TR2024-017
Learning-Based THz Multi-Layer Imaging With Model-Based Masks
-
- , "Learning-Based THz Multi-Layer Imaging With Model-Based Masks", International Conference on Infrared, Millimeter, and Terahertz Waves (IRMMW-THz), DOI: 10.1109/IRMMW-THz57677.2023.10299043, September 2023.BibTeX TR2024-017 PDF
- @inproceedings{Wang2023sep2,
- author = {Wang, Pu and Koike-Akino, Toshiaki and Boufounos, Petros T. and Tsujita, Wataru and Yamashita, Genki},
- title = {Learning-Based THz Multi-Layer Imaging With Model-Based Masks},
- booktitle = {International Conference on Infrared, Millimeter, and Terahertz Waves (IRMMW-THz)},
- year = 2023,
- month = sep,
- doi = {10.1109/IRMMW-THz57677.2023.10299043},
- issn = {2162-2035},
- isbn = {979-8-3503-3661-0},
- url = {https://www.merl.com/publications/TR2024-017}
- }
- , "Learning-Based THz Multi-Layer Imaging With Model-Based Masks", International Conference on Infrared, Millimeter, and Terahertz Waves (IRMMW-THz), DOI: 10.1109/IRMMW-THz57677.2023.10299043, September 2023.
-
MERL Contacts:
-
Research Areas:
Abstract:
This paper demonstrates a learning-based THz multi-layer pixel identification for non-destructive inspection. Specifically, we introduce a recurrent neural network that se- quentially learns features from THz spectrogram segments with masks from model-based sparse deconvolution. Initial performance evaluation on a three-layer sample with contents on all surfaces confirms the effectiveness of the proposed method.