We investigate the method to analyze interferometric plasmonic microscopy (IPM) images using a deep learning approach. An IPM image was generated by employing an optical model: the image intensity was formed by reflected and scattered fields. Convolutional neural network was utilized for the classification of IPM images. Conventional detection method based on fourier filtering was taken for comparison with the proposed method. It was confirmed that deep learning improves the performance significantly, in particular, robustness to noise. These results suggested applicability of deep learning beyond IPM images with higher efficiency.
|Title of host publication||Plasmonics|
|Subtitle of host publication||Design, Materials, Fabrication, Characterization, and Applications XX|
|Editors||Din Ping Tsai, Takuo Tanaka, Yu-Jung Lu|
|Publication status||Published - 2022|
|Event||Plasmonics: Design, Materials, Fabrication, Characterization, and Applications XX 2022 - San Diego, United States|
Duration: 2022 Aug 21 → 2022 Aug 25
|Name||Proceedings of SPIE - The International Society for Optical Engineering|
|Conference||Plasmonics: Design, Materials, Fabrication, Characterization, and Applications XX 2022|
|Period||22/8/21 → 22/8/25|
Bibliographical noteFunding Information:
This work was supported by the National Research Foundation (NRF) grants (NRF-2022R1A4A2000748) funded Korean Government and the Korea Medical Device Development Fund (RS-2020-KD000088).
© 2022 SPIE.
All Science Journal Classification (ASJC) codes
- Electronic, Optical and Magnetic Materials
- Condensed Matter Physics
- Computer Science Applications
- Applied Mathematics
- Electrical and Electronic Engineering