ELECTROENCEPHALOGRAM (EEG) BASED EMOTIONAL LIGHTING DESIGN USING DEEP-LEARNING FOR A USER-CENTRIC APPROACH

Abstract

This study proposes a methodology for using artificial intelligence (AI) and biometrics in spatial design. The research mainly applies a gated recurrent unit (GRU) model, a recurrent neural network (RNN), to analyze electroencephalogram (EEG) data and dynamically adjust lighting according to the user's emotional state. This study suggests an illumination adjustment system that modifies lighting according to the user's emotional state using the proposed method. Integration of EEG data can overcome the limitations of lighting systems. It can effectively target individual emotional responses. The GRU model represents a significant improvement in lighting design by addressing both cognitive and emotional user needs. The model's effectiveness in processing real-time data and adapting through incremental learning was evaluated. The model has shown a significant impact on emotional architecture and spatial design, with a focus on individual experience.