Emotion recognition method and system and storage medium

Abstract

The invention discloses an emotion recognition method and system and a storage medium. The method comprises the following steps: extracting face emotion features according to externally input video stream information to obtain a vibration image; performing pixel micro-seismic analysis according to the vibration image to form a vibration light ring; and performing emotional state recognition according to the vibrating light ring. According to the recognition method, face emotional features are extracted to form a vibration image, and the emotional state can be directly reflected; meanwhile, a vibration light ring is obtained through pixel micro-seismic analysis on the vibration image, the accuracy of emotion judgment can be improved and judgment errors can be reduced by adopting a pixel micro-seismic analysis technology, finally, the movement of the object and the emotional state of the object is described through the vibration light ring, and a corresponding emotional state diagram is displayed more visually. Therefore, by implementing the method, a more reasonable and accurate feature data value is provided for emotion recognition, the error of an emotion recognition result is reduced, the accuracy of emotion recognition is improved, and an effective judgment basis is provided for emotion prediction.

Description

technical field [0001] The invention relates to the technical field of intelligent recognition, in particular to an emotion recognition method, system and storage medium. Background technique [0002] Micro-expression is a person's inner reveal and cover-up, a term in psychology, and a branch of human behavior. People express their inner feelings to each other by making some expressions. Between different expressions or within a certain expression, the face often "leaks" other information. "Micro-expressions" can last as little as 1 / 25 of a second, and while a subconscious expression may last only a split second, it's an annoying quality that can easily reveal emotion. The duration of micro-expression is very short, and it is easier to capture by machine than human eyes. The application of deep learning algorithm makes the recognition of micro-expression very accurate. [0003] The methods for recognizing human emotions in the prior art rely on contact detection to obtain ...

AI Technical Summary

Problems solved by technology

[0005] In view of this, the embodiments of the present invention provide an emotion recognition method, system and storage medium to solve the problem that the deviation of emotion recognition results in the prior art cannot provide effective judgment basis for emotion prediction and cannot realize pre-safety management and control. technical problem

Images