Fear nervous emotion recognition method based on video eye movement and heart rate analysis

Abstract

The invention discloses a fear nervous emotion recognition method based on video eye movement and heart rate analysis, and proposes that a Relief feature selection method is used for optimal selection analysis of two physiological signals by utilizing a mode of combining eye movement in a video and remote non-contact heart rate estimation. Finally, a k-nearest neighbor (kNN) algorithm and a least squares support vector machine (LS-SVM) algorithm are adopted to recognize and judge the fear and tension mental state of the person. The method relates to the technical field of video analysis. The fear nervous emotion recognition method based on video eye movement and heart rate analysis can solve the problems that in the prior art, an abnormal emotion recognition method is restricted by the cooperation degree of a detected person, a test method is not secret and the test efficiency is low, an abnormal data processing mechanism is introduced through eye movement data analysis, the algorithm efficiency is improved, and a feature selection method is adopted, so that the feature dimension is reduced, and the abnormal emotion recognition of similar criminal knowers is effectively improved while the training time is shortened.

Description

technical field [0001] The invention relates to the technical field of video analysis, in particular to a method for identifying fear and tension based on video eye movement data analysis and non-contact heart rate analysis. Background technique [0002] Fear and tension is a kind of negative emotion that people show towards special scenes and goals in life. Fear refers to a strong depressive emotional experience that occurs when people face a certain dangerous situation and try to get rid of it but are powerless. Fear Psychology is what is commonly referred to as "fear". According to Kelly, fear is similar to threat but to a lesser degree. Fear arises when peripheral elements of a person's building system, rather than core building blocks, prove ineffective. Physical symptoms Symptoms of multisystem sympathetic reactivity, such as shortness of breath, shortness of breath, palpitation, etc. [0003] At present, the results of using physiological signal parameters to identif...

AI Technical Summary

Problems solved by technology

[0003] At present, the results of using physiological signal parameters to identify specific emotional states of people are more objective and true. However, since the collection of physiological signals is based on contact equipment, it is necessary to wear various physiological index collection equipment on the person under examination, so , because it involves personal freedom and privacy, this method is difficult to be widely used

[0004] The MIT laboratory led by Pichard recorded five physiological parameters and extracted 40 features to explore the feasibility of emotion recognition based on multiple physiological parameters. Kim used audio materials and video clips as eliciting materials to collect four physiological parameters of 200 subjects. Parameters, using the support vector machine algorithm to classify and recognize the four kinds of emotions, it is found that when the types of emotion recognition increase using the same algorithm, the recognition rate decreases. These methods can ensure the recognition accuracy while optimizing the multi-feature , to reduce the complexity of the entire recognition algorithm model, however, these algorithms often occupy larger computing resources. For the recognition and judgment of fear and tension, most of them focus on questionnaires and observation methods at present, but this method does not use human-computer interaction. And the capture of people's fear and tension in public safety scenarios

Images