Abstract
Image-based crack detection methods have been extensively studied due to their cost-effectiveness in terms of data acquisition and processing. However, automated crack detection still remain a challenging task due to complexity of image background and different patterns of cracks. To address these issues, this paper proposes a deep residual network with transfer learning for pixel-level crack detection on road surface images. The network was trained on 71 images of CrackForest dataset and tested on 47 images of it. Experimental results suggest that the deep residual network is superior to the existing algorithms with recall value and precision value of 84.90% and 93.57%, respectively.
| Original language | English |
|---|---|
| Publication status | Published - 2018 Jan 1 |
| Event | 35th International Symposium on Automation and Robotics in Construction and International AEC/FM Hackathon: The Future of Building Things, ISARC 2018 - Berlin, Germany Duration: 2018 Jul 20 → 2018 Jul 25 |
Other
| Other | 35th International Symposium on Automation and Robotics in Construction and International AEC/FM Hackathon: The Future of Building Things, ISARC 2018 |
|---|---|
| Country | Germany |
| City | Berlin |
| Period | 18/7/20 → 18/7/25 |
Fingerprint
All Science Journal Classification (ASJC) codes
- Computer Vision and Pattern Recognition
- Artificial Intelligence
- Building and Construction
Cite this
}
A deep residual network with transfer learning for pixel-level road crack detection. / Bang, Seongdeok; Park, Somin; Kim, Hongjo; Yoon, Yeo san; Kim, Hyoungkwan.
2018. Paper presented at 35th International Symposium on Automation and Robotics in Construction and International AEC/FM Hackathon: The Future of Building Things, ISARC 2018, Berlin, Germany.Research output: Contribution to conference › Paper
TY - CONF
T1 - A deep residual network with transfer learning for pixel-level road crack detection
AU - Bang, Seongdeok
AU - Park, Somin
AU - Kim, Hongjo
AU - Yoon, Yeo san
AU - Kim, Hyoungkwan
PY - 2018/1/1
Y1 - 2018/1/1
N2 - Image-based crack detection methods have been extensively studied due to their cost-effectiveness in terms of data acquisition and processing. However, automated crack detection still remain a challenging task due to complexity of image background and different patterns of cracks. To address these issues, this paper proposes a deep residual network with transfer learning for pixel-level crack detection on road surface images. The network was trained on 71 images of CrackForest dataset and tested on 47 images of it. Experimental results suggest that the deep residual network is superior to the existing algorithms with recall value and precision value of 84.90% and 93.57%, respectively.
AB - Image-based crack detection methods have been extensively studied due to their cost-effectiveness in terms of data acquisition and processing. However, automated crack detection still remain a challenging task due to complexity of image background and different patterns of cracks. To address these issues, this paper proposes a deep residual network with transfer learning for pixel-level crack detection on road surface images. The network was trained on 71 images of CrackForest dataset and tested on 47 images of it. Experimental results suggest that the deep residual network is superior to the existing algorithms with recall value and precision value of 84.90% and 93.57%, respectively.
UR - http://www.scopus.com/inward/record.url?scp=85053872134&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85053872134&partnerID=8YFLogxK
M3 - Paper
ER -