VIP member
Product details
Glass product quality inspection system
Glass products may produce quality defects such as bubbles and defects during the production process. In today's increasingly strict quality requirements and fierce market competition, the automatic inspection system for glass products that combines machine vision technology with motion control and network communication brings more than just a technological change to enterprises.
1、 System Overview
Schematic diagram of the quality inspection system, with the number of cameras configured according to actual needs
System purpose: Detection of non-conforming products
Testing object: Glass products
Defect objects: bubbles, scars, flaws, sand particles, stones, light distortion points, inclusions, etc
Maximum detection speed:230m/h
Detection width:2100mm (Adjustable)
Detection length: based on the length of the board (Adjustable)
Test result processing: The alarm light is loud; Display defect patterns;
Optional labeling machine
2、 System solution
Connection diagram of quality inspection system
1High performance linear scanning camera
Camera field of view selectionFov = 500mmMaximum line speed11kHzData transmission rate3×25MHzSuitable for the detection of roll paper.
Feature resolution:0.72mm2；
Number of pixels for the smallest feature:FpX =FpY=3 pixels；
Spatial resolution (distance between pixel centers mapped to the scene):RsX＝0.2mm/pixel，RsY＝0.4mm/pixel；
Image resolution: width direction of each cameraRi = 2048 pixels；
The fastest scanning speed of the camera:Ts =82μs/scan.
2Lower level computer
Linear array camera1.1one million time/The image information collected at a scanning speed of every second is captured through the high-speed multi-channel image acquisition card (high-performance acquisition card, with an acquisition speed of up to) of the lower computer160MB/s, with trigger input, exposure time control output, digitalI/O）Transferred to the corresponding lower level industrial control computer, with a data transmission capacity of up to150M/SThe lower level industrial control computer relies on pre written underlying software to achieve online image positioning, recognition, tracking, and detection, and transmits the detection result data to the upper level computer.
3Host
The upper computer that implements operations and result processing. Implement the operation of various system functions, glass product detection settings, glass product records and defect history management, and output processing results, including alarm, display, labeling, and printing detection information
4Processing of test results (labeling machine)
The host alarm light is on, the alarm sound is emitted, and the defect location is displayed; The alarm light at the receiving station is loud, assisting workers in removing waste materials; The labeling machine operates to post labels on defective products; Printer prints statistical information.
5Synchronous sensors
Install pulse encoder online, which generates continuous acquisition trigger signals.
6Light source
HFfluorescent lamp65W*4（30KHZ）
Brightness control: software automatic control
service life:6000Hour (replaced every six months)
3、 System functions
The system software isWindowXPChinese version
1Power on self-test
Check the connection of the camera;
Network connection;
Initialization work
2Set product parameters
Batch number, name, inspector, inspection time;
The width of the test substance;
The length of the test substance;
The number of columns for detecting the substance.
3Set defect detection level
Testing levels: high, standard, medium, and low (or graded according to the factory's requirements).
4Loading template images
5Set detection mode and range
Set the detection mode and range of each camera separately, and divide the overall coordinate system into6A section, corresponding to6Separate and test the products, and establish a corresponding database.
6Testing begins
Point like defects such as bubbles, sand particles, and stones are detected usingBlobAlgorithms are used for searching, highlighting features such as inclusions, tin deposits, and light distortion points. Linear defect detection can be identified using line detection or projection techniques0.12MMPattern deviation and color difference.
The system can quickly and accurately distinguish and identify common defects in the factory, such as pattern defects, bubbles, sand particles, stones, tin stains, light distortion points, inclusions, and glass integrity.
4、 System detection result output scheme
1The alarm emits different silver alarms, with yellow and green lights flashing in real-time;
2Use grayscale display to statically display defect patterns and their positions throughout the entire image, with the defect areas flashing with red blocks. Save and retrieve historical defect images;
3When the defective glass arrives at the receiving station (column by column), the red light flashes and the buzzer sounds to alert the worker of the defective product information. Alarm delay in100～180m/minAdjustable between speeds, with a speed of130m/minIt can be precise to the stack.
4The labeling machine operates based on control signals and attaches identification labels to the edges of defective sheets. (Note: This is an optional configuration)
5For each column of small roll products, separate the defect sheet database and print the defective information for each roll using a printer.
Online inquiry
-
Contacts
-
Company
-
Telephone
-
Email
-
WeChat
-
Verification Code
-
Message Content
-