Abstract

It is important for infrastructure managers to maintain a high standard to ensure user satisfaction during a lifecycle of infrastructures. Surveillance cameras and visual inspections have enabled progress toward automating the detection of anomalous features and assessing the occurrence of the deterioration. Frequently, collecting damage data constraints time consuming and repeated inspections. One-class damage detection approach has a merit that only the normal images enables us to optimize the parameters. Simultaneously, the visual explanation using the heat map enable us to understand the localized anomalous feature. We propose a civil-purpose application to automate one-class damage detection using the fully-convolutional data description (FCDD). We also visualize the explanation of the damage feature using the up-sampling-based activation map with the Gaussian upsampling from the receptive field of the fully convolutional network (FCN). We demonstrate it in experimental studies: concrete damage and steel corrosion in civil engineering. Furthermore, for developing to be more robust application, we apply our method to another outdoor domain that contains complex and noisy background in natural disaster dataset collected by different devices. Simultaneously, we propose deeper FCDDs focusing on another powerful backbones to improve the performance for damage detection, and implement ablation studies in disaster dataset.

Statistics