Emotion Recognition using PPG Based on Multi-Time Window PRV Feature Selection

Abstract

Emotion is a series of feelings triggered by specific objects or activities, influenced by various factors such as personality, mood, and motivation, and manifested in multiple forms. As a crucial part of emotion research, emotion recognition has found extensive use in various fields such as mental health monitoring, and intelligent system design. In the study of emotion recognition based on PPG, PPG have received considerable attention because of their objective nature and the ease with which they can be collected., and have been widely used in various emotion recognition systems. However, utilizing PPG signals for emotion recognition still presents several challenges. Traditional feature extraction from the entire signal can cause local features to be overshadowed by overall statistics, failing to capture critical local emotion-related features. To address these issues, this paper focuses on PPG, aiming to improve the efficiency of feature extraction and utilization in PPG emotion recognition tasks. This, in turn, broadens the potential applications of wearable devices for emotion monitoring. Experimental results demonstrate that compared to existing PPG emotion recognition methods, this method shows advantages on the DEAP public dataset.