Crack Segmentation by Leveraging Multiple Frames of Varying Illumination

Abstract

In this work, we present an automated inspection approach to assist remote visual examinations of nuclear power plant components. An automated approach would require detecting often low contrast cracks that could be surrounded by or even within textures with similar appearances such as welding, scratches, and grind marks. We propose a crack segmentation method for remote visual examination videos by aggregating the pixel-level classification confidence from multiple frames consisting of different illumination conditions. A dataset of 685 pixel-level ground truth images consisting of 37 cracks from remote visual examination videos is used for evaluation. The results show that the proposed method provides a significant improvement over hand-crafted feature based segmentation and 9% over convolutional neural network based method.

Cite

Text

Schmugge et al. "Crack Segmentation by Leveraging Multiple Frames of Varying Illumination." IEEE/CVF Winter Conference on Applications of Computer Vision, 2017. doi:10.1109/WACV.2017.121

Markdown

[Schmugge et al. "Crack Segmentation by Leveraging Multiple Frames of Varying Illumination." IEEE/CVF Winter Conference on Applications of Computer Vision, 2017.](https://mlanthology.org/wacv/2017/schmugge2017wacv-crack/) doi:10.1109/WACV.2017.121

BibTeX

@inproceedings{schmugge2017wacv-crack,
  title     = {{Crack Segmentation by Leveraging Multiple Frames of Varying Illumination}},
  author    = {Schmugge, Stephen J. and Rice, Lance and Lindberg, John and Grizzi, Robert and Joffe, Chris and Shin, Min C.},
  booktitle = {IEEE/CVF Winter Conference on Applications of Computer Vision},
  year      = {2017},
  pages     = {1045-1053},
  doi       = {10.1109/WACV.2017.121},
  url       = {https://mlanthology.org/wacv/2017/schmugge2017wacv-crack/}
}