CN111359912A - SMT production line intelligent detection system and method based on machine vision - Google Patents
SMT production line intelligent detection system and method based on machine vision Download PDFInfo
- Publication number
- CN111359912A CN111359912A CN202010298962.5A CN202010298962A CN111359912A CN 111359912 A CN111359912 A CN 111359912A CN 202010298962 A CN202010298962 A CN 202010298962A CN 111359912 A CN111359912 A CN 111359912A
- Authority
- CN
- China
- Prior art keywords
- detection
- pcb
- machine vision
- test
- production line
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/34—Sorting according to other particular properties
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/02—Measures preceding sorting, e.g. arranging articles in a stream orientating
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/36—Sorting apparatus characterised by the means used for distribution
- B07C5/361—Processing or control devices therefor, e.g. escort memory
- B07C5/362—Separating or distributor mechanisms
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/36—Sorting apparatus characterised by the means used for distribution
- B07C5/38—Collecting or arranging articles in groups
Landscapes
- General Factory Administration (AREA)
Abstract
An SMT production line intelligent detection system based on machine vision and a method thereof are composed of a cache unit, a control cabinet, a vision sensing unit, a panel turnover machine, a feeding machine, a vision detection device, six-axis robots, a host management and control system, a cloud service platform and the like, wherein a plurality of groups of vision detection devices and test systems are adopted to carry out multifunctional omnibearing test on PCB boards on the SMT production line, the personal safety in the test process is ensured, the machine vision intelligent detection of the PCB boards on the SMT production line is realized to replace manual detection, the detection accuracy and the detection efficiency of the PCB board assembly are improved, diversified test requirements can be met, the test system can be connected with a cloud end, the online release and receiving of test tasks through a mobile end are realized, various cloud services are provided for customers, the test data can be traced, the communication cost and the time cost are reduced, the invention adopts a modular design, the system has simple and clear structure and is convenient for related technicians to operate and maintain.
Description
Technical Field
The invention relates to a detection system, in particular to an SMT production line intelligent detection system and method based on machine vision, which can solve the problems of high detection difficulty, high labor cost, low detection efficiency and the like of an SMT production line by adopting technologies such as a machine vision algorithm, a sensor, an Internet of things, cloud service and the like.
Background
Surface Mount Technology (SMT) is a Technology in which Surface Mount Devices (SMD) are mounted on the Surface of a Printed Circuit Board (PCB) by means of specific processes, equipment and materials, and are soldered, cleaned and tested to finally complete assembly, SMT is the most important process Technology in the semiconductor assembly industry, and not only is the development of semiconductor devices towards the direction of chip, miniaturization, thinness, light weight, high reliability and multifunction, but also the development of national defense and Technology is promoted, and SMT is widely applied to the fields of internet of things equipment, communication equipment, aerospace and the like, and quality detection in the SMT assembly process mainly includes solder paste printing quality detection, post-mounting and post-reflow soldering detection, as well as cleaning detection, circuit detection, function detection and the like, SMT detection is an important step for improving the quality of SMT products, the quality of the product can be improved and the cost can be saved. However, most SMT production lines use traditional manual or ordinary equipment inspection methods and techniques during SMT operations, which results in high labor and cost, low work efficiency, high misoperation rate, and low testing efficiency, and cannot adapt to the development of SMT production in terms of product quality and speed. Therefore, there is a need for a smart detection system and method for SMT production line based on machine vision.
Disclosure of Invention
The invention aims to overcome the defects of the prior art, and provides an SMT production line intelligent detection system and method based on machine vision, which can solve the problems of high detection difficulty, high labor cost, low detection efficiency and the like of an SMT production line by adopting technologies such as a machine vision algorithm, a sensor, an Internet of things, cloud service and the like.
The technical scheme adopted by the invention for solving the technical problems is as follows: the SMT production line intelligent detection system and method based on machine vision provided by the invention comprises a PCB to be detected, a feeding channel, a cache unit, a sensor, a control cabinet, an electrical cabinet, a vision sensing unit, a plate turnover machine, a feeding machine, a machine vision detection device, a machine vision detection system, a six-axis industrial robot, a defective product channel, a blanking machine, a blanking channel, a host management and control system, a cloud service platform and the like;
the buffer unit comprises a controller, an operation panel and a lifting upright post, not only can be used for multilayer PCB board assemblies of a plurality of temporary storage and buffer SMT production lines, but also can be used for an accurate and stable lifting buffer platform.
The visual sensing unit comprises a visual camera and is used for preliminarily identifying the appearance shape of the piece to be detected and preliminarily eliminating detection such as wrong pieces, incomplete pieces and the like;
the board turnover machine is used for turning over the PCB after one side of the PCB is detected by the visual sensing unit, so that the subsequent visual sensing unit can conveniently detect the other side of the PCB, and the appearance quality can be preliminarily detected in an all-round way;
preferably, the feeding machine and the blanking machine are composed of a feeding channel, a blanking channel, a feeding mechanism, a blanking mechanism, a conveying belt, a driving wheel, a motor, a sensor, a controller, an alarm, an adjustable upright column and the like;
the sensor is used for acquiring the position information of the PCB to be detected, so that the PCB can be fed and discharged accurately;
the feeding machine is used for feeding the PCB of the SMT production line and transmitting the PCB to each detection unit through the feeding channel, the blanking machine is used for blanking qualified PCB pieces of the tested SMT production line and recovering finished products to a finished product warehouse through the blanking channel, the feeding and blanking processes are all full-automatic transmission, manual participation is not needed, and manpower and material resources are saved;
preferably, the machine vision detection device and the machine vision detection system comprise a quality detection unit, a silk-screen quality detection unit, a mounted component detection unit, a welding quality detection unit, a circuit detection unit, a component function detection unit and the like of a PCB to be detected in an SMT production line, and each detection unit is provided with a high-speed and high-precision intelligent camera;
the quality detection unit is mainly used for intelligently detecting the online detection of product appearance defects such as product surface scratches, dirt, concave-convex and the like through a machine vision detection device and a machine vision detection method;
the screen printing quality detection unit intelligently detects the phenomena of the area size, the position offset, the collapse of the soldering paste and the like of the soldering paste by using a detection device and a detection method of machine vision;
the surface-mounted component detection unit intelligently detects missing components, offset, wrong models and the like by using a machine vision detection device and method;
the welding quality detection unit mainly utilizes a machine vision detection device and method to intelligently detect the phenomena of cold welding, more metal compounds, insufficient soldering paste and the like;
the circuit detection unit intelligently detects whether the installation and welding of the PCB are wrong or not and whether the circuit connection relationship is wrong or not by using a detection device and a detection method of machine vision;
the function detection unit intelligently detects whether circuit hardware of the PCB is wrong or not and whether the product function is normal or not by using a machine vision detection device and method;
preferably, the machine vision detection device and the detection device of the machine vision detection system are provided with six industrial robots, the robots can move back and forth on the guide rails and are responsible for grabbing and placing the detected unqualified PCB onto the defective product channel, the unqualified products are conveyed to the repair station by the conveyor belt of the defective product channel, and the repaired unqualified products can be detected again through the feeding channel;
each function detection unit comprises a plurality of layers of detection stations, each layer of detection station can be automatically fed through a feeding device, the number of pieces to be detected placed on each layer of detection station can be adjusted according to production requirements, the detection requirements of a plurality of PCBs (printed circuit boards) can be met, and after one detection function is finished, the pieces are automatically conveyed to the next detection function unit through a conveying belt until the detection is finished;
preferably, each function detection unit of the machine vision detection device and the machine vision detection system is provided with a display screen, the detection process can be observed through the display screen, the display screen can display test data and test results in real time, each detection unit of the PCB detection device is positioned in the safety protection fence, and an alarm is arranged on a cabinet of each detection unit, so that the system can work normally and safely;
the machine vision detection device and the host management and control system of the machine vision detection system are composed of big data acquisition, analysis and application modules and the like, all the modules are connected with the test unit through the Internet of things, the interface of the management and control system can visually display the test process, the management and control system is provided for the test and production processes, and finally the test data are uploaded to the cloud service platform, so that remote online release of test tasks and remote fault detection can be conveniently carried out through the mobile terminal;
the defect feature extraction algorithm based on machine vision of the machine vision detection device and the machine vision detection system is an improved feature extraction algorithm based on a neural network, and the defect pattern recognition algorithm is an improved pattern recognition classification algorithm based on an SVM (support vector machine), so that the detection efficiency and the detection accuracy can be effectively improved.
The invention provides an SMT production line intelligent detection system and method based on machine vision, wherein the action flow of each detection unit comprises the following steps:
step S1: automatically conveying the PCB to be detected to the to-be-detected area of the feeding channel by using a conveying belt;
step S2: the test cabinet door is automatically opened, a feeding mechanism in the test cabinet places the PCB to be detected on a detection clamp in the detection unit, and the test cabinet door is automatically closed;
step S3: the electrical cabinet automatically powers on and starts the test device and the system;
step S4: intelligently detecting a test item of a to-be-tested piece, automatically opening a box door of a detection unit after detection is finished, and automatically compensating a next PCB to be tested after the detection unit is vacant;
step S5: analyzing and judging the test result of the PCB by using an intelligent analysis system based on machine learning, placing qualified products on a test station of the next test unit, and detecting unqualified PCBs and repairing the unqualified PCBs after the unqualified PCBs are repaired by a defective product channel;
step S6: after the retest is qualified, placing the PCB on a detection station of the next detection unit, and repairing or scrapping the PCB which is unqualified in the retest;
step S7: the PCB intelligent detection of the SMT production line is realized, the requirements of customers are finally met, and various management and service systems are formed.
Compared with the prior art, the invention has the following beneficial effects:
1. the SMT production line intelligent detection device and method based on machine vision have high accuracy, good detection consistency, high efficiency and low cost, and detection results are not influenced by artificial subjective factors;
2. the method can realize the detection of different items customized by customers, has simple and convenient operation process, is convenient to manage, and achieves intelligent detection. The detection device and the detection method also have the advantages of multiple functions, maneuverability, flexibility and self-adaptability, and can work in all weather in various severe environments;
3. related data are uploaded to a cloud management system through the Internet of things, and operating personnel and clients can monitor historical data, parameters and the running state of the system in real time through collected data so as to control the product quality, the normal running of the system and the like, so that basic guarantee is provided for further establishing an unmanned factory, the production automation level, the production capacity and the production efficiency of enterprises are improved, and more capital investment is saved for the enterprises.
Drawings
Other features, objects and advantages of the invention will become more apparent upon reading of the detailed description of non-limiting embodiments with reference to the following drawings:
FIG. 1 is a schematic diagram of an intelligent vision inspection system and apparatus according to the present invention;
FIG. 2 is a schematic diagram of an intelligent visual inspection process according to the present invention;
FIG. 3 is a flow chart of the intelligent image processing of the SMT production line according to the present invention;
FIG. 4 is a control flow chart of the visual inspection of the SMT production line according to the present invention;
FIG. 5 is a flow chart of the vision positioning system for SMT production according to the present invention.
In the figure:
1. the system comprises a PCB to be detected, 2, a sensor, 3, an adjustable upright post, 4, a feeding channel, 5, a safety protective guard, 6, an alarm, 7, a display screen, 8, an unqualified PCB, 9, a defective product channel, 10, a robot track, 11, a detection station, 12, a six-axis robot, 13, an electrical cabinet, 14, a control cabinet, 15, a discharging channel, 16, a cache unit, 17, an image acquisition module, 102, an image digital processing module, 103, an image enhancement module, 104, an image gray scale processing module, 105, an image segmentation module, 106, an image characteristic extraction module, 107, an image analysis module, 108, a defect identification module, 109, an industrial personal computer, 110, 111, a part to be detected, 112, a high-speed high-precision camera, 113, a light source, 114, a mouse, 115, a keyboard, 200, a vision sensing unit, 300, a panel turnover machine, 400, a feeding machine, 500. the system comprises a machine vision detection device, a machine vision detection system, a quality detection unit, a screen printing quality detection unit, a mounted component detection unit, a welding quality detection unit, a circuit detection unit and a component function detection unit, wherein the machine vision detection device comprises 501, the machine vision detection system, 510, the quality detection unit, 520, the screen printing quality detection unit, 530, the mounted component detection unit, 540, the welding quality detection unit, 550, the circuit detection unit and 560, the component function detection unit comprises 600, a blanking machine, 700, a host.
Detailed Description
The present invention will be described in detail with reference to specific examples. The following examples will assist those skilled in the art in further understanding the invention, but are not intended to limit the invention in any way. It should be noted that variations and modifications can be made by persons skilled in the art without departing from the spirit of the invention. All falling within the scope of the present invention.
In this embodiment, the SMT production line intelligent detection system and method based on machine vision provided by the invention include a PCB 1 to be detected, a feeding channel 4, a cache unit 100, a sensor 2, a control cabinet 14, an electrical cabinet 13, a vision sensing unit 200, a panel turnover machine 300, a feeding machine 400, a machine vision detection device 500, a machine vision detection system 501, a six-axis robot 12, a defective product channel 9, a blanking machine 600, a blanking channel 15, a host management and control system 700, a cloud service platform 800, and the like;
the buffer unit 100 comprises a controller, an operation panel and a lifting upright post 3, and can be used for multilayer, mass temporary storage and buffer storage of PCB plate assemblies of SMT production lines, and can be used for an accurate and stable lifting buffer storage platform.
The vision sensing unit 200 contain the vision camera for the appearance shape of treating the detection piece is treated in preliminary discernment, can tentatively get rid of detection such as wrong piece, incomplete piece, the panel turnover machine 300 be used for the upset through the PCB board behind the vision sensing unit 200 detection one side, the another side of subsequent vision sensing unit 200 detection PCB board of being convenient for accomplishes omnidirectional and tentatively detects appearance quality.
The feeding machine 400 and the blanking machine 600 are composed of a feeding channel 4, a blanking channel 15, a feeding mechanism, a blanking mechanism, a conveying belt, a driving wheel, a motor, a sensor 2, a controller, an alarm 6, an adjustable upright post 3 and the like; the sensor 2 is used for acquiring the position information of the PCB to be detected, so that the PCB can be fed and discharged accurately; the feeding machine 400 is used for feeding the PCB of the SMT production line and transmitting the PCB to each detection unit through the feeding channel 4, and the blanking machine 600 is used for blanking the tested qualified PCB of the SMT production line and recycling finished products to a finished product warehouse through the blanking channel 15; the feeding and discharging processes are all full-automatic conveying, manual participation is not needed, and manpower and material resources are saved;
the machine vision detection device 500 and the machine vision detection system 501 comprise a quality detection unit 510, a silk-screen quality detection unit 520, a mounted component detection unit 530, a welding quality detection unit 540, a circuit detection unit 550, a component function detection unit 560 and the like of a PCB to be detected in an SMT production line, and each detection unit is provided with a high-speed and high-precision intelligent camera; the quality detection unit 510 is mainly used for intelligently detecting the online detection of product appearance defects such as product surface scratches, dirt, concave-convex and the like through a machine vision detection device and a machine vision detection method; the screen printing quality detection unit 520 intelligently detects the phenomena of the area size, the position offset, the collapse of the soldering paste and the like of the soldering paste by using a machine vision detection device and method; the mounted component detecting unit 530 intelligently detects missing components, offset, wrong models and the like by using a machine vision detecting device and method; the welding quality detection unit 540 mainly utilizes a machine vision detection device and method to intelligently detect the phenomena of cold welding, more metal compounds, insufficient soldering paste and the like; the circuit detection unit 550 intelligently detects whether the installation and welding of the PCB board are wrong and whether the circuit connection relationship is wrong by using a detection device and a detection method of machine vision; the function detection unit 560 intelligently detects whether circuit hardware of the PCB is wrong and the product function is normal by using a machine vision detection apparatus and method.
The machine vision detection device 500 and the machine vision detection system 501 are provided with six-axis industrial robots 12, the robots can move back and forth on the guide rails 10 and are responsible for grabbing and placing unqualified PCB boards 8 to be detected onto the defective product channel 9, the unqualified products are conveyed to a repair station by a conveyor belt of the defective product channel 9, and the repaired unqualified products 8 are detected again through the feeding channel 4; each function detection unit all contain multilayer detection station 11, detection station 11 on every layer can both be through loading attachment automatic feeding, and the piece quantity that waits that every layer detection station 11 was placed can be adjusted according to the production demand, can satisfy the detection demand of many quantity, many varieties PCB board, through conveyer belt automatic conveying to next detection function unit after accomplishing and detecting a function, until detecting the end.
Each function detection unit of the machine vision detection device 500 and the machine vision detection system 501 is provided with a display screen 7, the detection process can be observed through the display screen 7, the display screen 7 can display test data and test results in real time, each detection unit of the PCB detection device is located in the safety protection fence 5, and the cabinet of each detection unit is provided with an alarm 6, so that the system can work normally and safely.
The host management and control system 700 of the machine vision detection device 500 and the machine vision detection system 501 is composed of big data acquisition, analysis, application modules and the like, all the modules are connected with the test unit through the Internet of things, the interface of the management and control system can visually display the test process, the management and control system is provided for the test and production process, finally the test data are uploaded to the cloud service platform 800, and remote online release of test tasks and remote fault detection are facilitated through the mobile terminal.
The defect feature extraction algorithm based on machine vision of the machine vision detection device 500 and the machine vision detection system 501 is a feature extraction algorithm improved based on a neural network, and the defect pattern recognition algorithm is a pattern recognition classification algorithm improved based on an SVM, so that the detection efficiency and the detection accuracy can be effectively improved.
The invention discloses an SMT production line intelligent detection system and method based on machine vision, wherein the action flow of each detection unit comprises the following steps:
step S1: the PCB to be detected is automatically conveyed to the area to be detected of the feeding channel 4 by a conveyor belt;
step S2: the test cabinet door is automatically opened, a feeding mechanism in the test cabinet places the PCB to be detected on a detection clamp in the detection unit, and the test cabinet door is automatically closed;
step S3: the electrical cabinet automatically powers on and starts the test device and the system;
step S4: intelligently detecting a test item of a to-be-tested piece, automatically opening a box door of a detection unit after detection is finished, and automatically compensating a next PCB to be tested after the detection unit is vacant;
step S5: analyzing and judging the test result of the PCB by using an intelligent analysis system based on machine learning, placing qualified products on a test station of the next test unit, and retesting the detected unqualified PCB after repairing the unqualified PCB by a defective product channel 9;
step S6: after the retest is qualified, placing the PCB on a detection station of the next detection unit, and repairing or scrapping the PCB which is unqualified in the retest;
step S7: the PCB intelligent detection of the SMT production line is realized, the requirements of customers are finally met, and various management and service systems are formed.
The SMT production line intelligent detection system and the SMT production line intelligent detection method based on machine vision are implemented, the detection system mainly comprises a PCB to be detected, a feeding channel, a cache unit, a sensor, a control cabinet, an electric cabinet, a vision sensing unit, a plate turnover machine, a feeding machine, a machine vision detection device, a machine vision detection system, a six-axis robot, a defective product channel, a blanking machine, a blanking channel, a host management and control system, a cloud service platform and the like, and the whole intelligent detection device and system are surrounded by a safety protective guard to ensure personal safety; the PCB to be detected is conveyed to each station through a feeding channel; the feeding channel penetrates through the concave inlets and outlets of the cabinet bodies of the detection units and is used for fully automatically conveying the PCB to be detected on the SMT production line; the PCB to be detected on the feeding channel can be temporarily stored on the cache unit and used for storing the piece to be detected in a buffering mode; the sensors are distributed on the conveying system and each detection station and are used for acquiring the position information of the piece to be detected, so that the piece to be detected can be conveniently and accurately grabbed in the subsequent process; the control cabinet consists of various PLC controllers, a singlechip and the like and is used for controlling the action of each unit of the whole machine vision intelligent test system; the electrical cabinet is used for supplying power, gas and other resources to the whole test system; the visual sensing unit is used for identifying the appearance shape of the piece to be detected and preliminarily eliminating wrong pieces, incomplete pieces and the like; the board turnover machine is used for detecting turnover of the PCB to be detected; the feeding machine is mainly used for automatically feeding and conveying the to-be-detected piece; the machine vision detection device is positioned on the transmission system and is arranged in a cabinet of each detection station; the machine vision detection system comprises six vision detection functional systems which are respectively used for the omnibearing and multifunctional detection of the PCB on the SMT production line; the six-axis robot is used for grabbing PCB boards which are detected to be unqualified on the SMT production line and placing the PCB boards on a unqualified product channel; the unqualified product channel is used for returning unqualified products to a repair station or a waste product warehouse; the blanking machine is used for transporting the qualified PCB on the SMT production line back to a finished product warehouse after the test is finished; the host management and control system platform is connected with each unit of the whole test system through the Internet of things, and a platform interface can visually display the test process; the cloud service platform is used for cloud computing and big data analysis detection piece data, can remotely issue and receive test tasks on line through the mobile terminal, and provides various cloud services for customers. The invention adopts a plurality of groups of machine vision detection devices and test systems to carry out multifunctional test and omnibearing test on the PCB on the SMT production line, ensures the personal safety in the test process, realizes the machine vision intelligent detection of the PCB on the SMT production line to replace manual detection, improves the detection accuracy and the detection efficiency of the PCB assembly, can meet diversified test requirements, can connect the test systems with the cloud, realizes the online release and reception of test tasks through a mobile terminal, provides various cloud services for customers, realizes the traceability of test data, and reduces the communication cost and the time cost.
Claims (9)
1. An SMT production line intelligent detection system and method based on machine vision, which comprises an SMT production line and a detection system, the detection system is characterized by mainly comprising a PCB (1) to be detected, a feeding channel (4), a cache unit (100), a sensor (2), a control cabinet (14), an electrical cabinet (13), a vision sensing unit (200), a plate turnover machine (300), a feeding machine (400), a machine vision detection device (500), a machine vision detection system (501), a six-axis robot (12), a defective product channel (9), a blanking machine (600), a blanking channel (15), a host management and control system (700) and a cloud service platform (800), wherein the cache unit (100) comprises a controller, an operation panel and a lifting upright post (3), and the cache unit (100) is used for PCB assembly of a multilayer temporary storage and cache SMT production line and can be used for an accurate and stable lifting cache platform.
2. The SMT production line intelligent detection system and method based on machine vision according to claim 1, wherein the vision sensing unit (200) comprises a vision camera, the vision camera is used for preliminarily recognizing appearance shapes of pieces to be detected, detection of wrong pieces, incomplete pieces and the like can be preliminarily eliminated, the board turnover machine (300) is used for overturning the PCB after one side of the PCB is detected by the vision sensing unit (200), detection of the other side of the PCB by the subsequent vision sensing unit (200) is facilitated, and omnibearing preliminary detection of appearance quality is achieved.
3. A machine vision based SMT production line intelligent detection system and method according to claim 1, the automatic feeding device is characterized in that the feeding machine (400) and the blanking machine (600) are composed of a feeding channel (4), a blanking channel (15), a feeding mechanism, a blanking mechanism, a conveying belt, a driving wheel, a motor, a sensor (2), a controller, an alarm (6) and an adjustable upright post (3), the sensor (2) is used for acquiring the position information of a PCB to be detected, so that the PCB can be accurately fed and blanked, the feeding machine (400) is used for feeding the PCB in an SMT production line, and is transmitted to each detection unit through a feeding channel (4), a blanking machine (600) is used for blanking qualified PCB pieces of the tested SMT production line, finished products are recovered to a finished product warehouse through a blanking channel (15), and the feeding and blanking processes are all full-automatic transmission without manual participation.
4. The SMT production line intelligent detection system and method based on machine vision according to claim 1, wherein the machine vision detection device (500) and the machine vision detection system (501) comprise a quality detection unit (510) of a PCB to be detected of the SMT production line, a silk screen quality detection unit (520), a mounted component detection unit (530), a welding quality detection unit (540), a circuit detection unit (550), and a component function detection unit (560), each detection unit is configured with a high-speed and high-precision intelligent camera, the quality detection unit (510) intelligently detects the on-line detection of product appearance defects such as scratch, dirt, and concave-convex on the surface of a product mainly through the machine vision detection device, and the silk screen quality detection unit (520) intelligently detects the area size of solder paste through the machine vision detection device, The detection device and the detection method mainly utilize the machine vision to intelligently detect the phenomena of cold welding, more metal compounds, insufficient soldering paste amount and the like, the circuit detection unit (550) utilizes the machine vision detection device to intelligently detect whether the installation and welding of the PCB are wrong or not and whether the circuit connection relationship is wrong or not, and the function detection unit (560) utilizes the machine vision detection device to intelligently detect whether the circuit hardware of the PCB is wrong or not and whether the product function is normal or not.
5. The SMT production line intelligent detection system and method based on machine vision as claimed in claim 1, wherein the machine vision detection device (500) and the machine vision detection system (501) are configured with six industrial robots (12), the robots can move back and forth on guide rails (10) and are responsible for grabbing and placing PCB boards (8) which are detected to be unqualified into defective product channels (9), the defective products are conveyed to repair stations through conveyor belts of the defective product channels (9), the repaired defective products (8) can be detected again through a feeding channel (4), each function detection unit comprises a plurality of layers of detection stations (11), each layer of detection stations (11) can be automatically fed through a feeding device, the number of pieces to be detected placed in each layer of detection stations (11) can be adjusted according to production requirements, and a large number of pieces can be met, The detection requirements of the various PCBs are met, and the PCB is automatically transmitted to the next detection functional unit through the conveyor belt after one detection function is completed until the detection is finished.
6. The SMT production line intelligent detection system and method based on machine vision according to claim 1, wherein each functional detection unit of the machine vision detection device (500) and the machine vision detection system (501) is provided with a display screen (7), a detection process can be observed through the display screen (7), the display screen (7) can display test data and test results in real time, each detection unit of the PCB detection device is located in the safety protection fence (5), and an alarm (6) is arranged on a cabinet of each detection unit, so that normal and safe operation of the system can be ensured.
7. The SMT production line intelligent detection system and method based on machine vision according to claim 1, wherein a host management and control system (700) of the machine vision detection device (500) and the machine vision detection system (501) is composed of big data acquisition, analysis and application modules, the modules and the test units are connected through the Internet of things, an interface of the management and control system can visually display a test process, the management and control system is provided for the test and production process, and finally test data are uploaded to a cloud service platform (800), so that remote online release of test tasks and remote fault detection can be performed through a mobile terminal.
8. The system and the method for intelligently detecting the SMT production line based on the machine vision according to claim 1, wherein the machine vision based defect feature extraction algorithms of the machine vision detection device (500) and the machine vision detection system (501) are feature extraction algorithms improved based on a neural network, and the defect pattern recognition algorithm is a pattern recognition classification algorithm improved based on an SVM, so that the detection efficiency and the detection accuracy can be effectively improved.
9. An SMT production line intelligent detection system and method based on machine vision according to claim 1, wherein the motion flow of each detection unit comprises the following steps:
step S1: the PCB to be detected is automatically conveyed to the area to be detected of the feeding channel (4) by a conveyor belt;
step S2: the test cabinet door is automatically opened, a feeding mechanism in the test cabinet places the PCB to be detected on a detection clamp in the detection unit, and the test cabinet door is automatically closed;
step S3: the electrical cabinet automatically powers on and starts the test device and the system;
step S4: intelligently detecting a test item of a to-be-tested piece, automatically opening a box door of a detection unit after detection is finished, and automatically compensating a next PCB to be tested after the detection unit is vacant;
step S5: analyzing and judging the test result of the PCB by using an intelligent analysis system based on machine learning, placing qualified products on a test station of the next test unit, and retesting the detected unqualified PCB after repairing the unqualified PCB by a defective product channel (9);
step S6: after the retest is qualified, placing the PCB on a detection station of the next detection unit, and repairing or scrapping the PCB which is unqualified in the retest;
step S7: the PCB intelligent detection of the SMT production line is realized, the requirements of customers are finally met, and various management and service systems are formed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010298962.5A CN111359912A (en) | 2020-04-08 | 2020-04-08 | SMT production line intelligent detection system and method based on machine vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010298962.5A CN111359912A (en) | 2020-04-08 | 2020-04-08 | SMT production line intelligent detection system and method based on machine vision |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111359912A true CN111359912A (en) | 2020-07-03 |
Family
ID=71201575
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010298962.5A Pending CN111359912A (en) | 2020-04-08 | 2020-04-08 | SMT production line intelligent detection system and method based on machine vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111359912A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111882557A (en) * | 2020-09-28 | 2020-11-03 | 成都睿沿科技有限公司 | Welding defect detection method and device, electronic equipment and storage medium |
| CN112007867A (en) * | 2020-07-22 | 2020-12-01 | 温州大学瓯江学院 | Distribution device based on big data |
| CN112487706A (en) * | 2020-11-13 | 2021-03-12 | 北京遥测技术研究所 | Automatic mounting parameter intelligent decision method based on ensemble learning |
| CN112827866A (en) * | 2020-12-30 | 2021-05-25 | 大连理工江苏研究院有限公司 | Intelligent product identification device and identification method for biological medicine |
| CN113319548A (en) * | 2021-05-31 | 2021-08-31 | 无锡恒久安泰智能制造有限公司 | Intelligent automatic production method for standard knot of tower crane |
| CN113484732A (en) * | 2021-07-26 | 2021-10-08 | 广东喜珍电路科技有限公司 | Method for improving PCB electrical measurement efficiency |
| CN113643286A (en) * | 2021-10-12 | 2021-11-12 | 南通海美电子有限公司 | Electronic component assembly detection method and system |
| CN113996556A (en) * | 2021-09-27 | 2022-02-01 | 深圳技术大学 | Cloud-integrated product sorting system and method |
| CN114324349A (en) * | 2021-12-02 | 2022-04-12 | 晋江海纳机械有限公司 | Device for counting yield of sanitary equipment based on big data |
| CN114779730A (en) * | 2022-06-16 | 2022-07-22 | 深圳市信润富联数字科技有限公司 | Detection control system |
| CN115366118A (en) * | 2022-08-09 | 2022-11-22 | 广东机电职业技术学院 | Relay detection system and method based on robot and vision technology |
-
2020
- 2020-04-08 CN CN202010298962.5A patent/CN111359912A/en active Pending
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112007867A (en) * | 2020-07-22 | 2020-12-01 | 温州大学瓯江学院 | Distribution device based on big data |
| CN111882557B (en) * | 2020-09-28 | 2021-01-05 | 成都睿沿科技有限公司 | Welding defect detection method and device, electronic equipment and storage medium |
| CN111882557A (en) * | 2020-09-28 | 2020-11-03 | 成都睿沿科技有限公司 | Welding defect detection method and device, electronic equipment and storage medium |
| CN112487706B (en) * | 2020-11-13 | 2023-10-17 | 北京遥测技术研究所 | Automatic mounting parameter intelligent decision method based on ensemble learning |
| CN112487706A (en) * | 2020-11-13 | 2021-03-12 | 北京遥测技术研究所 | Automatic mounting parameter intelligent decision method based on ensemble learning |
| CN112827866A (en) * | 2020-12-30 | 2021-05-25 | 大连理工江苏研究院有限公司 | Intelligent product identification device and identification method for biological medicine |
| CN113319548A (en) * | 2021-05-31 | 2021-08-31 | 无锡恒久安泰智能制造有限公司 | Intelligent automatic production method for standard knot of tower crane |
| CN113484732A (en) * | 2021-07-26 | 2021-10-08 | 广东喜珍电路科技有限公司 | Method for improving PCB electrical measurement efficiency |
| CN113996556A (en) * | 2021-09-27 | 2022-02-01 | 深圳技术大学 | Cloud-integrated product sorting system and method |
| CN113643286A (en) * | 2021-10-12 | 2021-11-12 | 南通海美电子有限公司 | Electronic component assembly detection method and system |
| CN114324349A (en) * | 2021-12-02 | 2022-04-12 | 晋江海纳机械有限公司 | Device for counting yield of sanitary equipment based on big data |
| CN114779730A (en) * | 2022-06-16 | 2022-07-22 | 深圳市信润富联数字科技有限公司 | Detection control system |
| CN115366118A (en) * | 2022-08-09 | 2022-11-22 | 广东机电职业技术学院 | Relay detection system and method based on robot and vision technology |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111359912A (en) | SMT production line intelligent detection system and method based on machine vision | |
| US11213860B2 (en) | Automatic magnetic core sorting system based on machine vision | |
| CN106770331B (en) | Workpiece defect detection system based on machine vision | |
| CN108174523A (en) | A kind of PCB appearances on-bne repair equipment and repair method | |
| CN113458764B (en) | Automatic disassembly equipment and method for apple mobile phone | |
| CN112415973B (en) | Intelligent control method and system for inner layer AOI (automated optical inspection) process | |
| CN112691926A (en) | Automatic quality detection device based on machine vision and working method thereof | |
| CN207798388U (en) | A kind of control valve automatic detection device | |
| CN112718552A (en) | Automatic quality defect inspection device for LED circuit board and working method thereof | |
| CN216174408U (en) | Display screen burning, detecting and labeling equipment | |
| CN212916627U (en) | Intelligent detection system of SMT production line based on machine vision | |
| CN112213325A (en) | AOI detection system and detection method thereof | |
| CN113723127B (en) | Intelligent checking channel for high-speed rail station inspection of passengers | |
| Teoh et al. | Automated visual inspection of surface mount PCBs | |
| CN111307817B (en) | Online detection method and system for PCB production process of intelligent production line | |
| CN112304965A (en) | Multi-scene fuse box visual detection system | |
| CN208399951U (en) | The equipment and system that the test result of circuit board is marked | |
| CN115744254A (en) | Door plant automated inspection conveyor | |
| CN215656519U (en) | System for automatically sorting products | |
| US7159754B2 (en) | Apparatus and method for corrective soldering | |
| CN212245279U (en) | High-speed board collecting and releasing machine with MES system and AGV car | |
| CN201269742Y (en) | Detection device | |
| CN112474383A (en) | Appearance detection device and detection method | |
| CN112747788A (en) | Plank check out test set and plank production line | |
| CN209853300U (en) | Automatic board device of putting of receiving of PCB board |
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 |