Small Defects Detection of Galvanized Strip Steel via Schatten-<i>p</i> Norm-Based Low-Rank Tensor Decomposition

Accurate and efficient white-spot defects detection for the surface of galvanized strip steel is one of the most important guarantees for the quality of steel production. It is a fundamental but “hard” small target detection problem due to its small pixel occupation in low-contrast images. By fully...

Full description

Saved in:
Bibliographic Details
Main Authors: Shiyang Zhou, Xuguo Yan, Huaiguang Liu, Caiyun Gong
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/25/8/2606
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849713830062456832
author Shiyang Zhou
Xuguo Yan
Huaiguang Liu
Caiyun Gong
author_facet Shiyang Zhou
Xuguo Yan
Huaiguang Liu
Caiyun Gong
author_sort Shiyang Zhou
collection DOAJ
description Accurate and efficient white-spot defects detection for the surface of galvanized strip steel is one of the most important guarantees for the quality of steel production. It is a fundamental but “hard” small target detection problem due to its small pixel occupation in low-contrast images. By fully exploiting the low-rank and sparse prior information of a surface defect image, a Schatten-<i>p</i> norm-based low-rank tensor decomposition (SLRTD) method is proposed to decompose the defect image into low-rank background, sparse defect, and random noise. Firstly, the original defect images are transformed into a new patch-based tensor mode through data reconstruction for mining valuable information of the defect image. Then, considering the over-shrinkage problem in the low-rank component estimation caused by a vanilla nuclear norm and a weighted nuclear norm, a nonlinear reweighting strategy based on a Schatten <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mi>p</mi></semantics></math></inline-formula>-norm is incorporated to improve the decomposition performance. Finally, a solution framework is proposed via a well-designed alternating direction method of multipliers to obtain the white-spot defect target image by a simple segmenting algorithm. The white-spot defect dataset from a real-world galvanized strip steel production line is constructed, and the experimental results demonstrate that the proposed SLRTD method outperforms existing state-of-the-art methods qualitatively and quantitatively.
format Article
id doaj-art-c357a1da277245efb7a3615a39cc7d3b
institution DOAJ
issn 1424-8220
language English
publishDate 2025-04-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj-art-c357a1da277245efb7a3615a39cc7d3b2025-08-20T03:13:51ZengMDPI AGSensors1424-82202025-04-01258260610.3390/s25082606Small Defects Detection of Galvanized Strip Steel via Schatten-<i>p</i> Norm-Based Low-Rank Tensor DecompositionShiyang Zhou0Xuguo Yan1Huaiguang Liu2Caiyun Gong3Key Laboratory of Metallurgical Equipment and Control Technology, Ministry of Education, Wuhan University of Science and Technology, Wuhan 430081, ChinaKey Laboratory of Metallurgical Equipment and Control Technology, Ministry of Education, Wuhan University of Science and Technology, Wuhan 430081, ChinaKey Laboratory of Metallurgical Equipment and Control Technology, Ministry of Education, Wuhan University of Science and Technology, Wuhan 430081, ChinaKey Laboratory of Metallurgical Equipment and Control Technology, Ministry of Education, Wuhan University of Science and Technology, Wuhan 430081, ChinaAccurate and efficient white-spot defects detection for the surface of galvanized strip steel is one of the most important guarantees for the quality of steel production. It is a fundamental but “hard” small target detection problem due to its small pixel occupation in low-contrast images. By fully exploiting the low-rank and sparse prior information of a surface defect image, a Schatten-<i>p</i> norm-based low-rank tensor decomposition (SLRTD) method is proposed to decompose the defect image into low-rank background, sparse defect, and random noise. Firstly, the original defect images are transformed into a new patch-based tensor mode through data reconstruction for mining valuable information of the defect image. Then, considering the over-shrinkage problem in the low-rank component estimation caused by a vanilla nuclear norm and a weighted nuclear norm, a nonlinear reweighting strategy based on a Schatten <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mi>p</mi></semantics></math></inline-formula>-norm is incorporated to improve the decomposition performance. Finally, a solution framework is proposed via a well-designed alternating direction method of multipliers to obtain the white-spot defect target image by a simple segmenting algorithm. The white-spot defect dataset from a real-world galvanized strip steel production line is constructed, and the experimental results demonstrate that the proposed SLRTD method outperforms existing state-of-the-art methods qualitatively and quantitatively.https://www.mdpi.com/1424-8220/25/8/2606galvanized strip steelsmall target detectionSchatten <i>p</i>-normtensor decomposition
spellingShingle Shiyang Zhou
Xuguo Yan
Huaiguang Liu
Caiyun Gong
Small Defects Detection of Galvanized Strip Steel via Schatten-<i>p</i> Norm-Based Low-Rank Tensor Decomposition
Sensors
galvanized strip steel
small target detection
Schatten <i>p</i>-norm
tensor decomposition
title Small Defects Detection of Galvanized Strip Steel via Schatten-<i>p</i> Norm-Based Low-Rank Tensor Decomposition
title_full Small Defects Detection of Galvanized Strip Steel via Schatten-<i>p</i> Norm-Based Low-Rank Tensor Decomposition
title_fullStr Small Defects Detection of Galvanized Strip Steel via Schatten-<i>p</i> Norm-Based Low-Rank Tensor Decomposition
title_full_unstemmed Small Defects Detection of Galvanized Strip Steel via Schatten-<i>p</i> Norm-Based Low-Rank Tensor Decomposition
title_short Small Defects Detection of Galvanized Strip Steel via Schatten-<i>p</i> Norm-Based Low-Rank Tensor Decomposition
title_sort small defects detection of galvanized strip steel via schatten i p i norm based low rank tensor decomposition
topic galvanized strip steel
small target detection
Schatten <i>p</i>-norm
tensor decomposition
url https://www.mdpi.com/1424-8220/25/8/2606
work_keys_str_mv AT shiyangzhou smalldefectsdetectionofgalvanizedstripsteelviaschattenipinormbasedlowranktensordecomposition
AT xuguoyan smalldefectsdetectionofgalvanizedstripsteelviaschattenipinormbasedlowranktensordecomposition
AT huaiguangliu smalldefectsdetectionofgalvanizedstripsteelviaschattenipinormbasedlowranktensordecomposition
AT caiyungong smalldefectsdetectionofgalvanizedstripsteelviaschattenipinormbasedlowranktensordecomposition