CN113670923B - Omnibearing high-precision detection device and method for defects of notebook computer shell - Google Patents

Omnibearing high-precision detection device and method for defects of notebook computer shell Download PDF

Info

Publication number
CN113670923B
CN113670923B CN202110968422.8A CN202110968422A CN113670923B CN 113670923 B CN113670923 B CN 113670923B CN 202110968422 A CN202110968422 A CN 202110968422A CN 113670923 B CN113670923 B CN 113670923B
Authority
CN
China
Prior art keywords
notebook computer
image
conveyor belt
cmos camera
computer
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
Application number
CN202110968422.8A
Other languages
Chinese (zh)
Other versions
CN113670923A (en
Inventor
李磊
黄海鸿
周帮来
李新宇
刘志峰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hefei University of Technology
Original Assignee
Hefei University of Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Hefei University of Technology filed Critical Hefei University of Technology
Priority to CN202110968422.8A priority Critical patent/CN113670923B/en
Publication of CN113670923A publication Critical patent/CN113670923A/en
Application granted granted Critical
Publication of CN113670923B publication Critical patent/CN113670923B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20081—Training; Learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20084—Artificial neural networks [ANN]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20092—Interactive image processing based on input by user
    • G06T2207/20104—Interactive definition of region of interest [ROI]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20212—Image combination
    • G06T2207/20221—Image fusion; Image merging
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30108—Industrial image inspection
    • G06T2207/30164—Workpiece; Machine component

Abstract

The invention discloses an omnibearing high-precision detection device and method for a shell defect of a notebook computer, wherein the detection device consists of an industrial personal computer, an image acquisition module, a lower computer control module, a conveyer belt, a flip mechanism and a jacking rotating mechanism; the industrial personal computer and the lower computer control module control the notebook computer to stop when reaching a specified detection station, and the servo motor is used for controlling the flip mechanism and the jacking rotating mechanism to perform angle adjustment on each surface of the notebook computer in a coordinated manner, so that images of each surface of the notebook computer are collected through the image collection module and provided for the industrial personal computer, and the collected images of each surface are classified and fused, and then the classification and identification of image characteristics-defect forms are performed. The invention can realize the omnibearing high-precision detection of the defects of the notebook computer shell.

Description

Omnibearing high-precision detection device and method for defects of notebook computer shell
Technical Field
The invention relates to an omnibearing high-precision detection device and method for defects of a notebook computer shell.
Background
The surface of the notebook computer shell can be provided with defects such as scratches, bumps and pits, at present, manual detection is generally adopted for detecting the defects of the surface of the notebook computer shell, the problems of long detection time, different standards, high cost and the like exist in manual detection, in addition, in a visual detection system, a shooting camera always shoots at an angle, the defect detection accuracy is reduced due to the fact that the shooting angle is not in place, and the detection efficiency is seriously reduced.
Disclosure of Invention
The invention aims to solve the defects in the prior art, and provides an omnibearing high-precision detection device and method for detecting the defects of the shell of a notebook computer, so that the omnibearing automatic detection of the defects of the shell of the notebook computer on multiple surfaces can be realized, and the detection precision and detection efficiency of the defects of the shell of the notebook computer are improved.
In order to achieve the aim of the invention, the invention adopts the following technical scheme:
the invention relates to an omnibearing high-precision detection device for detecting defects of a notebook computer shell, which is characterized in that a conveyer belt is arranged, and a first conveyer belt and a second conveyer belt are sequentially arranged on the conveyer belt along the conveying direction;
a push rod assembly is arranged on one side of the first conveyor belt, and a rubber block is arranged at the movable end of the push rod cylinder and used for adjusting the initial position of a notebook computer on the first conveyor belt, so that the notebook computer is positioned in the middle position of the first conveyor belt;
the same sides of the first conveying belt and the second conveying belt are provided with a flip mechanism, and the flip mechanism is provided with a supporting plate between the first conveying belt and the second conveying belt; the support plate is fixedly provided with a linkage assembly through a support seat, the support end of the linkage assembly is movably hinged with the support seat through a hinge bolt, one end of the linkage assembly is connected with the movable end of the rodless cylinder on the support plate, and the other end of the linkage assembly is provided with a first vacuum chuck group for adsorbing the A surface of the notebook computer; the rodless cylinder drives the linkage assembly to act and drives the A surface of the notebook computer on the first vacuum chuck group to perform flip action;
the bottom between the first conveying belt and the second conveying belt is provided with a jacking rotating mechanism, the jacking rotating mechanism is characterized in that a first servo motor is fixed on a mounting plate at the bottom, the first servo motor is connected with a ball screw through a first coupler, the ball screw penetrates through a middle plate, and a top plate is arranged at the top of the ball screw; the middle plate is provided with a rotating mechanism, the middle plate at one side of the rotating mechanism is provided with a guide rod, the bottom of the guide rod is arranged on the mounting plate, and the top of the guide rod is arranged on the top plate; the first servo motor drives the ball screw to rotate and drives a rotating mechanism on the middle plate to move up and down along the ball screw between the mounting plate and the top plate;
the rotating mechanism is provided with a second servo motor at the middle position of the middle plate and is connected with one end of a rotating shaft through a second coupler and a speed reducer, the other end of the rotating shaft is connected with one end of a connecting shaft, a sucker mounting frame is mounted at the other end of the connecting shaft, and second vacuum sucker groups are mounted on the sucker mounting frame and are uniformly distributed in a rectangular shape and used for adsorbing the D surface of a notebook computer; the second servo motor drives the connecting shaft to rotate and drives the notebook computer D surface on the second vacuum chuck group to do rotary motion at the bottom between the first conveying belt and the second conveying belt; and the second servo motor adjusts the rotation speed of the rotation mechanism;
a detection station is arranged between the first conveyor belt and the second conveyor belt, and a first image acquisition module and a second image acquisition module are arranged on the detection station;
the first image acquisition module is characterized in that a first movable frame group is arranged at the bottom of one side of the first conveyor belt, a first CMOS camera and a first linear light source are arranged on the first movable frame group, the first linear light source is arranged below the first CMOS camera, the first CMOS camera is positioned above the detection station, and the first CMOS camera is connected with the industrial personal computer through a first image acquisition card; adjusting imaging definition of the first CMOS camera by adjusting brightness of the first linear light source;
the second image acquisition module is characterized in that a second movable frame group is arranged at the bottom of the other side of the first conveyor belt, the second CMOS camera and a second linear light source are arranged on the second movable frame group, the second linear light source is arranged below the second CMOS camera, and the second CMOS camera is positioned at the side edge of the detection station; the second CMOS camera is connected with the industrial personal computer through a second image acquisition card; adjusting the imaging definition of the second CMOS camera by adjusting the brightness of the second linear light source;
the industrial personal computer is also in signal transmission with a lower computer control module through serial port communication, and the lower computer control module is respectively connected with the first conveyor belt, the second conveyor belt, the flip mechanism and the jacking rotating mechanism and used for controlling the operation of the device; the industrial personal computer classifies and fuses the images acquired by the first image acquisition module and the second image acquisition module through an image processing algorithm to obtain omnibearing notebook shell image information, and then uses a deep learning model to classify and identify image characteristics-defect forms of the notebook shell image information.
The omnibearing high-precision detection device for the defects of the notebook computer shell is also characterized in that:
taking the central position of the horizontal plane of the area between the first conveying belt and the second conveying belt as an origin O, taking the conveying direction along the first conveying belt as an X-axis direction, taking the direction parallel to the horizontal plane and vertical to the conveying direction of the first conveying belt as a Y-axis direction, and taking the vertical upward direction vertical to the horizontal plane as a Z-axis direction, so as to establish a world coordinate system;
let the size of the notebook computer at the detection station be a×b, and the size of the image of the first CMOS camera shooting field of view be L 1 ×W 1 ；
When the notebook computer is positioned at the origin O position of the world coordinate system, the installation position of the first CMOS camera in the Z-axis direction is adjusted so that the shooting view field of the first CMOS camera is vertically and perpendicularly perpendicular to the XOY plane of the world coordinate system, and the pixel distances from the left and right edges of the A plane of the notebook computer to the left and right edges of the shooting view field of the first CMOS camera in the picture of the shooting view field of the first CMOS camera are c 1 The pixel distance from the upper and lower edges of the A surface of the notebook computer to the upper and lower edges of the shooting view field of the first CMOS camera is d 1 And satisfies:
Figure BDA0003225066760000031
the second CMOS camera shoots a view field picture with the size L 2 ×W 2 ；
When the notebook computer is positioned at the original point O position of the world coordinate system, the second CMOS camera is adjusted to be positioned in the positive directions of the Y axis and the Z axis of the world coordinate system, so that the shooting view field of the second CMOS camera is vertically and positively seen on the XOZ plane of the world coordinate system; and the pixel distances from the left and right edges of the B surface of the notebook computer in the picture of the second CMOS camera shooting view field to the left and right edges of the second CMOS camera shooting view field are c 2 The pixel distance from the upper and lower edges of the B surface of the notebook computer to the upper and lower edges of the shooting view field of the second CMOS camera is d 2 And satisfies:
Figure BDA0003225066760000032
the invention relates to a detection method of an omnibearing high-precision detection device for detecting defects of a notebook computer shell, which is characterized by comprising the following steps:
step 1: the notebook computer to be detected is horizontally placed on a first running conveyor belt after being closed, and the opening and closing end is along the conveying direction of the conveyor belt and is on the same side as the push rod assembly; the push rod assembly adjusts the position of the notebook computer on the first conveyor belt under the action of the push rod air cylinder, so that the push rod assembly resets after the notebook computer is positioned at the middle position on the first conveyor belt;
step 2: the first image acquisition module acquires the digital image on the detection station in real time and sends the digital image to the industrial personal computer; the industrial personal computer identifies the characteristic points of the received digital image according to the position information and the number of the prestored characteristic points on the surface of the notebook computer, if the number of the identified characteristic points is matched with the number of the prestored characteristic points, the position deviation value of the characteristic points in the digital image is calculated according to the position information of the prestored characteristic points, if the position deviation value is smaller than the set threshold value, the notebook computer to be detected is judged to reach the designated position, and a shutdown signal is sent to the lower computer control module so as to control the first conveyor belt and the second conveyor belt to stop running;
step 3: after the industrial personal computer receives a shutdown completion signal fed back by the lower computer control module, acquiring an A-plane image of the notebook computer to be detected by using the first image acquisition module;
step 4: after the acquisition of the A-plane image is completed, the industrial personal computer sends an acquisition signal to a lower computer control module, and the acquisition signal is used for controlling the flip mechanism and the bottom jacking rotating mechanism to act, so that the bottom jacking rotating mechanism performs jacking action to drive the second vacuum chuck group to suck the D-plane of the notebook computer, and meanwhile, the flip mechanism drives the first vacuum chuck group to suck the A-plane of the notebook computer, and then 90-degree flip action is performed to open the notebook computer;
step 5: when the flip action is completed, the first vacuum chuck group loosens the A surface of the notebook computer, the lower computer control module controls the jacking rotating mechanism to rotate, and the rotating mechanism is set to drive the notebook computer to stop after rotating for a delta angle through the second servo motor, so that the industrial personal computer controls the first image acquisition module to acquire C surface images of the notebook computer at a stopping interval, and controls the second image acquisition module to acquire A, B surface images of the notebook computer until A, B, C surface images are acquired after the notebook computer rotates for one circle, and then the images are sent to the industrial personal computer for classification and fusion treatment;
step 6: the lower computer control module controls the flip mechanism to drive the first vacuum chuck group to suck the A surface of the notebook computer, and then 90-degree closing action is performed to close the notebook computer;
step 7: the lower computer control module controls the jacking rotating mechanism to drive the notebook computer to rotate 180 degrees, and then controls the flip mechanism to drive the first vacuum chuck group to suck the A surface of the notebook computer and then turn over 90 degrees, so that the bottom of the notebook computer faces the second image acquisition module; the second image acquisition module acquires a D-plane image of the notebook computer;
step 8: after the acquisition of the D-plane image is completed, the industrial personal computer sends a reset signal to the lower computer control module so as to control the flip mechanism to drive the first vacuum chuck group to reversely flip by 90 degrees, so that after the notebook computer is reset, an operation signal is sent to the lower computer control module so as to control the first conveyor belt and the second conveyor belt to resume operation;
step 9: the industrial personal computer carries out histogram equalization processing and median filtering processing on the acquired image, and then cuts the processed image according to a proportion, so as to establish a gray scale template based on an image pyramid;
the industrial personal computer corrects the positions and the directions of the A-plane image, the B-plane image and the C-plane image of the notebook computer according to the gray template; and combining the corrected images into an image and dividing the region of interest to obtain a region of interest sample set, wherein the region of interest sample set is used for training and optimizing a deep learning model to realize classification and identification of image feature-defect forms.
Compared with the prior art, the invention has the beneficial effects that:
1. compared with the traditional detection mode, the invention utilizes the machine vision mode to detect, and realizes the omnibearing detection of the defects of the notebook computer shell by the cooperative action of the flip mechanism and the jacking rotating mechanism and the cooperation of the image acquisition module.
2. According to the invention, the proportional relation range between the size of the camera view field and the size of the notebook computer is determined, the installation position of the camera is fixed, the imaging condition of the detected defects of the notebook computer shell in the camera view field is determined, the unified detection of the sizes of different notebook computer shells is satisfied, and the size diversity of detection objects is expanded.
3. The invention receives the visual signal through the industrial personal computer, calculates the position deviation value according to the pre-stored ideal position information of the detection object to adjust the position relation of the detection object, judges that the detection object reaches the designated position, controls the conveyor belt to stop for image acquisition, and improves the detection precision and efficiency.
4. According to the invention, the rotation speed and the rotation angle of the jacking rotation mechanism are precisely controlled through the servo motor, the image information of the A, B, C surface of the notebook computer is continuously collected for a plurality of times, the industrial personal computer classifies and fuses the collected images of all the surfaces, and the defect information is judged through the deep learning model, so that the high-precision detection of the defects of the shell of the notebook computer is realized.
Drawings
FIG. 1 is a diagram of a world coordinate system of the present invention;
FIG. 2 is a flow chart of the detection of the present invention;
FIG. 3 is a schematic diagram of an axial measurement of the detection platform of the present invention;
FIG. 4 is a block diagram of a flip mechanism of the present invention;
FIG. 5 is a block diagram of a jacking and rotating mechanism according to the present invention;
FIG. 6 is a view of a rotating portion of the mechanism of FIG. 5;
FIG. 7 is a schematic diagram of a notebook computer B, C for collection according to the present invention;
fig. 8 shows the side B, C of the notebook computer rotated by 30 °;
FIG. 9 is a schematic view of a notebook computer rotated 180 degrees;
reference numerals in the drawings: 1 a first conveyor belt; 2 a second conveyor belt; 3, a flip mechanism; 4, a rodless cylinder; 5, a first vacuum chuck group; 6, linking the supporting seat of the assembly; 7, lifting the rotating mechanism; 8, a first servo motor; 9 a rotating mechanism; 10, a coupling; 11 ball screw; 12 guide rods; 13 mounting plates; 14 a middle plate; 15 a second servo motor; a 16 speed reducer; 17 a rotation shaft; 18 connecting shafts; 19 sucker mounting frames; a second vacuum chuck group 20; a first image acquisition module 21; a second image acquisition module 22; a first CMOS camera 23; a first linear light source 24; 25 a first movable frame set; a second CMOS camera 26; a second linear light source 27; 28 a second set of mobile racks; 29 a push rod assembly; 30 push rod cylinder; 31 a support plate; 32 linkage components; 33 rubber blocks.
Detailed Description
In this embodiment, as shown in fig. 3, a conveying belt is provided for an omnibearing high-precision detection device for a defect of a notebook computer casing, and the conveying belt is sequentially provided with a first conveying belt 1 and a second conveying belt 2 along a conveying direction;
a push rod assembly 29 is arranged on one side of the first conveyor belt 1, and a rubber block 33 is arranged at the movable end of a push rod cylinder 30 of the push rod assembly 29 and used for adjusting the initial position of a notebook computer on the first conveyor belt 1, so that the notebook computer is positioned in the middle position on the first conveyor belt 1, and the notebook computer can conveniently and accurately move to a detection station;
as shown in fig. 4, a flip mechanism 3 is disposed on the same side of the first conveyor belt 1 and the second conveyor belt 2, and a support plate 31 is disposed between the first conveyor belt 1 and the second conveyor belt 2 in the flip mechanism 3; the support plate 31 is fixedly provided with a linkage assembly 32 through the support seat 6, the support end of the linkage assembly 32 is movably hinged with the support seat 6 through a hinge bolt, one end of the linkage assembly 32 is connected with the movable end of the rodless cylinder 4 on the support plate 31, and the other end of the linkage assembly 32 is provided with a first vacuum chuck group 5 for adsorbing the A surface of the notebook computer; the rodless cylinder 4 drives the linkage assembly 32 to act and drives the A surface of the notebook computer on the first vacuum chuck group 5 to perform flip-up action;
as shown in fig. 3, a lifting rotating mechanism 7 is installed at the bottom between the first conveyor belt 1 and the second conveyor belt 2, as shown in fig. 5, the lifting rotating mechanism 7 is formed by fixing a first servo motor 8 on a mounting plate 13 at the bottom, the first servo motor 8 is connected with a ball screw 11 through a first coupling 10, the ball screw 11 penetrates through an intermediate plate 14, and a top plate is installed at the top of the ball screw 11; the rotary mechanism 9 is arranged on the middle plate 14, the guide rod 12 is arranged on the middle plate 14 at one side of the rotary mechanism 9, the bottom of the guide rod 12 is arranged on the mounting plate 13, and the top of the guide rod 12 is arranged on the top plate; the first servo motor 8 drives the ball screw 11 to rotate and drives the rotating mechanism 9 on the middle plate 14 to move up and down along the ball screw 11 between the mounting plate 13 and the top plate; the lifting rotating mechanism 7 drives the notebook computer to move up and down at the detection station, so that the detection of defects of the notebook computer shell is facilitated;
as shown in fig. 6, the rotating mechanism 9 is provided with a second servo motor 15 in the middle position of the middle plate 14, and is connected with one end of a rotating shaft 17 through a second coupler and a speed reducer 16, the other end of the rotating shaft 17 is connected with one end of a connecting shaft 18, a sucker mounting frame 19 is mounted at the other end of the connecting shaft 18, and a second vacuum sucker group 20 is mounted on the sucker mounting frame 19 and uniformly distributed in a rectangular shape for adsorbing the D surface of the notebook computer; the second servo motor 15 drives the connecting shaft 17 to rotate and drives the notebook computer D surface on the second vacuum chuck group 20 to do rotary motion at the bottom between the first conveyor belt 1 and the second conveyor belt 2; the second servo motor 15 adjusts the rotation speed of the rotation mechanism 9 to control the rotation of the notebook computer;
a detection station is arranged between the first conveyor belt 1 and the second conveyor belt 2, and a first image acquisition module 21 and a second image acquisition module 22 are arranged on the detection station;
the first image acquisition module 21 is characterized in that a first movable frame group 25 is arranged at the bottom of one side of the first conveyor belt 1, a first CMOS camera 23 and a first linear light source 24 are arranged on the first movable frame group 25, the first linear light source 24 is arranged below the first CMOS camera 23, the first CMOS camera 23 is positioned above a detection station, and the first CMOS camera 23 is connected with an industrial personal computer through a first image acquisition card; and adjusts the imaging sharpness of the first CMOS camera 23 by adjusting the brightness of the first linear light source 14;
the second image acquisition module 22 is provided with a second movable frame group 28 at the bottom of the other side of the first conveyor belt 1, a second CMOS camera 26 and a second linear light source 27 are arranged on the second movable frame group 28, the second linear light source 27 is arranged below the second CMOS camera 26, and the second CMOS camera 26 is positioned at the side of the detection station; the second CMOS camera 26 is connected with the industrial personal computer through a second image acquisition card; and adjusts the imaging sharpness of the second CMOS camera 26 by adjusting the brightness of the second linear light source 27;
taking the central position of the horizontal plane of the area between the first conveyor belt 1 and the second conveyor belt 2 as an origin O, taking the conveying direction along the first conveyor belt 1 as an X-axis direction, taking the direction parallel to the horizontal plane and vertical to the conveying direction of the first conveyor belt 1 as a Y-axis direction, and taking the vertical upward direction vertical to the horizontal plane as a Z-axis direction, so as to establish a world coordinate system, as shown in figure 1;
let the size of the notebook computer at the detection station be a×b, and the size of the image of the field of view photographed by the first CMOS camera 23 be L 1 ×W 1 ；
When the notebook computer is at the origin O position of the world coordinate system, the mounting position of the first CMOS camera 23 in the Z-axis direction is adjusted so that the shooting field of view is perpendicularly and perpendicularly seen through the XOY plane of the world coordinate system, and the pixel distances from the left and right edges of the a plane of the notebook computer to the left and right edges of the shooting field of view of the first CMOS camera 23 in the frame of the shooting field of view of the first CMOS camera 23 are c 1 The pixel distance from the upper and lower edges of the A surface of the notebook computer to the upper and lower edges of the shooting view field of the first CMOS camera 23 is d 1 And satisfies:
Figure BDA0003225066760000071
the second CMOS camera 26 takes a field of view of L 2 ×W 2 ；
When the notebook computer is positioned at the original point O position of the world coordinate system, the second CMOS camera 26 is adjusted to be positioned in the positive directions of the Y axis and the Z axis of the world coordinate system, so that the shooting view field of the second CMOS camera is vertically and positively seen on the XOZ plane of the world coordinate system; and the pixel distances from the left and right edges of the B surface of the notebook computer in the picture of the field of view shot by the second CMOS camera 26 to the left and right edges of the field of view shot by the second CMOS camera 26 are c 2 The pixel distance from the upper and lower edges of the B surface of the notebook computer to the upper and lower edges of the shooting view field of the second CMOS camera 26 is d 2 And satisfies:
Figure BDA0003225066760000072
the industrial personal computer is in signal transmission with the lower computer control module through serial port communication, and the lower computer control module is respectively connected with the first conveyor belt 1, the second conveyor belt 2, the flip mechanism 3 and the jacking rotating mechanism 7 and used for controlling the operation of the device; the industrial personal computer classifies and fuses the images acquired by the first image acquisition module 21 and the second image acquisition module 22 through an image processing algorithm to obtain the image information of the omnibearing notebook shell, so that the image information of the omnibearing notebook shell is classified and identified in an image characteristic-defect mode by adopting a deep learning model, and a convolutional neural network, an improved VGG19 deep learning model and the like can be adopted according to the characteristics and the defect types of the images;
in this embodiment, as shown in fig. 2, an omnibearing high-precision detection method for detecting defects of a notebook computer casing is performed according to the following steps:
step 1: the notebook computer to be detected is horizontally placed on the first conveyor belt 1 after being closed, and the opening and closing end is along the conveying direction of the conveyor belt and is on the same side as the push rod assembly 29; the push rod assembly 29 adjusts the position of the notebook computer on the first conveyor belt 1 under the action of the push rod air cylinder 30, so that the push rod assembly 29 is reset after the notebook computer is positioned at the middle position on the first conveyor belt 1; the notebook computer is moved to the middle position of the conveying belt, so that the notebook computer can be accurately moved to the detection station;
step 2: the first image acquisition module 21 acquires the digital image on the detection station in real time and sends the digital image to the industrial personal computer; the industrial personal computer carries out feature point identification on the received digital image according to the position information and the number of the pre-stored feature points on the surface of the notebook computer, if the number of the identified feature points is matched with the number of the pre-stored feature points, the position deviation value of the feature points in the digital image is calculated according to the pre-stored feature point position information, if the position deviation value is smaller than a set threshold value, the notebook computer to be detected is judged to reach a designated position, and an outage signal is sent to a lower computer control module so as to control the first conveyor belt 1 and the second conveyor belt 2 to stop running;
step 3: after the industrial personal computer receives the shutdown completion signal fed back by the lower computer control module, the first image acquisition module 21 is utilized to acquire an A-plane image of the notebook computer to be detected;
step 4: after the acquisition of the A-plane image is completed, the industrial personal computer sends an acquisition signal to a lower computer control module, and the acquisition signal is used for controlling the flip mechanism 3 and the bottom jacking and rotating mechanism 7 to act, so that the bottom jacking and rotating mechanism 7 performs jacking action to drive the second vacuum chuck group 20 to suck the D-plane of the notebook computer, and meanwhile, the flip mechanism 3 drives the first vacuum chuck group 5 to suck the A-plane of the notebook computer, and then 90-degree flip action is performed to open the notebook computer;
step 5: after the flip action is completed, the first vacuum chuck group 5 loosens the A surface of the notebook computer, the lower computer control module controls the jacking rotating mechanism 7 to rotate, and the rotating mechanism 9 is set to drive the notebook computer to stop after rotating for a delta angle through the second servo motor 15, as shown in fig. 7 and 8, so that the industrial personal computer controls the first image acquisition module 21 to acquire C surface images of the notebook computer at a stopping interval, controls the second image acquisition module 22 to acquire A, B surface images of the notebook computer until the acquisition of A, B, C surface images is completed after the notebook computer rotates for one circle, and then the images are sent to the industrial personal computer to be classified and fused, wherein the rotating angle delta can be adjusted at multiple angles of 30 degrees, 60 degrees, 90 degrees and the like preliminarily according to the size and the number of the defects of the shell of the notebook computer;
step 6: the lower computer control module controls the flip mechanism 3 to drive the first vacuum chuck group 5 to suck the A surface of the notebook computer, and then 90-degree closing action is performed to close the notebook computer;
step 7: after the lower computer control module controls the jacking rotating mechanism 7 to drive the notebook computer to rotate 180 degrees, as shown in fig. 9, the flip mechanism 3 is controlled to drive the first vacuum chuck group 5 to suck the surface A of the notebook computer and then turn over 90 degrees, so that the bottom of the notebook computer faces the second image acquisition module 22; so that the second image acquisition module acquires the D-plane image of the notebook computer;
step 8: after the acquisition of the D-plane image is completed, the industrial personal computer sends a reset signal to the lower computer control module to control the flip mechanism 3 to drive the first vacuum chuck group 5 to reversely flip 90 degrees, so that after the notebook computer is reset, an operation signal is sent to the lower computer control module to control the first conveyor belt 1 and the second conveyor belt 2 to resume operation;
step 9: the industrial personal computer carries out histogram equalization processing on the acquired image; then median filtering is carried out on the processed image; then the processed image is cut in proportion, so that a gray scale template based on an image pyramid is established and used for detecting defect characteristic areas, reducing defect matching search areas, rapidly positioning the positions of target defects, and facilitating the next step of manufacturing a defect data area sample set after collecting defect information;
the industrial personal computer corrects the positions and the directions of the A-side image, the B-side image and the C-side image of the notebook computer according to the gray template; merging the corrected images into one image and dividing the region of interest; thereby obtaining a sample set of the region of interest; and finally, establishing and optimizing a deep learning model based on the region-of-interest sample set, and finally classifying and identifying image feature-defect forms by using the deep learning model, wherein a convolutional neural network, an improved VGG19 deep learning model and the like can be adopted according to the features and the types of defects of the image.

Claims (3)

1. The omnibearing high-precision detection device for the defects of the notebook computer shell is characterized by comprising a conveying belt, wherein a first conveying belt (1) and a second conveying belt (2) are sequentially arranged on the conveying belt along the conveying direction;
a push rod assembly (29) is arranged on one side of the first conveyor belt (1), and a rubber block (33) is arranged at the movable end of a push rod cylinder (30) of the push rod assembly (29) and is used for adjusting the initial position of a notebook computer on the first conveyor belt (1) so that the notebook computer is positioned at the middle position on the first conveyor belt (1);
the same sides of the first conveying belt (1) and the second conveying belt (2) are provided with a flip mechanism (3), and the flip mechanism (3) is provided with a supporting plate (31) between the first conveying belt (1) and the second conveying belt (2); the support plate (31) is fixedly provided with a linkage assembly (32) through a support seat (6), the support end of the linkage assembly (32) is movably hinged with the support seat (6) through a hinge bolt, one end of the linkage assembly (32) is connected with the movable end of the rodless cylinder (4) on the support plate (31), and the other end of the linkage assembly (32) is provided with a first vacuum chuck group (5) for adsorbing the A surface of the notebook computer; the rodless cylinder (4) drives the linkage assembly (32) to act and drives the A surface of the notebook computer on the first vacuum chuck group (5) to perform flip action;
the lifting and rotating mechanism (7) is arranged at the bottom between the first conveying belt (1) and the second conveying belt (2), a first servo motor (8) is fixed on a mounting plate (13) at the bottom of the lifting and rotating mechanism (7), the first servo motor (8) is connected with a ball screw (11) through a first coupler (10), the ball screw (11) penetrates through a middle plate (14), and a top plate is arranged at the top of the ball screw (11); a rotating mechanism (9) is arranged on the middle plate (14), a guide rod (12) is arranged on the middle plate (14) at one side of the rotating mechanism (9), the bottom of the guide rod (12) is arranged on the mounting plate (13), and the top of the guide rod is arranged on the top plate; the first servo motor (8) drives the ball screw (11) to rotate and drives the rotating mechanism (9) on the middle plate (14) to move up and down along the ball screw (11) between the mounting plate (13) and the top plate;
the rotating mechanism (9) is characterized in that a second servo motor (15) is arranged in the middle of the middle plate (14), one end of a rotating shaft (17) is connected through a second coupler and a speed reducer (16), the other end of the rotating shaft (17) is connected with one end of a connecting shaft (18), a sucker mounting frame (19) is mounted at the other end of the connecting shaft (18), and second vacuum sucker groups (20) are mounted on the sucker mounting frame (19) and are uniformly distributed in a rectangular shape and used for adsorbing the D surface of a notebook computer; the second servo motor (15) drives the connecting shaft (17) to rotate and drives the notebook computer D surface on the second vacuum chuck group (20) to do rotary motion at the bottom between the first conveying belt (1) and the second conveying belt (2); and the second servo motor (15) adjusts the rotation speed of the rotation mechanism (9);
a detection station is arranged between the first conveying belt (1) and the second conveying belt (2), and a first image acquisition module (21) and a second image acquisition module (22) are arranged on the detection station;
the first image acquisition module (21) is characterized in that a first movable frame group (25) is arranged at the bottom of one side of the first conveyor belt (1), a first CMOS camera (23) and a first linear light source (24) are arranged on the first movable frame group (25), the first linear light source (24) is arranged below the first CMOS camera (23), the first CMOS camera (23) is arranged above the detection station, and the first CMOS camera (23) is connected with an industrial personal computer through a first image acquisition card; and adjusting imaging sharpness of the first CMOS camera (23) by adjusting brightness of the first linear light source (24);
the second image acquisition module (22) is characterized in that a second movable frame group (28) is arranged at the bottom of the other side of the first conveyor belt (1), the second CMOS camera (26) and a second linear light source (27) are arranged on the second movable frame group (28), the second linear light source (27) is arranged below the second CMOS camera (26), and the second CMOS camera (26) is positioned at the side edge of the detection station; the second CMOS camera (26) is connected with the industrial personal computer through a second image acquisition card; and adjusting the imaging sharpness of the second CMOS camera (26) by adjusting the brightness of the second linear light source (27);
the industrial personal computer is also in signal transmission with a lower computer control module through serial port communication, and the lower computer control module is respectively connected with the first conveyor belt (1), the second conveyor belt (2), the flip mechanism (3) and the jacking rotating mechanism (7) and used for controlling the operation of the device; the industrial personal computer classifies and fuses the images acquired by the first image acquisition module (21) and the second image acquisition module (22) through an image processing algorithm to obtain omnibearing notebook shell image information, and then uses a deep learning model to classify and identify image characteristics-defect forms of the notebook shell image information.
2. The omnibearing high-precision detection device for detecting defects of notebook computer shells according to claim 1, which is characterized in that:
taking the central position of the horizontal plane of the area between the first conveyor belt (1) and the second conveyor belt (2) as an origin O, taking the conveying direction along the first conveyor belt (1) as an X-axis direction, taking the direction parallel to the horizontal plane and perpendicular to the conveying direction of the first conveyor belt (1) as a Y-axis direction, and taking the vertical upward direction perpendicular to the horizontal plane as a Z-axis direction, so as to establish a world coordinate system;
let the size of the notebook computer at the detection station be a×b, and the picture size of the shooting field of view of the first CMOS camera (23) be L 1 ×W 1 ；
When the notebook computer is in the world seatWhen the origin point O of the standard system is positioned, the installation position of the first CMOS camera (23) in the Z-axis direction is adjusted so that the shooting view field is vertically and vertically ahead of the world coordinate system XOY surface, and the pixel distances from the left and right edges of the notebook computer A surface to the left and right edges of the shooting view field of the first CMOS camera (23) in the picture of the shooting view field of the first CMOS camera (23) are c 1 The pixel distance from the upper and lower edges of the A surface of the notebook computer to the upper and lower edges of the shooting view field of the first CMOS camera (23) is d 1 And satisfies:
Figure FDA0003225066750000021
the second CMOS camera (26) shoots a field of view picture with a size L 2 ×W 2 ；
When the notebook computer is positioned at the original point O position of the world coordinate system, the second CMOS camera (26) is adjusted to be positioned in the positive directions of the Y axis and the Z axis of the world coordinate system, so that the shooting view field of the notebook computer is vertically and positively seen on the XOZ plane of the world coordinate system; and the pixel distances from the left and right edges of the B surface of the notebook computer in the picture of the visual field shot by the second CMOS camera (26) to the left and right edges of the visual field shot by the second CMOS camera (26) are all c 2 The pixel distance from the upper and lower edges of the B surface of the notebook computer to the upper and lower edges of the shooting view field of the second CMOS camera (26) is d 2 And satisfies:
Figure FDA0003225066750000031
3. the detection method of the omnibearing high-precision detection device for detecting the defects of the shell of the notebook computer according to claim 1, which is characterized by comprising the following steps:
step 1: the notebook computer to be detected is horizontally placed on a first running conveyor belt (1) after being closed, and the opening and closing end is along the conveying direction of the conveyor belt and is on the same side as the push rod assembly (29); the push rod assembly (29) adjusts the position of the notebook computer on the first conveyor belt (1) under the action of the push rod air cylinder (30), so that the push rod assembly (29) is reset after the notebook computer is positioned at the middle position on the first conveyor belt (1);
step 2: the first image acquisition module (21) acquires the digital image on the detection station in real time and sends the digital image to the industrial personal computer; the industrial personal computer identifies the characteristic points of the received digital image according to the position information and the number of the prestored characteristic points on the surface of the notebook computer, if the number of the identified characteristic points is matched with the number of the prestored characteristic points, the position deviation value of the characteristic points in the digital image is calculated according to the position information of the prestored characteristic points, if the position deviation value is smaller than the set threshold value, the notebook computer to be detected is judged to reach the designated position, and a shutdown signal is sent to the lower computer control module so as to control the first conveyor belt (1) and the second conveyor belt (2) to stop running;
step 3: after the industrial personal computer receives a shutdown completion signal fed back by the lower computer control module, acquiring an A-plane image of the notebook computer to be detected by using the first image acquisition module (21);
step 4: after the acquisition of the A-plane image is completed, the industrial personal computer sends an acquisition signal to a lower computer control module, and the acquisition signal is used for controlling the flip mechanism (3) and the bottom jacking rotating mechanism (7) to act, so that the bottom jacking rotating mechanism (7) performs jacking action to drive the second vacuum chuck group (20) to suck the D-plane of the notebook computer, and meanwhile, the flip mechanism (3) drives the first vacuum chuck group (5) to suck the A-plane of the notebook computer, and then 90-degree flip action is performed to open the notebook computer;
step 5: after the flip action is finished, the first vacuum chuck group (5) loosens the A surface of the notebook computer, the lower computer control module controls the jacking rotating mechanism (7) to rotate, the rotating mechanism (9) is set to drive the notebook computer to stop after rotating for delta angles through the second servo motor (15), so that the industrial personal computer controls the first image acquisition module (21) to acquire C surface images of the notebook computer in a stopping interval, and controls the second image acquisition module (22) to acquire A, B surface images of the notebook computer until A, B, C surface images are acquired after the notebook computer rotates for one circle and then sent to the industrial personal computer to be classified and fused;
step 6: the lower computer control module controls the flip mechanism (3) to drive the first vacuum chuck group (5) to suck the A surface of the notebook computer, and then 90-degree closing action is performed to close the notebook computer;
step 7: the lower computer control module controls the jacking rotating mechanism (7) to drive the notebook computer to rotate 180 degrees, and then controls the flip mechanism (3) to drive the first vacuum chuck group (5) to suck the A surface of the notebook computer and then turn over 90 degrees, so that the bottom of the notebook computer faces the second image acquisition module (22); the second image acquisition module acquires a D-plane image of the notebook computer;
step 8: after the acquisition of the D-plane image is completed, the industrial personal computer sends a reset signal to the lower computer control module so as to control the flip mechanism (3) to drive the first vacuum chuck group (5) to reversely flip by 90 degrees, so that after the notebook computer is reset, an operation signal is sent to the lower computer control module so as to control the first conveyor belt (1) and the second conveyor belt (2) to resume operation;
step 9: the industrial personal computer carries out histogram equalization processing and median filtering processing on the acquired image, and then cuts the processed image according to a proportion, so as to establish a gray scale template based on an image pyramid;
the industrial personal computer corrects the positions and the directions of the A-plane image, the B-plane image and the C-plane image of the notebook computer according to the gray template; and combining the corrected images into an image and dividing the region of interest to obtain a region of interest sample set, wherein the region of interest sample set is used for training and optimizing a deep learning model to realize classification and identification of image feature-defect forms.
CN202110968422.8A 2021-08-23 2021-08-23 Omnibearing high-precision detection device and method for defects of notebook computer shell Active CN113670923B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110968422.8A CN113670923B (en) 2021-08-23 2021-08-23 Omnibearing high-precision detection device and method for defects of notebook computer shell

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110968422.8A CN113670923B (en) 2021-08-23 2021-08-23 Omnibearing high-precision detection device and method for defects of notebook computer shell

Publications (2)

Publication Number Publication Date
CN113670923A CN113670923A (en) 2021-11-19
CN113670923B true CN113670923B (en) 2023-06-27

Family

ID=78545122

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110968422.8A Active CN113670923B (en) 2021-08-23 2021-08-23 Omnibearing high-precision detection device and method for defects of notebook computer shell

Country Status (1)

Country Link
CN (1) CN113670923B (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115684208B (en) * 2023-01-04 2023-03-10 广州弘高科技股份有限公司 Automatic feeding and discharging control system of AVI detection machine
CN115993363B (en) * 2023-02-20 2023-11-17 苏州天准科技股份有限公司 Detection device for side edge of notebook computer
CN116223500B (en) * 2023-02-21 2023-10-27 苏州天准科技股份有限公司 Notebook computer appearance detection equipment and detection method

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH0868763A (en) * 1994-08-31 1996-03-12 Sanyo Electric Co Ltd Apparatus and method for confirming image
WO2017088469A1 (en) * 2015-11-27 2017-06-01 武汉精测电子技术股份有限公司 High-precision automatic optical detection system and method based on mechanical arm
CN109580641A (en) * 2018-12-12 2019-04-05 江苏大学 A kind of notebook computer panel automatic detection device

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH0868763A (en) * 1994-08-31 1996-03-12 Sanyo Electric Co Ltd Apparatus and method for confirming image
WO2017088469A1 (en) * 2015-11-27 2017-06-01 武汉精测电子技术股份有限公司 High-precision automatic optical detection system and method based on mechanical arm
CN109580641A (en) * 2018-12-12 2019-04-05 江苏大学 A kind of notebook computer panel automatic detection device

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
基于AOI的电阻应变计检测系统;路新科;;电子设计工程(第08期);全文 *
安全壳缺陷检测系统;张曙光;郑鹏尚;赵灿松;王春苗;;仪器仪表用户(第07期);全文 *

Also Published As

Publication number Publication date
CN113670923A (en) 2021-11-19

Similar Documents

Publication Publication Date Title
CN113670923B (en) Omnibearing high-precision detection device and method for defects of notebook computer shell
CN105759720B (en) Manipulator track and localization online recognition and method for correcting error based on computer vision
CN111729869B (en) Machine vision analysis's car power device detecting system
CN109821763B (en) Fruit sorting system based on machine vision and image identification method thereof
CN112845194B (en) Defect detecting apparatus and detecting method thereof
CN112697804A (en) Mobile phone middle frame visual detection system and detection method thereof
CN114252452A (en) Online detection device and method for appearance defects and contour dimension of small-sized revolving body
CN113670934B (en) Omnibearing rapid detection device and method for defects of notebook computer shell
CN111307812A (en) Welding spot appearance detection method based on machine vision
CN109085179A (en) A kind of board surface flaw detection device and detection method
CN108759676B (en) Checkerboard-based transmission case end face large-size form and position tolerance detection device and method
US11327028B2 (en) Ceramic ball automatic sorting system and method
CN116797569A (en) Wafer defect detection method and system
CN111323425A (en) Multi-camera visual detection device and method
CN111203389A (en) Visual detection device and detection method
CN215416732U (en) Visual inspection system of ceramic substrate
CN109013388B (en) Automatic sorting system and method for ceramic balls
CN208721584U (en) A kind of board surface flaw detection device
CN220690817U (en) Appearance process measuring device for liquid crystal display screen
CN117078674B (en) Intelligent PCBA appearance defect detection method and detection system
CN113640318B (en) Four mirror surface intellectual detection system equipment of monocular
CN112819824B (en) Neural network for vision inspection system, inspection system and method including the same
CN117030714B (en) Mobile phone middle frame appearance detection mechanism and detection method
CN220160598U (en) OLED module AOI detection equipment
CN114136987B (en) Device and method for detecting deformation defect of lens

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant