An Accurate Deep Key-Point Prediction Model With Low-Level Texture Refinement and High-Level Semantic Enhancement for Bolt Vertex Detection in Industrial Machine Systems

The safety status of intelligent monitoring bolts is crucial for ensuring the stability of industrial building and structural systems. Considering the complexity of the industrial production environment, the collected bolt images may have diverse and complex features, which will seriously affect the...

Full description

Saved in:
Bibliographic Details
Main Authors: Jiaqi Liu, Yingbo Wang, Mingyue Lang, Fengyuan Zuo
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10909128/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850093113344786432
author Jiaqi Liu
Yingbo Wang
Mingyue Lang
Fengyuan Zuo
author_facet Jiaqi Liu
Yingbo Wang
Mingyue Lang
Fengyuan Zuo
author_sort Jiaqi Liu
collection DOAJ
description The safety status of intelligent monitoring bolts is crucial for ensuring the stability of industrial building and structural systems. Considering the complexity of the industrial production environment, the collected bolt images may have diverse and complex features, which will seriously affect the recognition accuracy of bolt vertices and easily lead to false alarms and missed alarms of bolt looseness. To address these challenges, this paper proposes a bolt key-point detection method guided by masked-based dual-views model with texture refinement and semantic enhancement. Firstly, inspired by multimodal technology, a dual-views deep fusion method is designed based on the original and relief views to fully utilize their respective advantages to obtain robust key features. Secondly, low-level texture refinement and high-level semantic enhancement modules are designed to improve the edge texture and high-level semantic details of the bolt area, respectively. Finally, we established the gradient update of the above network for key-point recognition loss, target detection loss and cross entropy loss. In addition, we introduced a masked-based unsupervised pre-training paradigm based on convolutional structure to enhance the feature representation ability of the above model. In experiments and discussions, we analyzed the effectiveness of the proposed method in bolt vertex recognition tasks and achieved 0.987 AP and 0.743 Acc. In real-world application cases, a large number of experimental results have demonstrated the necessity of the proposed method.
format Article
id doaj-art-da33b7fd9b3f4c119abace083432201d
institution DOAJ
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-da33b7fd9b3f4c119abace083432201d2025-08-20T02:41:58ZengIEEEIEEE Access2169-35362025-01-0113482264823810.1109/ACCESS.2025.354733210909128An Accurate Deep Key-Point Prediction Model With Low-Level Texture Refinement and High-Level Semantic Enhancement for Bolt Vertex Detection in Industrial Machine SystemsJiaqi Liu0Yingbo Wang1Mingyue Lang2Fengyuan Zuo3https://orcid.org/0000-0001-9642-5705Ansteel Group Mining Company Ltd., Anshan, ChinaEnergy Control Center, Ansteel Group Mining Company Ltd., Anshan, ChinaAnsteel Group Mining Company Ltd., Anshan, ChinaCollege of Information Science and Engineering, Northeastern University, Shenyang, ChinaThe safety status of intelligent monitoring bolts is crucial for ensuring the stability of industrial building and structural systems. Considering the complexity of the industrial production environment, the collected bolt images may have diverse and complex features, which will seriously affect the recognition accuracy of bolt vertices and easily lead to false alarms and missed alarms of bolt looseness. To address these challenges, this paper proposes a bolt key-point detection method guided by masked-based dual-views model with texture refinement and semantic enhancement. Firstly, inspired by multimodal technology, a dual-views deep fusion method is designed based on the original and relief views to fully utilize their respective advantages to obtain robust key features. Secondly, low-level texture refinement and high-level semantic enhancement modules are designed to improve the edge texture and high-level semantic details of the bolt area, respectively. Finally, we established the gradient update of the above network for key-point recognition loss, target detection loss and cross entropy loss. In addition, we introduced a masked-based unsupervised pre-training paradigm based on convolutional structure to enhance the feature representation ability of the above model. In experiments and discussions, we analyzed the effectiveness of the proposed method in bolt vertex recognition tasks and achieved 0.987 AP and 0.743 Acc. In real-world application cases, a large number of experimental results have demonstrated the necessity of the proposed method.https://ieeexplore.ieee.org/document/10909128/Masked-based unsupervised pre-trainingdual-views deep fusionkey-point detectiontexture refinementsemantic enhancement
spellingShingle Jiaqi Liu
Yingbo Wang
Mingyue Lang
Fengyuan Zuo
An Accurate Deep Key-Point Prediction Model With Low-Level Texture Refinement and High-Level Semantic Enhancement for Bolt Vertex Detection in Industrial Machine Systems
IEEE Access
Masked-based unsupervised pre-training
dual-views deep fusion
key-point detection
texture refinement
semantic enhancement
title An Accurate Deep Key-Point Prediction Model With Low-Level Texture Refinement and High-Level Semantic Enhancement for Bolt Vertex Detection in Industrial Machine Systems
title_full An Accurate Deep Key-Point Prediction Model With Low-Level Texture Refinement and High-Level Semantic Enhancement for Bolt Vertex Detection in Industrial Machine Systems
title_fullStr An Accurate Deep Key-Point Prediction Model With Low-Level Texture Refinement and High-Level Semantic Enhancement for Bolt Vertex Detection in Industrial Machine Systems
title_full_unstemmed An Accurate Deep Key-Point Prediction Model With Low-Level Texture Refinement and High-Level Semantic Enhancement for Bolt Vertex Detection in Industrial Machine Systems
title_short An Accurate Deep Key-Point Prediction Model With Low-Level Texture Refinement and High-Level Semantic Enhancement for Bolt Vertex Detection in Industrial Machine Systems
title_sort accurate deep key point prediction model with low level texture refinement and high level semantic enhancement for bolt vertex detection in industrial machine systems
topic Masked-based unsupervised pre-training
dual-views deep fusion
key-point detection
texture refinement
semantic enhancement
url https://ieeexplore.ieee.org/document/10909128/
work_keys_str_mv AT jiaqiliu anaccuratedeepkeypointpredictionmodelwithlowleveltexturerefinementandhighlevelsemanticenhancementforboltvertexdetectioninindustrialmachinesystems
AT yingbowang anaccuratedeepkeypointpredictionmodelwithlowleveltexturerefinementandhighlevelsemanticenhancementforboltvertexdetectioninindustrialmachinesystems
AT mingyuelang anaccuratedeepkeypointpredictionmodelwithlowleveltexturerefinementandhighlevelsemanticenhancementforboltvertexdetectioninindustrialmachinesystems
AT fengyuanzuo anaccuratedeepkeypointpredictionmodelwithlowleveltexturerefinementandhighlevelsemanticenhancementforboltvertexdetectioninindustrialmachinesystems
AT jiaqiliu accuratedeepkeypointpredictionmodelwithlowleveltexturerefinementandhighlevelsemanticenhancementforboltvertexdetectioninindustrialmachinesystems
AT yingbowang accuratedeepkeypointpredictionmodelwithlowleveltexturerefinementandhighlevelsemanticenhancementforboltvertexdetectioninindustrialmachinesystems
AT mingyuelang accuratedeepkeypointpredictionmodelwithlowleveltexturerefinementandhighlevelsemanticenhancementforboltvertexdetectioninindustrialmachinesystems
AT fengyuanzuo accuratedeepkeypointpredictionmodelwithlowleveltexturerefinementandhighlevelsemanticenhancementforboltvertexdetectioninindustrialmachinesystems