Autonomous surface crack identification of concrete structures based on an improved one-stage object detection algorithm

•An improved YOLOv4 network adopting the pruning technique and the EvoNorm-S0 structure was presented.•The pruning technique is used to light-weight the network structure, and the EvoNorm-S0 could improve the detection accuracy.•Compared with the original network, the mAP50 of the improved network i...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Engineering structures Ročník 272; s. 114962
Hlavní autori: Wu, Peirong, Liu, Airong, Fu, Jiyang, Ye, Xijun, Zhao, Yinghao
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: Kidlington Elsevier Ltd 01.12.2022
Elsevier BV
Predmet:
Algorithms
Computer vision
Concrete
Concrete structures
Crack identification
Cracks
Deep learning
Object recognition
Structural engineering
Structural inspection
Structural safety
Surface cracks
YOLO series
Crack identification
Deep learning
Computer vision
Structural inspection
YOLO series
ISSN:0141-0296, 1873-7323
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:•An improved YOLOv4 network adopting the pruning technique and the EvoNorm-S0 structure was presented.•The pruning technique is used to light-weight the network structure, and the EvoNorm-S0 could improve the detection accuracy.•Compared with the original network, the mAP50 of the improved network is increased from 91.69% to 92.54% (after 100 epochs of training for both the original and the improved YOLOv4), and the weight of the proposed network is just 54.9% of the original ones.•Compared with three other leading algorithms in this field, the proposed network can identify the largest number of target species in the image, and has the highest mAP50. oncrete is a widely used material in the infrastructure system. However, this material is susceptible to several factors that eventually create concrete cracks. Thus, accurately identifying the cracks' size and location in concrete structures is crucial for structural safety evaluation. In this research, an improved YOLOv4 network adopting the pruning technique and the EvoNorm-S0 structure was put forward to better identify concrete cracks from many misleading targets. The pruning technique is used to light-weight the network structure, and the EvoNorm-S0 could improve the detection accuracy. The results indicate that compared with the original YOLOv4, the mAP50 of the improved network is increased from 91.69% to 92.54% when both models are trained for 100 epochs, and the 1-Batch inference time is reduced by 15.9%. Moreover, the weight of the proposed network is just 54.9% of the original ones. The proposed network was also compared with three other leading algorithms in this field (i.e., SSD300, YOLOv3, and YOLO X-L) using the same dataset. The results show that the proposed network can not only correctly classify the largest number of objects with a fast calculation speed, but also has the highest mAP50. Thus, this proposed network exhibits several advantages for detecting concrete cracks and is a desirable tool for practical engineering.
Bibliografia:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0141-0296
1873-7323
DOI:10.1016/j.engstruct.2022.114962