A novel approach for detection of deception using Smoothed Pseudo Wigner-Ville Distribution (SPWVD)

• DOI: 10.4236/jbise.2013.61002
• Publication Date: 2013-01-24T11:07:36Z
• Authors: Elias Ebrahimzadeh; Seyed Mohammad Alavi; Ahmad Bijar; Alireza Pakkhesal
• Keywords: Lie detection; Feature vector

For many years, the uncertainty of lie-detection systems has been one concerns defense related agencies. Clearly results these must be generalized by a high value accuracy to acceptable judicial systems. In this paper, new method based on P300-based component proposed for lie-detection. regard, test protocol is designed Odd-ball paradigm concealed information recognition. This was done 32 people and their brain signals were acquired. After preprocessing, classic features are extracted from each single trial. that, time-frequency (TF) transformation applied sweeps TF produced thereupon. Then, best combinational feature vector selected in order improve classifier accuracy. Finally, Guilty Innocent persons classified KNN MLP. We found that combination Time-Frequency Classic have better ability achieve higher amount The obtained show can detect deception 89.73% which than other previously reported methods.