Transparent plate defect detection device based on machine vision and detection method thereof

A transparent plate and defect detection technology, applied in measuring devices, instruments, scientific instruments, etc., can solve the problems of missed detection efficiency, false detection of defects on the surface of transparent plates, etc., to improve efficiency, improve detection accuracy, and save labor. Effect

Pending Publication Date: 2019-11-22
TAIYUAN UNIVERSITY OF SCIENCE AND TECHNOLOGY
View PDF0 Cites 16 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] In order to overcome the deficiencies in the prior art and solve the technical problems of false detection, missed detection and low efficiency in the detection of internal defects and surf...

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Transparent plate defect detection device based on machine vision and detection method thereof
  • Transparent plate defect detection device based on machine vision and detection method thereof
  • Transparent plate defect detection device based on machine vision and detection method thereof

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0037] Such as figure 1 with figure 2 A machine vision-based transparent plate defect detection device shown includes an industrial camera 1, an industrial lens 13, an industrial light source 2, a stepping motor and a dark box 12, wherein:

[0038] The bottom of the dark box 12 is respectively provided with a feed port and a discharge port according to the conveying direction of the conveyor belt 10. The conveyor belt 10 runs through the feed port and the discharge port of the dark box 12 and extends to the outside of the dark box 12. The transparent plate to be tested 11 placed on the conveyor belt 10;

[0039] In said obscura 12:

[0040] A door-shaped hanger 3 is arranged across the top of the conveyor belt 10, the industrial light source 2 is arranged at the root of the column of the door-shaped hanger 3, the industrial light source 2 is set as a line light source, and the beam direction is parallel to the surface of the transparent plate 11 to be tested , the light be...

Embodiment 2

[0057] The structure of a machine vision-based transparent plate defect detection device in Embodiment 2 is the same as that in Embodiment 1, and will not be repeated here.

[0058] In Example 2, the device and detection method provided by the present invention are applied to the detection of surface scratches, internal cracks, gravel and air bubbles on a transparent PEO plastic plate with a thickness of 4mm. Industrial lens, the object distance is 241.55mm, the field of view is 379.27mm*200mm, the exposure time is 4.6ms, and the captured image resolution is 4096×2160.

[0059] The selection range of the binarization noise reduction threshold is 46, the low-pass Gaussian filter uses a filter template of 15×15, and the selection range of the binarization enhancement threshold is 49.

[0060] The measured scratch width is 6.5mm, with a relative error of 3.2%; the scratch length is 42.7mm, with a relative error of 4.1%; the bubble width is 3.5mm, with a relative error of 3.3%; th...

Embodiment 3

[0062] The structure of a machine vision-based transparent plate defect detection device in Embodiment 3 is the same as that in Embodiment 1, and will not be repeated here.

[0063] In this embodiment 3, the device and detection method provided by the present invention are applied to the detection of air holes, surface scratches, internal cracks, gravel and air bubbles on a 1mm thick transparent PYREX glass plate, and a black and white industrial camera with 8.9 million pixels is used, and the focal length is selected It is a 25mm industrial lens, the object distance is 268.38mm, the field of view is 151.7mm*80mm, the exposure time is 4.2ms, and the resolution of the collected image is 4096×2160. The selection range of the binarization noise reduction threshold is 47, the low-pass Gaussian filter uses a filter template of 15×15, and the selection range of the binarization enhancement threshold is 51.

[0064] The measured air hole width is 6.0mm, with a relative error of 2.4%;...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

PropertyMeasurementUnit
Heightaaaaaaaaaa
Lengthaaaaaaaaaa
Widthaaaaaaaaaa
Login to view more

Abstract

The invention discloses a transparent plate defect detection device based on machine vision and a detection method thereof, belongs to the technical field of intelligent defect detection of transparent materials, and solves the technical problems of false detection, missing detection and low efficiency in detection of internal defects and surface defects of transparent plates. According to the technical scheme, the device comprises a detection system composed of an industrial camera, an industrial lens, an industrial light source and a stepping motor, the detection system is placed in a cameraobscura, the light source is installed on the side surfaces of a conveying belt and a detected material, and the industrial camera is installed at a vertical position above the detected material forshooting. The detection system adopts a computer to realize Ethernet communication with the industrial camera to obtain an image, and processes the image by adopting a method of first binarization noise reduction, then Gaussian filtering and binarization enhancement to obtain contour information. The defect size information and the defect area are calculated in a mode of comparing by adopting a shooting graduated scale. According to the invention, the labor can be saved, the cost is reduced, the detection accuracy is improved, and the size detection error range is 0.5-5.0%.

Description

technical field [0001] The invention belongs to the technical field of defect intelligent detection of transparent materials, and relates to a machine vision-based transparent plate defect detection device and its detection method, in particular to a transparent glass material defect based on machine vision, lighting technology and image processing Detection method. Background technique [0002] At this stage, glass-based flat transparent materials are being used more and more widely as raw materials. In order to make transparent materials more suitable for various working environments, material manufacturers turn to the production of plates with different colors, different light transmittances and surface shapes, which greatly increases the probability of material defects. Common defects include surface scratches , internal cracks, grit and air bubbles. It is difficult to accurately and quickly judge the size of defects generated in the process of producing transparent ma...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): G01N21/958
CPCG01N21/958
Inventor 薛彬李亚杰吴志生孟庆森刘栋茌文清
Owner TAIYUAN UNIVERSITY OF SCIENCE AND TECHNOLOGY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products