Dispersion-Independent Terahertz Classification Based on Geometric Algebra for Substance Detection

Zhou, S. L.; Valchev, D. G.; Dinovitser, A.; Chappell, J. M.; Iqbal, A.; Ng, B. W-H.; Kee, T. W. and Abbott, D. (2016). Dispersion-Independent Terahertz Classification Based on Geometric Algebra for Substance Detection. In: 41st International Conference on Infrared, Millimeter, and Terahertz waves (IRMMW-THz), 25-30 Sep 2016, Copenhagen, Denmark.

DOI: https://doi.org/10.1109/IRMMW-THz.2016.7758626

Abstract

We demonstrate and validate Geometric Algebra (GA) based terahertz (THz) signal classification of various powders in tablet form of various thicknesses, and compare the results with a conventional Support Vector Machine (SVM) approach. By using geometric algebra we can perform classification independently of dispersion and hence independently of the transmission path length through the sample. In principle, it may be possible to extend the GA coordinate-free transformation to other types of pulsed signals, such as pulsed microwaves or even acoustic signals in such fields as seismology. The classifier is available for download at Github, https://github.com/swuzhousl/Shengling-zhou/blob/geometric-algebra-classifier/GAclassifier/

Viewing alternatives

Metrics

Item Actions

Export

You can export this page using these formats Digital Object Identifier - DOI