Detection Algorithm of Structural Surface Cracks Based on Class Activation Map
|
|
|||||||||||
Bibliographic Details
| Author(s): |
Boqiang Xu
(Department of Bridge Engineering, Tongji University, Shanghai, China)
Chao Liu (Department of Bridge Engineering, Tongji University, Shanghai, China) |
||||
|---|---|---|---|---|---|
| Medium: | conference paper | ||||
| Language(s): | English | ||||
| Conference: | IABSE Congress: Bridges and Structures: Connection, Integration and Harmonisation, Nanjing, People's Republic of China, 21-23 September 2022 | ||||
| Published in: | IABSE Congress Nanjing 2022 | ||||
|
|||||
| Page(s): | 1216-1223 | ||||
| Total no. of pages: | 8 | ||||
| DOI: | 10.2749/nanjing.2022.1216 | ||||
| Abstract: |
The computer vision algorithm based on deep learning has achieved excellent performance in structural surface damage detection, but the accurate detection algorithm has high requirements for the quantity and quality of data sets. This paper presents a method based on class activation map (CAM), which can detect the crack position and distribution only by image-level data labeling. Firstly, a classification model Vgg16-Crack is trained based on the transfer learning method, and the accuracy and generalization ability of the model are tested by the confusion matrix. Then, based on the CAM algorithm, this paper improves and optimizes the current Grad-CAM++ algorithm, and takes the CAM generated by Vgg16-Crack as the result of crack detection. Finally, the method proposed in this paper is tested in the field. The test result shows that the method proposed in this paper can realize the accurate detection of structural surface cracks. |
||||
| Keywords: |
crack detection computer vision transfer learning Convolutional Neural Networks (CNN) Class Activation Map(CAM)
|
||||
| Copyright: | © 2022 International Association for Bridge and Structural Engineering (IABSE) | ||||
| License: | This creative work is copyrighted material and may not be used without explicit approval by the author and/or copyright owner. |
||||