Abstract—In recent years, researches on the olfactory have been actively conducted. As one of models of olfactory function, there is KIII model proposed by Freeman et al. There have been some researches on the classification using KIII model. These class distinctions are performed by the particular feature, the amount of statistics, namely, the standard deviation of the time series signal in the KIII model. However, as the identification rates of them are low, there need to improve identification rates. In this study, we propose a high performance feature extraction method in the classification using Freeman’s olfactory KIII model, making use of the cepstrum analysis often used in speech recognition field. Finally, through computer simulations, it is verified that the proposed method is superior to the conventional method.
Index Terms—Freeman, KIII model, Fourier transform, discrete cosine transform
Cite: Masanao Obayashi, Ryohei Suda, Takashi Kuremoto, and Shingo Mabu, "A Class Identification Method Using Freeman’s Olfactory KIII Model," Journal of Image and Graphics, Vol. 4, No. 2, pp. 130-135, December 2016. doi: 10.18178/joig.4.2.130-135
Copyright © 2012-2021 Journal of Image and Graphics, All Rights Reserved