Steel low-magnification loose image intelligent grading method

CN117152085BActive Publication Date: 2026-07-03JIANGSU YONGGANG GROUP CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
JIANGSU YONGGANG GROUP CO LTD
Filing Date
2023-09-01
Publication Date
2026-07-03

AI Technical Summary

Technical Problem

In existing technologies, the detection of low-magnification porosity in steel relies on manual grading. The results are greatly affected by human factors, have low accuracy, and cannot accurately reflect the internal quality of the steel, thus affecting production guidance and process optimization.

Method used

A low-magnification digital inspection system combined with image analysis and recognition methods is used to identify loose defects through deep learning and convolutional neural networks. The ResNeSt network model is used to label and rate loose defects, calculate the range of loose areas, the number of voids and their density, and achieve automated rating.

Benefits of technology

It provides more accurate porosity detection data, reduces the subjectivity and fatigue-induced bias of manual rating, improves the automation efficiency and accuracy of rating, and supports the optimization of production processes.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117152085B_ABST
    Figure CN117152085B_ABST
Patent Text Reader

Abstract

The application provides a steel low-magnification loose image intelligent grading method and relates to the technical field of image processing, and comprises the following steps: step 1: collecting a low-magnification structure image of a continuous casting round billet sample; step 2: identifying the loose defects of the low-magnification structure image; step 3: identifying the key features of the loose defects; and step 4: automatically grading the loose defects of the continuous casting round billet sample based on the key features of the loose defects. The application extracts and quantifies the loose low-magnification structure defect information by using the image recognition and image processing technology, realizes the automatic grading of the loose steel by combining the image, and can provide more abundant low-magnification structure information to realize the standardization and intelligentization of the steel low-magnification loose defect detection.
Need to check novelty before this filing date? Find Prior Art