CN207937374U - A kind of pad solder defect detecting device - Google Patents
A kind of pad solder defect detecting device Download PDFInfo
- Publication number
- CN207937374U CN207937374U CN201820256245.4U CN201820256245U CN207937374U CN 207937374 U CN207937374 U CN 207937374U CN 201820256245 U CN201820256245 U CN 201820256245U CN 207937374 U CN207937374 U CN 207937374U
- Authority
- CN
- China
- Prior art keywords
- weldment
- detected
- data
- camera
- connect
- 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
Landscapes
- Image Processing (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The utility model discloses a kind of pad solder defect detecting devices, including：Light source is set to above pad solder part to be detected, for providing light；Data acquisition components, for acquiring weldment apparent condition image data to be detected and transmitting；Whether data processing and control assembly, connect with light source and data acquisition components respectively, for handling appearance status image data, obtain qualified judging result and storing and display, be additionally operable to control light source and data acquisition components operation respectively；Adjusting and sorter assemblies, it is connect respectively with data acquisition components and data processing and control assembly, for adjusting the position of data acquisition components to acquire apparent condition image data, it is additionally operable to, when judging result is unqualified, sort unqualified weldment under data processing and the control of control assembly.Pad solder defect detecting device provided by the utility model can improve detection efficiency, accuracy of detection, and production quality is effectively ensured.
Description
Technical field
The utility model is related to industrial vision detection technique fields, and in particular to a kind of pad solder defect detecting device.
Background technology
With the development of electronics industry, especially in precise electronic spare and accessory parts, the small spare and accessory parts production process of volume, weldering
Disk welding quality will directly affect entire product quality.
It is directed to pad solder quality testing in small spare and accessory parts production process at present, it is single to weld since pad volume is small
Disk size is within 4mm, and artificial detection needs to rely on after electron microscope is amplified and visually recognizes, and detection efficiency is low, main
The property seen is strong, it is difficult to ensure detection quality.
Therefore it provides a kind of pad solder defect detecting device, being detected by way of machine vision can be significantly
Detection efficiency, accuracy of detection are improved, cost of labor is reduced, meets modern science production requirement, control production quality is existing
There is technology urgent problem to be solved.
Utility model content
The utility model needs to rely on after electron microscope is amplified for artificial detection in the prior art and visually distinguishes
Know, detection efficiency is low, subjectivity is strong, it is difficult to which the problem of ensureing detection quality provides a kind of pad solder defect detecting device, lead to
The mode for crossing machine vision is detected and can greatly improve detection efficiency, accuracy of detection, reduces cost of labor, meets modern section
Learn production requirement, control production quality.
The utility model provides a kind of pad solder defect detecting device, including:
Light source is set to above pad solder part to be detected, for providing light；
Data acquisition components, for acquiring weldment apparent condition image to be detected under the light that the light source provides
Data are simultaneously transmitted；
Data processing and control assembly are connect with the light source and data acquisition components respectively, for the appearance
Status image data are handled, and are obtained the whether qualified judging result of the weldment and are stored and display, are additionally operable to point
The light source and data acquisition components operation are not controlled；
Adjusting and sorter assemblies, connect respectively with the data acquisition components and data processing and control assembly, for adjusting
The position of the data acquisition components is saved to acquire the apparent condition image data, is additionally operable to when the judging result be not conform to
When lattice, unqualified weldment is sorted under the data processing and the control of control assembly.
Preferably, the data acquisition components include：
Camera lens, for being amplified to weldment apparent condition image to be detected；
Camera is set to side of the camera lens far from the weldment, for acquiring weldment appearance shape to be detected
State image data is simultaneously transmitted.
Preferably, the data processing and control assembly include：
Data processor is connect with the camera, for handling the apparent condition image data, to obtain
State the whether qualified judging result of weldment；
Controller is connect with the data processor, light source, data acquisition components and adjusting and sorter assemblies respectively,
For controlling the data processor, light source and data acquisition components and adjusting and sorter assemblies operation respectively.
Preferably, the adjusting and sorter assemblies include：
Adjuster is connect with the camera and the controller respectively, for being adjusted under the control of the controller
The position of the camera is to acquire the apparent condition image data；
Sorter is connect with the controller, is used for when the judging result is unqualified, in the control of the controller
Unqualified weldment is sorted under system.
Preferably, the data processor is treated journey to the apparent condition image data and includes：Primary Stage Data
Pretreatment, off-line training and on-line checking.
Preferably, the Primary Stage Data preprocessing process is specially：To manually delimiting acquisition interesting image regions, eliminate
Part background influence, while pad locations positioning is carried out to the region of interest area image of acquisition using algorithm.
Preferably, the off-line training process is specially：Before being detected to weldment to be detected, a large amount of welding are acquired
Qualified image is as positive sample, and the image with pad defect is as negative sample, by setting training parameter, to positive negative sample
It is trained on the basis of general network model, obtains training network model.
Preferably, the on-line checking process is specially：Pass through the network model pair during the calling off-line training
The image of acquisition detects in real time, obtains the whether qualified judging result of the weldment.
Preferably, the data processing and control assembly further include memory, are connect with the data processor, for depositing
Storage treated apparent condition image data, each weldment to be detected whether qualified judging result, and storage is a large amount of
Weld the negative sample of the positive sample of qualified image and the image with pad defect.
Preferably, the data processing and control assembly further include display, are connect with the data processor, for showing
Show the whether qualified judging result of each weldment to be detected.
One or more technical solutions that the utility model provides, have at least the following technical effects or advantages：It is logical
The mode for crossing machine vision is detected and can improve detection efficiency, accuracy of detection, reduces cost of labor, and product life is effectively ensured
Yield and quality.
Description of the drawings
It is required in being described below to embodiment in order to illustrate more clearly of the technical scheme in the embodiment of the utility model
Attached drawing to be used is briefly described, it should be apparent that, the accompanying drawings in the following description is only some realities of the utility model
Example is applied, it for those of ordinary skill in the art, without creative efforts, can also be according to these attached drawings
Obtain other attached drawings.
Fig. 1 is the pad solder defect detecting device structure chart that the utility model embodiment provides.
Specific implementation mode
In order to solve pad solder defect detecting device existing in the prior art need to rely on electron microscope carry out
It is visually recognized after amplification, detection efficiency is low, subjectivity is strong, it is difficult to which it is scarce to provide a kind of pad solder for the problem of ensureing detection quality
Detection device is fallen into, concrete thought is as follows：Including light source, it is set to above pad solder part to be detected, for providing light；Number
According to acquisition component, for acquiring weldment apparent condition image data to be detected under the light that light source provides and transmitting；Number
According to processing and control assembly, it is connect respectively with light source and data acquisition components, to appearance status image data
Reason obtains the whether qualified judging result of weldment and stores and show, is additionally operable to control light source and data acquisition group respectively
Part is run；Adjusting and sorter assemblies, connect respectively with data acquisition components and data processing and control assembly, for adjusting data
The position of acquisition component is additionally operable to acquiring apparent condition image data when judging result is unqualified, in data processing and
Unqualified weldment is sorted under the control of control assembly.Pad solder defect detecting device energy provided by the utility model
Detection efficiency, accuracy of detection are enough improved, cost of labor is reduced, production quality is effectively ensured.
As shown in Figure 1, the utility model provides a kind of pad solder defect detecting device, a kind of pad solder defect inspection
Device is surveyed, including:Light source 1 is set to above pad solder part to be detected, for providing light, to ensure Image Acquisition ring
For border light source homogeneous constant without shadow, the imaging of pad appearance state is apparent；Data acquisition components 2, the light for being provided in the light source 1
Weldment apparent condition image data to be detected is acquired under line and is transmitted；Data processing and control assembly 3, respectively with the light
Source 1 and data acquisition components 2 connect, and for handling the apparent condition image data, obtaining the weldment is
The judging result of no qualification is simultaneously stored and is shown, is additionally operable to control the light source 1 respectively and data acquisition components 2 are run；It adjusts
Section and sorter assemblies 4, connect respectively with the data acquisition components 2 and data processing and control assembly 3, for adjusting the number
According to the position of acquisition component 2 to acquire the apparent condition image data, it is additionally operable to when the judging result is unqualified,
Unqualified weldment is sorted under the data processing and the control of control assembly 3.Wherein data processing and control assembly 3
It is connected by input/output cards using serial communication mode between adjusting and sorter assemblies 4.
In conjunction with shown in Fig. 1, the data acquisition components 2 include：Camera lens 21, for weldment apparent condition to be detected
Image is amplified, and camera lens 21 is telecentric lens, ensures also obtain ideal Image Acquisition in the case that single pad is smaller
Effect；Camera 22 is set to side of the camera lens 21 far from the weldment, for acquiring weldment appearance shape to be detected
State image data is simultaneously transmitted, and in the present embodiment, camera 22 is the high-resolution face battle array industrial camera with global exposure mode.
Specifically, the data processing and control assembly 3 include：Data processor 31 connect with the camera 22, is used for
The apparent condition image data is handled, to obtain the whether qualified judging result of the weldment；Controller 32, point
It is not connect with the data processor 31, light source 1, data acquisition components 2 and adjusting and sorter assemblies 4, for controlling respectively
The data processor 31, light source 1 and data acquisition components 2 and adjusting and sorter assemblies 4 are run.In the present embodiment,
Camera 22 is connected with data processor 31 by USB3.0.
Specifically, the adjusting and sorter assemblies 4 include：Adjuster 41, respectively with the camera 22 and the control
Device 32 connects, for adjusting the position of the camera 22 under the control of the controller 32 to acquire the apparent condition image
Data, in the present embodiment, adjuster 41 are manipulator, can carry the movement of camera 22 and capture；Sorter 42, with the control
Device 32 connect, for when the judging result be it is unqualified when, under the control of the controller 32 to unqualified weldment into
Row sorting.
In the present embodiment, the data processor 31 is treated journey to the apparent condition image data and includes：
Primary Stage Data pretreatment, off-line training and on-line checking.The Primary Stage Data preprocessing process is specially：To manually delimiting acquisition
Interesting image regions eliminate part background influence, while carrying out pad position to the region of interest area image of acquisition using algorithm
Set positioning.Off-line training process is specially：Before being detected to weldment to be detected, acquires the qualified image of a large amount of welding and make
For positive sample, the image with pad defect is as negative sample, by setting training parameter, to positive negative sample in universal network mould
It is trained on the basis of type, obtains training network model.The on-line checking process is specially：By calling the off-line training
Network model in the process detects the image of acquisition in real time, obtains the whether qualified judging result of the weldment.
Specifically, the data processing and control assembly 3 further include memory 33, are connect with the data processor 31,
For the apparent condition image data that stores that treated, each weldment to be detected whether qualified judging result, Yi Jicun
The negative sample of the positive sample of a large amount of qualified images of welding of storage and the image with pad defect；Display 34, with the data
Processor 31 connects, the judging result whether qualified for showing each weldment to be detected.
In conclusion application scheme at least has following advantageous effects compared with the existing technology：
1) detection efficiency, accuracy of detection can be improved by being detected by way of machine vision；
2) cost of labor is reduced, production quality is effectively ensured.
Above disclosed is only a kind of the utility model preferred embodiment, cannot limit this practicality with this certainly
Novel interest field, those skilled in the art can understand all or part of the processes for realizing the above embodiment, and according to
Equivalent variations made by the utility model claims, still fall within the scope of the utility model.
Claims (3)
1. a kind of pad solder defect detecting device, which is characterized in that including:
Light source (1) is set to above pad solder part to be detected, for providing light；
Data acquisition components (2) include the telecentric lens for being amplified to weldment apparent condition image to be detected
(21), it is set to side of the camera lens (21) far from the weldment, for acquiring weldment apparent condition figure to be detected
As data and the camera of transmission (22), the camera (22) is the high-resolution face battle array industrial camera with global exposure mode；
The data processor (31) being connect with the camera (22) by USB3.0, and respectively with the data processor
(31), the controller (32) of camera (22) connection of light source (1), data acquisition components (2)；
It adjusts and sorter assemblies (4) includes：The adjuster being connect respectively with the camera (22) and the controller (32)
(41), the sorter (42) being connect with the controller (32), adjuster (41) are manipulator；
The controller adjusts the position of the camera (22) for control machinery hand, weldment appearance to be detected is captured in movement
Status image data, and control the sorter (42) and sort the weldment detected.
2. pad solder defect detecting device according to claim 1, which is characterized in that the data processing and control group
Part (3) further includes memory (33), is connect with the data processor (31), for storing treated apparent condition picture number
According to, each weldment to be detected whether qualified judging result, and the positive samples of a large amount of qualified images of welding of storage and
The negative sample of image with pad defect.
3. pad solder defect detecting device according to claim 1, which is characterized in that the data processing and control group
Part (3) further includes display (34), is connect with the data processor (31), for whether showing each weldment to be detected
Qualified judging result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201820256245.4U CN207937374U (en) | 2018-02-09 | 2018-02-09 | A kind of pad solder defect detecting device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201820256245.4U CN207937374U (en) | 2018-02-09 | 2018-02-09 | A kind of pad solder defect detecting device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN207937374U true CN207937374U (en) | 2018-10-02 |
Family
ID=63652697
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201820256245.4U Active CN207937374U (en) | 2018-02-09 | 2018-02-09 | A kind of pad solder defect detecting device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN207937374U (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109623143A (en) * | 2019-01-11 | 2019-04-16 | 东莞塔菲尔新能源科技有限公司 | A kind of on-line monitoring device and monitoring method for battery laser welding |
| CN110530884A (en) * | 2019-10-09 | 2019-12-03 | 易思维（杭州）科技有限公司 | Laser Welding protects eyeglass defect detecting device |
-
2018
- 2018-02-09 CN CN201820256245.4U patent/CN207937374U/en active Active
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109623143A (en) * | 2019-01-11 | 2019-04-16 | 东莞塔菲尔新能源科技有限公司 | A kind of on-line monitoring device and monitoring method for battery laser welding |
| CN110530884A (en) * | 2019-10-09 | 2019-12-03 | 易思维（杭州）科技有限公司 | Laser Welding protects eyeglass defect detecting device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR102102874B1 (en) | Automatic optical inspection system and operating method thereof | |
| CN110659660B (en) | Automatic optical detection classification equipment using deep learning system and training equipment thereof | |
| CN205317678U (en) | Metal product printing defect on -line measuring system based on machine vision | |
| CN101539533A (en) | Automatic detecting device and method for battery internal defect | |
| CN207937374U (en) | A kind of pad solder defect detecting device | |
| CN107271447A (en) | A kind of PLC outward appearances and LED status detection means based on machine vision | |
| CN108480239A (en) | Workpiece quick sorting method based on stereoscopic vision and device | |
| CN107024478A (en) | A kind of part processes automatic defect identifying system | |
| CN111079831A (en) | Intelligent optical detection sample characteristic and flaw automatic marking method and device | |
| CN110560376A (en) | Product surface defect detection method and device | |
| CN104833925A (en) | Fuel cell bipolar plate detection method and system based on machine vision | |
| CN202066794U (en) | Full-automatic key detection device | |
| CN100562132C (en) | The assembling of numerical camera mould and test macro and method | |
| CN104655659A (en) | X-ray detection system for PCB production | |
| CN116330319B (en) | State monitoring system and method for carrying manipulator | |
| CN108132256A (en) | A kind of welding quality machine vision high-precision detecting method | |
| CN107907546A (en) | A kind of defects of vision detection method and system based on intelligent mobile terminal | |
| CN109894763B (en) | Solar module junction box positioning and assembling system and positioning and assembling method thereof | |
| CN207198050U (en) | A kind of PLC outward appearances and LED status detection means based on machine vision | |
| CN105548210A (en) | Detection method for American-standard power source extension wire based on machine vision | |
| CN214375120U (en) | PCB board detection mechanism | |
| CN107096727A (en) | A kind of intelligent machining system | |
| CN208026642U (en) | A kind of pad of engine defect detecting device | |
| CN208623750U (en) | Intelligent industrial camera | |
| CN104324893A (en) | Automatic optical fiber end face detection system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| PP01 | Preservation of patent right | ||
| PP01 | Preservation of patent right |
Effective date of registration: 20201019 Granted publication date: 20181002 |
|
| PD01 | Discharge of preservation of patent | ||
| PD01 | Discharge of preservation of patent |
Date of cancellation: 20201210 Granted publication date: 20181002 |