Multi-mode intelligent emotion sensing system

Abstract

The invention provides a multi-mode intelligent emotion sensing system comprising a collecting module, a recognition module, and a fusion module. The recognition module consists of an emotion-based emotion recognition unit, a voice-based emotion recognition unit, a behavior-based emotion recognition unit, and a physiological-signal-based emotion recognition unit. All emotion recognition units in the recognition module carry out recognition on multi-mode information to obtain emotion components including emotion types and emotional intensities; and the fusion module fuses the emotion components of the recognition module to realize accurate sensing of body emotions. The multi-mode intelligent emotion sensing system has the following beneficial effects: compared with the prior art, the system is capable of identify emotions accurately to detect real emotions of users with thinking and behaving in disagreement effectively.

Description

technical field [0001] The invention relates to the technical field of emotion recognition, in particular to a multimodal intelligent emotion perception system. Background technique [0002] The current technology is as follows: [0003] 1. An intelligent gender and emotion recognition detection system and method based on vision and voice, the system includes an image-based emotion and gender recognition module, which is used to recognize the emotions of people in the car according to the face image, and perform vehicle detection based on the face. The gender recognition of the people in the car; the voice-based emotion and gender recognition module is used to carry out the emotion recognition of the people in the car according to the human voice, and the gender recognition of the people in the car according to the voice; the fusion module is used to identify the gender The results of matching and emotion recognition are fused, and sent to the personalized intelligent voice...

AI Technical Summary

Problems solved by technology

[0008] 1. Poor integration: The signal collected by the existing technology is relatively single, and the measurement of multiple signals needs to be connected to a variety of different devices

It is impossible to collect multiple information at the same time, that is, it is unable to perceive a variety of information that can represent emotions in the subject's current emotional state

[0009] 2. Low accuracy: Relying on single-modal signals cannot accurately identify real emotions, and there will be duplicity

[0010] 3. Ignoring the measurement of emotional intensity: Existing technologies often only focus on the identification of emotional types, while ignoring the measurement of emotional intensity

Ignoring the measurement of emotional intensity will not be able to achieve real-time and accurate perception of emotional changes

[0011] 4. The influence of human behavior on emotional perception is ignored: most of the existing technologies use subjective judgment, expression information, voice information, etc. to identify emotions, while ignoring the mapping relationship between human behavior information and emotions

[0012] 5. The influence of physiological signals on emotional perception is ignored: most of the existing technologies ignore the influence of physiological signals on emotional perception

Emotion recognition based on subjective judgment, facial expression, and voice information in the prior art is simple and intuitive, and the measurement is simple, but the results are often affected by subjective factors, easy to disguise, and sometimes cannot truly reflect the emotional state of the subject at the time

Images