC, then the PLC can write a DB (memory) block and write 1 into a 40001 address of the DB block, and the system can continuously read the address at the beginning of detection to judge whether a detection requirement exists.
When a detection requirement exists, the system calls the camera module to scan data, so that front scanning data and side scanning data are obtained respectively, the data are stored in a binary npy format in a compressed manner, and after npy is rendered, the data are shown in fig. 4 and 5, wherein different gray scales represent different heights on a workpiece.
After data exist, the data are preprocessed by using the data preprocessing algorithm provided by the invention, firstly, noise reduction and filtering are carried out, dead spots which cannot be scanned by a three-dimensional camera are removed by a numerical filtering invalid point algorithm and a radius discrete point removing algorithm, and the data volume is reduced by carrying out micro down-sampling by voxel down-sampling.
Then the coarse positioning of the PIN is followed, with given PCB layout and data, in horizontal and vertical and depth direction, by zone definition, resulting in the PIN foot diagram as shown in fig. 6, where the raised part is the PIN foot.
After rough positioning, the final base plane is found by continuously cutting and judging the jump of the number of three-dimensional point clouds in the upper threshold range and the lower threshold range of a cutting plane through the cutting jump surface query method provided by the invention for the first time, and the rendering graph shown in the figure 7 is obtained.
Then, the volume is obtained through integration, and the volume is obtained according to a definition formula of double integration:
since the resolution of the three-dimensional point cloud camera is 12um, which is an infinite interval tending to 0 in a physical sense, the heights of discrete point cloud data points are summed by subtracting all depth coordinates and base planes and summing to obtain a volume, wherein x and y represent horizontal and vertical coordinates of a base plane, f (x, y) corresponds to a height coordinate under the coordinates, and σ represents the area of the base plane, and since the interval λ of the point cloud data is only 12 micrometers, i.e., λ is close to 0, the base plane can be divided into Δ σ of a square area element with the side length of 12 micrometers i For Δ σ i Each point of (1), his abscissa and ordinate(ξ i ,η i ) Corresponding height f (xi) i ,η i ) The height coordinate of each point cloud is obtained, so that the volume is converted into tiny cuboids, the bottom surface of each cuboid is a square with the side length of 12 micrometers, the height of each cuboid is the point cloud, and the process of integrating and solving the volume is converted into the operation of point cloud height summation.
In the side surface, due to the rotational scanning, as shown in fig. 8 and 9, all coordinates Z need to be transformed, the center of the rotation circle is used as the origin, and the center point P of the outer surface is used as the center point P 1 As starting points, other points are based on P 1 And performing approximate conversion. The distance from the center of the rotating circle can be calculated by coordinates to obtain R and P 2 Is the next point swept by the camera, and the distance dy is fixed, 12.5, which is the pitch of the camera scan. P 2 ' is the actual point, i.e. the point to be calculated is deduced to yield P 2 ' Z coordinate:
and finally, judging whether the volume is in the defect judgment area or not through a given threshold value, thereby carrying out defect judgment. And communicates by means of writing a DB block, and if defective, then at the address 4002 of the DB block, writing 111, indicating that 3 PINs are all defective, and writing 0 if normal, e.g. 101, indicating that the second PIN is normal and the first third is defective. And writes 4003 address to 1, indicating that the test has been completed and returns the result. And then the PLC continuously reads the data to judge whether a detection result is obtained or not, and when the address of 4003 is read to be 1, the next workpiece is entered.
Examples
For example, when the workpiece 1 enters the production line, an activation signal is triggered first to initialize all the points of the DB block, then the PLC causes the DB4001 to write as 1, the detection computer continuously reads the signal of the position, when 1 is detected, the camera is controlled to scan, the camera scans the front data and the side data of the workpiece respectively, after the scanning is completed, the detection computer performs calculation on the front data and the side data to obtain volumes 17, 16, 17 of the front three PINs, the volumes of the side three PINs are 7,8,7, the volumes of the final three PINs are 24, 24, 24, then the total soldering tin amount is 28, the volumes scanned by the PINs are subtracted from the total soldering tin amount, namely the soldering tin of the side and the back which cannot be scanned are 4,4,4, a given threshold value is 3.5-5, and the PINs are within the range, so that the three PINs are normal, 000 is written into the 4002 address, and then 1 is written into the 4003 address. PLC reads 4003 and is 1 back, can control to produce the line and get into next and produce the line.
The protection content of the present invention is not limited to the above embodiments. Variations and advantages that may occur to those skilled in the art may be incorporated into the invention without departing from the spirit and scope of the inventive concept, and the scope of the appended claims is intended to be protected.
Claims (10)
1. A soldering tin defect detection method for solving volume based on integration is characterized by comprising the following steps:
step one, a Programmable Logic Controller (PLC) in a defect detection production line sends a PCB defect detection requirement, a three-dimensional point cloud line scanning camera is controlled to scan a workpiece needing to detect whether defects exist or not and the defect degree to obtain three-dimensional point cloud data of the workpiece, and the data are stored in a computer;
step two, filtering and denoising the three-dimensional point cloud data of the workpiece to be detected obtained in the step one;
step three, performing primary coarse positioning on the region of interest on the three-dimensional point cloud data; the interested area is an area with a single PIN PIN;
step four, carrying out fine positioning on the region of interest on the three-dimensional point cloud data;
step five, integrating to obtain the volume of the region of interest obtained in the step four;
and step six, carrying out threshold judgment on the obtained volume of the region of interest, and judging whether soldering tin of the PCB has defects.
2. The defect detection method of claim 1, wherein in the first step, the three-dimensional point cloud data is npy file acquired by controlling a three-dimensional point cloud line scanning camera, and comprises three-dimensional point cloud data for imaging the front surface of a PCB workpiece and scanned side three-dimensional point cloud data obtained by rotating the workpiece, and comprises data of X, Y, Z three coordinates.
3. The defect detection method of claim 1, wherein in the first step, the controlling the three-dimensional point cloud line scanning camera refers to running a camera control program for secondary development based on a Dynamic Link Library (DLL) for secondary development given by a third-party camera and document data on a Ubuntu operating system, and the functions including controlling camera scanning, camera memory clearing and camera data storage in multiple formats are realized.
4. The defect detection method of claim 1, wherein in the second step, the filtering and denoising employs a data filtering invalid point algorithm based on three-dimensional point cloud data and a discrete point removing algorithm based on statistical radius to remove noise point cloud data generated by light or camera.
5. The defect detection method of claim 4, wherein the data filtering invalid point algorithm is implemented by judging the magnitude of the Z-direction coordinate value to uniformly remove data with the maximum Z-direction coordinate generated in a corresponding area due to the fact that light rays in a part of the area on the workpiece to be detected cannot be reflected; the discrete point removing algorithm is that each point in the point cloud is drawn with a sphere with a preset radius, and if the number of other points in the sphere is smaller than a preset point cloud number threshold value, the point is judged as a special case and deleted.
6. The defect detection method of claim 1, wherein in step three, the preliminary coarse positioning can obtain a rough area with a single PIN; the preliminary rough positioning is based on a design drawing of a PCB workpiece and point cloud data, and the position positioning of a horizontal surface and a vertical surface is carried out through the horizontal coordinates, the vertical coordinates and the height coordinates of the point cloud data.
7. The defect detection method of claim 1, wherein in the fourth step, the fine positioning refers to finding a reference surface of the PCB by a cutting type jump surface query, and in a subsequent volume solving process, height data of the reference surface, namely, a bottom surface of an irregular PIN, converts a region of interest of the coarse positioning into a soldering tin region only containing the PIN, and the data of the fine positioning refers to point cloud data only containing the PIN reference surface and soldering tin higher than the reference surface;
the cutting type jump surface query means that in the vertical direction, the point cloud number of the current height is continuously calculated in an interval set according to the actual situation, then the point cloud number of the current height is subtracted by the last calculated point cloud number to obtain a point cloud number difference, and when the point cloud number difference suddenly jumps and is 100 times different from the previous point cloud number difference, the jump plane is the reference plane of the PCB.
8. The defect detection method of claim 1, wherein in the fifth step, the volume calculation is performed by converting the process of integrating and calculating the volume into the sum of the heights of the point clouds based on the characteristics of the point cloud data.
9. The defect detection method according to claim 1, wherein in the sixth step, the threshold judgment is to judge whether the workpiece is defective or not based on the total amount of solder and a threshold given by empirical knowledge; the threshold is the volume calculated by a large number of defect-free workpiece samples, and when the detected volume exceeds the threshold, the workpiece is a defect workpiece, otherwise, the workpiece is a normal workpiece; and after the judgment is finished, communicating with a production line PLC in a memory DB block writing mode.
10. A defect detection system for implementing the defect detection method according to any one of claims 1 to 9, wherein the defect detection system comprises: the device comprises a three-dimensional point cloud line scanning device, a computer and a production line PLC;
the three-dimensional point cloud line scanning device comprises process equipment and a three-dimensional point cloud line scanning camera; the process equipment is provided with a workpiece table for mounting a workpiece to be detected; the three-dimensional point cloud line scanning camera is used for line scanning the PCB to acquire height data and brightness data of the workpiece;
the computer is used for deploying an algorithm used in the detection method, acquiring camera data and communicating with the three-dimensional point cloud line scanning camera and the production line PLC;
the production line PLC can send out the detection demand to the PCB defect.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211324836.8A CN115825104A (en) | 2022-10-27 | 2022-10-27 | Soldering tin defect detection method and system based on integration volume calculation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211324836.8A CN115825104A (en) | 2022-10-27 | 2022-10-27 | Soldering tin defect detection method and system based on integration volume calculation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115825104A true CN115825104A (en) | 2023-03-21 |
Family
ID=85525545
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211324836.8A Pending CN115825104A (en) | 2022-10-27 | 2022-10-27 | Soldering tin defect detection method and system based on integration volume calculation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115825104A (en) |
-
2022
- 2022-10-27 CN CN202211324836.8A patent/CN115825104A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102496161B (en) | Method for extracting contour of image of printed circuit board (PCB) | |
| CN111982921B (en) | Method and device for detecting hole defects, conveying platform and storage medium | |
| CN110503638B (en) | Spiral adhesive quality online detection method | |
| CN112767399B (en) | Semiconductor bonding wire defect detection method, electronic device and storage medium | |
| JPH07260701A (en) | Recognition method of area of inspection | |
| JP5123244B2 (en) | Shape defect inspection device, shape modeling device, and shape defect inspection program | |
| CN108802051B (en) | System and method for detecting bubble and crease defects of linear circuit of flexible IC substrate | |
| CN112102254A (en) | Wood surface defect detection method and system based on machine vision | |
| CN113222955A (en) | Gear size parameter automatic measurement method based on machine vision | |
| JP3414967B2 (en) | Bump inspection method | |
| JP3009205B2 (en) | Inspection method and apparatus | |
| CN113705564A (en) | Pointer type instrument identification reading method | |
| CN115825104A (en) | Soldering tin defect detection method and system based on integration volume calculation | |
| CN112419274A (en) | Solder paste detection method, system, electronic device and medium | |
| WO2000028309A1 (en) | Method for inspecting inferiority in shape | |
| JP4814116B2 (en) | Mounting board appearance inspection method | |
| CN114965272A (en) | Testing method of chip defect detection platform | |
| CN114942246A (en) | Method and equipment for detecting welding quality of MiniLED based on 3D confocal sensor | |
| JP2924859B2 (en) | Appearance inspection method and device | |
| CN113379678A (en) | Circuit board detection method and device, electronic equipment and storage medium | |
| CN112697813A (en) | AOI special scanning operation method | |
| JP2970855B2 (en) | Inspection method for semiconductor memory device | |
| CN112184637A (en) | 3D prints forming part defect and assesses device on line | |
| CN113450331B (en) | Special-shaped component pin detection method | |
| JPH06201603A (en) | Method for generating inspection data in missing inspection of electronic component and method for detecting missing of the component |
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 |