Crack-Detection Algorithm Integrating Multi-Scale Information Gain with Global–Local Tight–Loose Coupling

In this study, an improved target-detection model based on information theory is proposed to address the difficulties of crack-detection tasks, such as slender target shapes, blurred boundaries, and complex backgrounds. By introducing a multi-scale information gain mechanism and a global–local featu...

Full description

Saved in:
Bibliographic Details
Main Authors: Yun Bai, Zhiyao Li, Runqi Liu, Jiayi Feng, Biao Li
Format: Article
Language:English
Published: MDPI AG 2025-02-01
Series:Entropy
Subjects:
Online Access:https://www.mdpi.com/1099-4300/27/2/165
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849719281127784448
author Yun Bai
Zhiyao Li
Runqi Liu
Jiayi Feng
Biao Li
author_facet Yun Bai
Zhiyao Li
Runqi Liu
Jiayi Feng
Biao Li
author_sort Yun Bai
collection DOAJ
description In this study, an improved target-detection model based on information theory is proposed to address the difficulties of crack-detection tasks, such as slender target shapes, blurred boundaries, and complex backgrounds. By introducing a multi-scale information gain mechanism and a global–local feature coupling strategy, the model has significantly improved feature extraction and expression capabilities. Experimental results show that, on a single-crack dataset, the model’s mAP@50 and mAP@50-95 are 1.6% and 0.8% higher than the baseline model RT-DETR, respectively; on a multi-crack dataset, these two indicators are improved by 1.2% and 1.0%, respectively. The proposed method shows good robustness and detection accuracy in complex scenarios, providing new ideas and technical support for in-depth research in the field of crack detection.
format Article
id doaj-art-ac58babc3aaa4e889d20382102dfd4f7
institution DOAJ
issn 1099-4300
language English
publishDate 2025-02-01
publisher MDPI AG
record_format Article
series Entropy
spelling doaj-art-ac58babc3aaa4e889d20382102dfd4f72025-08-20T03:12:11ZengMDPI AGEntropy1099-43002025-02-0127216510.3390/e27020165Crack-Detection Algorithm Integrating Multi-Scale Information Gain with Global–Local Tight–Loose CouplingYun Bai0Zhiyao Li1Runqi Liu2Jiayi Feng3Biao Li4School of Aviation, Inner Mongolia University of Technology, Hohhot 010021, ChinaSchool of Aviation, Inner Mongolia University of Technology, Hohhot 010021, ChinaSchool of Aviation, Inner Mongolia University of Technology, Hohhot 010021, ChinaSchool of Aviation, Inner Mongolia University of Technology, Hohhot 010021, ChinaSchool of Aviation, Inner Mongolia University of Technology, Hohhot 010021, ChinaIn this study, an improved target-detection model based on information theory is proposed to address the difficulties of crack-detection tasks, such as slender target shapes, blurred boundaries, and complex backgrounds. By introducing a multi-scale information gain mechanism and a global–local feature coupling strategy, the model has significantly improved feature extraction and expression capabilities. Experimental results show that, on a single-crack dataset, the model’s mAP@50 and mAP@50-95 are 1.6% and 0.8% higher than the baseline model RT-DETR, respectively; on a multi-crack dataset, these two indicators are improved by 1.2% and 1.0%, respectively. The proposed method shows good robustness and detection accuracy in complex scenarios, providing new ideas and technical support for in-depth research in the field of crack detection.https://www.mdpi.com/1099-4300/27/2/165crack detectionmulti-scale information gainmutual information optimizationconditional entropy minimization
spellingShingle Yun Bai
Zhiyao Li
Runqi Liu
Jiayi Feng
Biao Li
Crack-Detection Algorithm Integrating Multi-Scale Information Gain with Global–Local Tight–Loose Coupling
Entropy
crack detection
multi-scale information gain
mutual information optimization
conditional entropy minimization
title Crack-Detection Algorithm Integrating Multi-Scale Information Gain with Global–Local Tight–Loose Coupling
title_full Crack-Detection Algorithm Integrating Multi-Scale Information Gain with Global–Local Tight–Loose Coupling
title_fullStr Crack-Detection Algorithm Integrating Multi-Scale Information Gain with Global–Local Tight–Loose Coupling
title_full_unstemmed Crack-Detection Algorithm Integrating Multi-Scale Information Gain with Global–Local Tight–Loose Coupling
title_short Crack-Detection Algorithm Integrating Multi-Scale Information Gain with Global–Local Tight–Loose Coupling
title_sort crack detection algorithm integrating multi scale information gain with global local tight loose coupling
topic crack detection
multi-scale information gain
mutual information optimization
conditional entropy minimization
url https://www.mdpi.com/1099-4300/27/2/165
work_keys_str_mv AT yunbai crackdetectionalgorithmintegratingmultiscaleinformationgainwithgloballocaltightloosecoupling
AT zhiyaoli crackdetectionalgorithmintegratingmultiscaleinformationgainwithgloballocaltightloosecoupling
AT runqiliu crackdetectionalgorithmintegratingmultiscaleinformationgainwithgloballocaltightloosecoupling
AT jiayifeng crackdetectionalgorithmintegratingmultiscaleinformationgainwithgloballocaltightloosecoupling
AT biaoli crackdetectionalgorithmintegratingmultiscaleinformationgainwithgloballocaltightloosecoupling