Micro-expression lie detector based on laser-induced graphene sensor

Abstract

The invention discloses a micro-expression lie detector based on a laser-induced graphene sensor, which is characterized in that laser-induced graphene is utilized to prepare a flexible electronic skin sensor, and the flexible electronic skin sensor is attached to the facial muscle of a person to be examined to detect the change of the facial muscle so as to obtain the characteristic waveform of the facial expression; when the examined person generates the micro-expression, the waveform has corresponding characteristic change, and a lie detection result is obtained.

Description

technical field [0001] The invention belongs to the technical field of polygraph, in particular to a micro-expression polygraph based on a laser-induced graphene sensor. Background technique [0002] When people deliberately provide lies, they will generate a certain amount of psychological pressure, and this psychological pressure will cause a series of physiological reactions, such as abnormal breathing rate, rapid heartbeat, increased blood pressure, slightly increased body temperature, sweaty palms, and slight muscle tremors. Wait. Since these physiological responses are controlled by the human autonomic nervous system, they cannot be restrained by human subjective will. The polygraph can accurately depict abnormal physiological changes, reflecting the emotional state of the subject at that time. At present, polygraphs on the market mainly measure a person's psychological changes through three aspects, namely pulse, respiration and skin resistance. However, an innocen...

AI Technical Summary

Problems solved by technology

[0004] The existing micro-expression recognition methods are to obtain the facial images of the persons being interrogated through visual methods, and then perform micro-expression recognition based on deep neural networks. Under the premise of sufficient neural network training, for complex problems such as micro-expression recognition, the neural network needs a large amount of training data to be able to fully train. However, the existing micro-expression databases only have dozens or hundreds of videos, which is far from The conditions for sufficient training of the corresponding neural network cannot be reached, so the micro-expression recognition method based on deep features is not very robust

Images