G Goldswain and J Tapson 2006 Meas. Sci. Technol. 17 2711 doi:10.1088/0957-0233/17/10/025
G Goldswain and J Tapson
Show affiliationsWe investigate using a kernel learning machine, specifically kernel ridge regression (KRR), to predict volume fractions in typical industrial electrical impedance tomography (EIT) applications. The 'curse of dimensionality' associated with applying such methods to physically captured EIT training data is overcome with a new training method, involving sampling of training data during rapid random repositioning of a set of physical objects in the measurement plane. We compare the performance to multi-layer perceptron (MLP) neural networks which appear to be the most common computational intelligence approach to the EIT reconstruction problem. We use empirically trained static situations so as to compare the results to previous research. Dynamic situations are also investigated, and KRR is shown to outperform MLP methods in both cases. Furthermore, KRR is shown to be a useful tool in EIT for extracting process information from industrial flows without first performing conventional image reconstruction.
07.05.Mh Neural networks, fuzzy logic, artificial intelligence
Issue 10 (October 2006)
Received 10 April 2006, in final form 27 July 2006
Published 31 August 2006
G Goldswain and J Tapson 2006 Meas. Sci. Technol. 17 2711
P Bolognesi et al 2001 J. Phys. B: At. Mol. Opt. Phys. 34 3193
B. T. Soifer et al. 1999 ApJ 513 207
Shinji Tsujikawa et al 2004 Class. Quantum Grav. 21 5767
T S Ibrahim et al 2001 Phys. Med. Biol. 46 2545
Donald C. Morton 2003 ApJS 149 205
J. Albert et al. 2007 ApJ 667 358
Jens Woitas et al. 2002 ApJ 580 336
Iwan Jensen 2000 J. Phys. A: Math. Gen. 33 5953
Nikodem J Popławski 2007 Class. Quantum Grav. 24 3013