Automatic Recognition of Road Damage Based on Lightweight Attentional Convolutional Neural Network
An efficient road damage detection system can reduce the risk of road defects to motorists and road maintenance costs to traffic management authorities, for which a lightweight end-to-end road damage detection network is proposed in this paper, aiming at fast and automatic accurate identification an...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-12-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/22/24/9599 |
| _version_ | 1797455355329904640 |
|---|---|
| author | Han Liang Seong-Cheol Lee Suyoung Seo |
| author_facet | Han Liang Seong-Cheol Lee Suyoung Seo |
| author_sort | Han Liang |
| collection | DOAJ |
| description | An efficient road damage detection system can reduce the risk of road defects to motorists and road maintenance costs to traffic management authorities, for which a lightweight end-to-end road damage detection network is proposed in this paper, aiming at fast and automatic accurate identification and classification of multiple types of road damage. The proposed technique consists of a backbone network based on a combination of lightweight feature detection modules constituted with a multi-scale feature fusion network, which is more beneficial for target identification and classification at different distances and angles than other studies. An embedded lightweight attention module was also developed that can enhance feature information by assigning weights to multi-scale convolutional kernels to improve detection accuracy with fewer parameters. The proposed model generally has higher performance and fewer parameters than other representative models. According to our practice tests, it can identify many types of road damage based on the images captured by vehicle cameras and meet the real-time detection required when piggybacking on mobile systems. |
| first_indexed | 2024-03-09T15:53:16Z |
| format | Article |
| id | doaj.art-183de7eca7034f2a959f068b8d875658 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-09T15:53:16Z |
| publishDate | 2022-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-183de7eca7034f2a959f068b8d8756582023-11-24T17:52:13ZengMDPI AGSensors1424-82202022-12-012224959910.3390/s22249599Automatic Recognition of Road Damage Based on Lightweight Attentional Convolutional Neural NetworkHan Liang0Seong-Cheol Lee1Suyoung Seo2Department of Civil Engineering, Kyungpook National University, Daegu 37224, Republic of KoreaDepartment of Civil Engineering, Kyungpook National University, Daegu 37224, Republic of KoreaDepartment of Civil Engineering, Kyungpook National University, Daegu 37224, Republic of KoreaAn efficient road damage detection system can reduce the risk of road defects to motorists and road maintenance costs to traffic management authorities, for which a lightweight end-to-end road damage detection network is proposed in this paper, aiming at fast and automatic accurate identification and classification of multiple types of road damage. The proposed technique consists of a backbone network based on a combination of lightweight feature detection modules constituted with a multi-scale feature fusion network, which is more beneficial for target identification and classification at different distances and angles than other studies. An embedded lightweight attention module was also developed that can enhance feature information by assigning weights to multi-scale convolutional kernels to improve detection accuracy with fewer parameters. The proposed model generally has higher performance and fewer parameters than other representative models. According to our practice tests, it can identify many types of road damage based on the images captured by vehicle cameras and meet the real-time detection required when piggybacking on mobile systems.https://www.mdpi.com/1424-8220/22/24/9599object detectionlightweight networkattention mechanismroad damagecomputer vision |
| spellingShingle | Han Liang Seong-Cheol Lee Suyoung Seo Automatic Recognition of Road Damage Based on Lightweight Attentional Convolutional Neural Network Sensors object detection lightweight network attention mechanism road damage computer vision |
| title | Automatic Recognition of Road Damage Based on Lightweight Attentional Convolutional Neural Network |
| title_full | Automatic Recognition of Road Damage Based on Lightweight Attentional Convolutional Neural Network |
| title_fullStr | Automatic Recognition of Road Damage Based on Lightweight Attentional Convolutional Neural Network |
| title_full_unstemmed | Automatic Recognition of Road Damage Based on Lightweight Attentional Convolutional Neural Network |
| title_short | Automatic Recognition of Road Damage Based on Lightweight Attentional Convolutional Neural Network |
| title_sort | automatic recognition of road damage based on lightweight attentional convolutional neural network |
| topic | object detection lightweight network attention mechanism road damage computer vision |
| url | https://www.mdpi.com/1424-8220/22/24/9599 |
| work_keys_str_mv | AT hanliang automaticrecognitionofroaddamagebasedonlightweightattentionalconvolutionalneuralnetwork AT seongcheollee automaticrecognitionofroaddamagebasedonlightweightattentionalconvolutionalneuralnetwork AT suyoungseo automaticrecognitionofroaddamagebasedonlightweightattentionalconvolutionalneuralnetwork |