|
Structural Monitoring and Maintenance Volume 9, Number 3, September 2022 , pages 289-303 DOI: https://doi.org/10.12989/smm.2022.9.3.289 |
||
|
Pixel-based crack image segmentation in steel structures using atrous separable convolution neural network |
||
Quoc-Bao Ta, Quang-Quang Pham, Yoon-Chul Kim, Hyeon-Dong Kam and Jeong-Tae Kim
|
||
| Abstract | ||
| In this study, the impact of assigned pixel labels on the accuracy of crack image identification of steel structures is examined by using an atrous separable convolution neural network (ASCNN). Firstly, images containing fatigue cracks collected from steel structures are classified into four datasets by assigning different pixel labels based on image features. Secondly, the DeepLab v3+ algorithm is used to determine optimal parameters of the ASCNN model by maximizing the average mean-intersection-over-union (mIoU) metric of the datasets. Thirdly, the ASCNN model is trained for various image sizes and hyper-parameters, such as the learning rule, learning rate, and epoch. The optimal parameters of the ASCNN model are determined based on the average mIoU metric. Finally, the trained ASCNN model is evaluated by using 10% untrained images. The result shows that the ASCNN model can segment cracks and other objects in the captured images with an average mIoU of 0.716. | ||
| Key Words | ||
| atrous convolution; crack identification; Deeplabv3+ network; semantic segmentation; steel structure; vision image | ||
| Address | ||
| (1) Quoc-Bao Ta, Quang-Quang Pham, Hyeon-Dong Kam, Jeong-Tae Kim: Department of Ocean Engineering, Pukyong National University, Nam-gu, Busan 48513, Republic of Korea; (2) Yoon-Chul Kim: Department of Civil Engineering, Yonsei University, Seodaemun-gu, Seoul 03722, Republic of Korea. | ||
