Crime detector system on the basis of stress assessment utilizing Photoplethysmograph and Galvanic Skin Response signals

Document Type : Original Research

Authors

1 north khorasan university of medical sciences

2 Mashhad university of medical sciences

Abstract

Aims: The main objective of this article was to design a crime detector system. The most notable effect of the stress of lying and denying the crime appears in circulatory and perspiration system. The Galvanic Skin Response (GSR) and Photoplethysmogrph (PPG) signals were used to achieve this goal.
Methods: The data set included PPG signals and GSR recorded by an inductive test of classic polygraph device. Features of time and frequency were extracted from the signals. Statistical analysis was used to determine the most productive features. Consequently, data was classified and accuracy coefficient was calculated through applying these features to Linear Discriminate Analysis )LDA(. All procedures were conducted using MATLAB 2011 software.
Results: 20 people participated in the study. The mean age of participants was 36±6 years. Our study showed that time feature in GSR signal and frequency feature in PPG had the most impacts on crime discrimination. Finally determination of crime was done with accuracy coefficient of 80% through applying Time feature in GSR signal and Frequency feature in PPG on the linear network.
Conclusion: According to the findings, the new method introduced in this study has a comparable accuracy to the results of former methods while offering a more comfortable recording and less diagnostic costs. This new method can be suggested for use as a lie screening system.

Keywords