Unlock instant, AI-driven research and patent intelligence for your innovation.

Industrial detection adaptive classification method and device

An adaptive classification and industrial technology, applied in measuring devices, sorting, optical instrument testing, etc., can solve problems such as inability to quickly adjust segmentation parameters and reference parameters, and achieve the effect of saving manpower and improving efficiency

Active Publication Date: 2019-02-22
BEIJING LUSTER LIGHTTECH
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The purpose of this application is to provide an industrial detection adaptive classification method and device to solve the problem that the prior art cannot quickly adjust segmentation parameters and reference parameters

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
  • Industrial detection adaptive classification method and device
  • Industrial detection adaptive classification method and device
  • Industrial detection adaptive classification method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0071] figure 1 is the imaging image of the LCD defect, figure 2 is the defect image extracted with a segmentation threshold of 157, image 3 In order to segment the defect image extracted with a threshold of 167, it can be seen that the difference in the segmentation threshold will have a relatively large impact on the feature extraction of the defect area. In order to solve the problem that different segmentation parameters and benchmark parameters have a huge impact on the final detection results of defects, and the parameter adjustment work brings time-consuming and labor-intensive problems.

[0072] This application provides an adaptive classification method for industrial detection, see Figure 4 , the method includes:

[0073] Step S100, acquiring images of test samples, the test samples include defective products, good products, passed inspection products and missed inspection products;

[0074] Before determining the test samples, it is necessary to determine the...

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

No PUM Login to View More

Abstract

The invention discloses an industrial detection adaptive classification method and device. A segmentation parameter feature extraction range and a reference parameter optimization range can be presetaccording to the method. All segments are automatically traversed according to the given segmentation parameter feature extraction range, formation reference parameter information is recorded into a document, and then an optimal segmentation line is searched according to automatic search and traversal of the reference parameter optimization range. According to the method, a computer automaticallyclassifies, the manpower is greatly saved, the efficiency is improved, and the parameter full traversal optimization in the case of large batches of samples is supported.

Description

technical field [0001] The present application relates to the technical field of classification methods, in particular to an adaptive classification method and device for industrial detection. Background technique [0002] In the LCD industry, there are two main difficulties in the detection of defects. One is the problem of defect imaging, which is mainly due to the design of the optical path; the other is the detection problem of "imaging defects". For the sake of convenience, this patent refers to dust, foreign matter, etc. as interference objects, and "imaging defects" as defects. [0003] For the detection of "defects", traditional machine vision algorithms often use image processing methods such as spatial filtering, image segmentation, edge extraction, morphological operations, and Blob analysis to extract suspicious targets, and there are defects and interference in suspicious targets; then calculate The area, energy, gray value, size, aspect ratio and other multi-d...

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): G01M11/00G01N21/94B07C5/342B07C5/02B07C5/00
CPCB07C5/00B07C5/02B07C5/3422G01M11/00G01N21/94
Inventor 王太兴姚毅解三霞路建伟马增婷
Owner BEIJING LUSTER LIGHTTECH