Abstract
Deception detection is becoming an interesting filed in different areas related to security, criminal investigation, law enforcement and terrorism detection. Recently non-verbal features have become essential features for deception detection process. One of the most important kind of these features is facial expression. The importance of these expressions come from the idea that Human face contain different expressions each of which is directly related to a certain state. In this research paper, facial expressions' data are collected for 102 participants (25 women and 77 men) as video clips. There are 504 clips for lie response and 384 for truth response (total 888 video clips). Facial expressions in a form of Action Units (AUs) are extracted for each frame with video clip. The AUs are encoded based on Facial Action Coding System (FACS) which are 18 AUs. These are: AU 1, 2, 4, 5, 6, 7, 9, 10, 12, 14, 15, 17, 20, 23, 25, 26, 28 and 45. Based on the collected data, only six AUs are the most effective and have a direct impact on the discrimination process between liar and truth teller. These AUs are AU 6, 7, 10, 12, 14 and 28
Highlights
Deception is defined as concealing the truth from individuals using face and body gestures [1]
Another study performed by Demyanov et al [12] based on Action Units (AUs) detection for 270 participants that were taken from Mafia TV show and the archived accuracy was 70.26%
This work uses the process of selection for the most repeated AUs vectors or in other words, each AU may occur in the frame within a video clip so the decision for the most repeated AUs takes over all frames within a single video clip
Summary
Deception is defined as concealing the truth from individuals using face and body gestures [1]. LITERATURE REVIEW Recently, different studies are performed in the area of DDS. These studies depend on one of two kinds of features either verbal or non-verbal. In another study performed by Azar and Campisi [9] in this study, the designed system depends on infrared imaging through detecting and measuring the temperature change in the nose area. This study was performed on 16 participants, and the achieved detection accuracy was about 83.5%. Another study performed by Demyanov et al [12] based on AUs detection for 270 participants that were taken from Mafia TV show and the archived accuracy was 70.26%. All the mentioned studies suffer from several drawbacks like: a limited number of participants, performed in a constrained environment and there is no optimization process on the selected features. This work is performed on a real database that contains 102 subjects and is performed in unconstrained environments, the optimization process is performed on the extracted features to select only the effective ones
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
