Two-stage hybrid algorithm for recognition of industrial slab numbers with data quality improvement
Abstract As the unique recognition of each slab, the accurate recognition of slab number is especially critical for the hot rolling production process. However, the collected data are often of low quality due to poor production environment conditions, making traditional deep learning algorithms face...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Springer
2022-12-01
|
Series: | Complex & Intelligent Systems |
Subjects: | |
Online Access: | https://doi.org/10.1007/s40747-022-00933-0 |
_version_ | 1797806474605363200 |
---|---|
author | Qingqing Liu Xianpeng Wang Xiangman Song |
author_facet | Qingqing Liu Xianpeng Wang Xiangman Song |
author_sort | Qingqing Liu |
collection | DOAJ |
description | Abstract As the unique recognition of each slab, the accurate recognition of slab number is especially critical for the hot rolling production process. However, the collected data are often of low quality due to poor production environment conditions, making traditional deep learning algorithms face more significant challenges in slab numbers recognition. In this paper, a two-stage hybrid algorithm based on convolutional neural network and Transformer is proposed to identify industrial slab numbers. In the first stage, an improved CycleGAN (HybridCy) is developed to enhance the quality of real-world unpaired data. In the second stage, a multi-scale hybrid vision transformer model (MSHy-Vit) is proposed to identify slab numbers of the improved data output of stage one. The experimental results on industrial slab data show that HybridCy exhibits stable and efficient performance. Even for low-quality data with severe geometric distortion, HybridCy can accomplish quality improvement, which can help to improve recognition accuracy. In addition, the MSHy-Vit achieves superior accuracy in the recognition of slab numbers in comparison to existing methods in the literature. |
first_indexed | 2024-03-13T06:07:52Z |
format | Article |
id | doaj.art-edfa2878944a4b1cbc73153a3aa7cec7 |
institution | Directory Open Access Journal |
issn | 2199-4536 2198-6053 |
language | English |
last_indexed | 2024-03-13T06:07:52Z |
publishDate | 2022-12-01 |
publisher | Springer |
record_format | Article |
series | Complex & Intelligent Systems |
spelling | doaj.art-edfa2878944a4b1cbc73153a3aa7cec72023-06-11T11:29:35ZengSpringerComplex & Intelligent Systems2199-45362198-60532022-12-01933367338410.1007/s40747-022-00933-0Two-stage hybrid algorithm for recognition of industrial slab numbers with data quality improvementQingqing Liu0Xianpeng Wang1Xiangman Song2National Frontiers Science Center for Industrial Intelligence and Systems Optimization, Key Laboratory of Data Analytics and Optimization for Smart Industry (Northeastern University), Ministry of EducationLiaoning Engineering Laboratory of Data Analytics and Optimization for Smart Industry, Liaoning Key Laboratory of Manufacturing System and Logistics OptimizationNational Frontiers Science Center for Industrial Intelligence and Systems OptimizationAbstract As the unique recognition of each slab, the accurate recognition of slab number is especially critical for the hot rolling production process. However, the collected data are often of low quality due to poor production environment conditions, making traditional deep learning algorithms face more significant challenges in slab numbers recognition. In this paper, a two-stage hybrid algorithm based on convolutional neural network and Transformer is proposed to identify industrial slab numbers. In the first stage, an improved CycleGAN (HybridCy) is developed to enhance the quality of real-world unpaired data. In the second stage, a multi-scale hybrid vision transformer model (MSHy-Vit) is proposed to identify slab numbers of the improved data output of stage one. The experimental results on industrial slab data show that HybridCy exhibits stable and efficient performance. Even for low-quality data with severe geometric distortion, HybridCy can accomplish quality improvement, which can help to improve recognition accuracy. In addition, the MSHy-Vit achieves superior accuracy in the recognition of slab numbers in comparison to existing methods in the literature.https://doi.org/10.1007/s40747-022-00933-0Slab numbers recognitionData quality enhancementTransformerCycleGAN |
spellingShingle | Qingqing Liu Xianpeng Wang Xiangman Song Two-stage hybrid algorithm for recognition of industrial slab numbers with data quality improvement Complex & Intelligent Systems Slab numbers recognition Data quality enhancement Transformer CycleGAN |
title | Two-stage hybrid algorithm for recognition of industrial slab numbers with data quality improvement |
title_full | Two-stage hybrid algorithm for recognition of industrial slab numbers with data quality improvement |
title_fullStr | Two-stage hybrid algorithm for recognition of industrial slab numbers with data quality improvement |
title_full_unstemmed | Two-stage hybrid algorithm for recognition of industrial slab numbers with data quality improvement |
title_short | Two-stage hybrid algorithm for recognition of industrial slab numbers with data quality improvement |
title_sort | two stage hybrid algorithm for recognition of industrial slab numbers with data quality improvement |
topic | Slab numbers recognition Data quality enhancement Transformer CycleGAN |
url | https://doi.org/10.1007/s40747-022-00933-0 |
work_keys_str_mv | AT qingqingliu twostagehybridalgorithmforrecognitionofindustrialslabnumberswithdataqualityimprovement AT xianpengwang twostagehybridalgorithmforrecognitionofindustrialslabnumberswithdataqualityimprovement AT xiangmansong twostagehybridalgorithmforrecognitionofindustrialslabnumberswithdataqualityimprovement |