A Defect Detection Method Based on Gradient Multi-threshold Optimization

A multi-threshold optimization and defect detection technology, applied in optical testing flaws/defects, etc., can solve problems such as inaccurate thresholds, complex algorithms, and long optimal thresholding iterations.

Inactive Publication Date: 2017-12-19
深圳市雅汇恒科技有限公司
View PDF3 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to provide a defect detection method based on gradient multi-threshold optimization, which is used to solve problems such as inaccurate thresholds, thresholds that cannot change with image fluctuations, too long iterations of optimal thresholding, and complex algorithms, etc., and has defect detection effects Good, fast extraction speed, etc.

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
  • A Defect Detection Method Based on Gradient Multi-threshold Optimization
  • A Defect Detection Method Based on Gradient Multi-threshold Optimization
  • A Defect Detection Method Based on Gradient Multi-threshold Optimization

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0042] In the production process, the wood defect detection system such as figure 1As shown, the upper two sides are the light source, the middle one is the acquisition camera, and the lower one is the conveyor belt. The system consists of a conveyor platform, industrial camera, image acquisition card, computer and defect detection software. The camera adopts the Guppy_PRO series industrial camera, which collects grayscale images, and the resolution can be adjusted freely. Since the images involved in defect detection are collected by black and white industrial cameras, they can be directly converted into digital images through the image acquisition card without considering the conversion of color space.

[0043] Such as figure 2 As shown, the present invention is based on a gradient multi-threshold optimized defect detection method, which specifically includes the following steps:

[0044] Step 1. Collect an image of the object to be detected, convert it into a digital ima...

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 present invention is a method for optimizing defect detection based on gradient multi-thresholds. Firstly, the optimized threshold is obtained through a simplified mean value clustering algorithm; secondly, 100 modules in each sample gradient image are counted through a normal distribution model, and the calculated Obtain a dynamic threshold; again, by dividing the sample image into blocks, and based on a statistical method, extract the maximum value of the pixel and the maximum value of the pixel difference in the module; finally, on the basis of modularization, judge through multiple thresholds and obtain the output The modules are combined into a complete image, and the median filter is performed on it to obtain the image of the defect detection result. The present invention improves the accuracy of the algorithm and reduces the time cost of the algorithm in the iterative process through the simplified mean value clustering algorithm; based on the statistics and normal distribution model, the edge is extracted in the gradient image, which significantly increases the accuracy and accuracy of the algorithm. Treatment effect: the invention can quickly and accurately detect the defects of wood, and improve the application range of detection and the quality of produced wood.

Description

technical field [0001] The invention relates to a defect detection method based on gradient multi-threshold optimization. Background technique [0002] In image segmentation processing, an image is divided into components that have a strong correlation with real-world objects or regions, which are mainly divided into the following three categories: global-based segmentation, edge-based segmentation, and local-based segmentation. At present, image segmentation is mainly used for the fine extraction of image edges. For the detection of strong edges, it mainly focuses on global segmentation and local segmentation, that is, thresholding segmentation. However, for some specific occasions, such as wood defect detection, etc., in addition to filtering In addition to removing the noise of the image itself, it is also necessary to remove the annual rings of the wood, etc., and retain the defective parts of the wood. Thresholding segmentation is an ideal method for background and obj...

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
Patent Type & Authority Patents(China)
IPC IPC(8): G01N21/88
Inventor 李俊高银
Owner 深圳市雅汇恒科技有限公司
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

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap