Language selection

Government of Canada / Gouvernement du Canada

Search

Anomaly detection using 1D convolutional neural networks for surface enhanced Raman scattering

From National Research Council Canada

DOIResolve DOI: https://doi.org/10.1117/12.2576447
AuthorSearch for: 1; Search for: 1
EditorSearch for: Valenta, Christopher R.; Search for: Shaw, Joseph A.; Search for: Kimata, Masafumi
Affiliation
  1. National Research Council of Canada. Metrology Research Centre
FormatText, Article
ConferenceSPIE Future Sensing Technologies, November 9-13, 2020, Online Only
Subjectsurface-enhanced Raman scattering; convolutional neural networks; raman spectroscopy; anomaly detection; one-class classification; deep learning; pattern recognition
Abstract
An accurate supervised classification technique requires a large training database with an equal number of samples in each category. However, in practice, data class imbalance is naturally inherent in detection and identification tasks. In an extreme case, one category of data has a majority of training samples (positive class), causing over-classifying. In these circumstances, the negative classes are either absent, poorly sampled or not well defined. Deep one-class classifiers are artificial neural networks developed to overfit the positive class samples. This unique situation constrains the network model to be trained data features just with the knowledge of the positive class. One well-known application of one-class classifiers is for anomaly detection problem, where the model stands out outliers. In this study, we proposed using a one-dimensional CNN model for anomaly detection of Surface-Enhanced Raman Spectra applicable for Portable Raman Spectrometer in field investigations.
Publication date
PublisherSPIE
In
  • SPIE Future Sensing Technologies, 115250S-2 (November 2020).
Series
  • Proceedings of SPIE 11525.
LanguageEnglish
Peer reviewedYes
Export citationExport as RIS
Report a correctionReport a correction (opens in a new tab)
Record identifierbdd4f5e8-f157-4b17-820b-354e2187097c
Record created2021-12-17
Record modified2021-12-20
Date modified: