Micro-expression generation detection method and detection system

Abstract

The micro-expression generation detection method comprises the following steps: 1) recording normal electroencephalogram data and normal facial video data of a testee before stimulating and inducing micro-expressions; 2) stimulating and inducing micro-expressions, and recording electroencephalogram data and facial video data; 3) marking timestamp data, and matching electroencephalogram time data with facial video time data; 4) obtaining the starting point time Tsm of the response of the electroencephalogram time data, and the starting and ending frames and the top frame of the facial expression change; and 5) judging whether the the micro-expression occurs or not. According to the method, the electroencephalogram signals and the facial image information are associated through the timestamps, the electroencephalogram signals and the facial video data are processed respectively, whether the micro-expressions occur or not is judged in combination with the processing result, and the accuracy of the judgment result is high.

Description

technical field [0001] The invention relates to the technical field of EEG signal and facial video signal processing, in particular to a method for detecting whether micro-expressions occur. Background technique [0002] As a short-lived facial expression that occurs within 1 / 25s to 1 / 2s, micro-expression is considered to be a natural emotional expression that is difficult to control when people are depressed or trying to hide their true emotions, so it is a very important clue in lie detection , its importance and wide application scenarios have received more and more attention. [0003] At present, the detection of micro-expression is mainly through the method of expression recognition, which can separate the specific expression state from a given static image or dynamic video sequence to a certain extent, so as to determine the psychological emotion of the recognized object. Realize the computer's understanding and recognition of human facial expressions, but the detecti...

AI Technical Summary

Problems solved by technology

Its shortcomings are: 1. The patent does not disclose the specific method of time-correlating EEG signals and facial action pictures. Both processing EEG signals and facial picture recognition have a certain delay and cannot solve the problem of data synchronization; 2. Through Establishing a BP neural network for facial action detection requires a lot of calculation and data processing, which cannot handle a large amount of data; 3. The essence of this patent is to recognize facial pictures through EEG signals, not to separately identify and then combine EEG signals and facial pictures Judgment, its accuracy is not enough

Images