Estimating perceived indoor air quality and environmental satisfaction using a camera

Abstract

Although indoor environmental monitoring data are increasingly available, they often fail to accurately capture personal environmental perception, comfort, and satisfaction. This disconnection has led to a growing interest in using real-time feedback from building users to improve the control of heating, ventilation, and air-conditioning systems and reduce energy waste. However, traditional questionnaire-based surveys only provide a snapshot at the time of investigation and suffer from a variety of bias sources. To address this, this paper proposes a novel camera-based method for estimating perceived indoor air quality and environmental satisfaction by analysing facial expressions. We conducted indoor air pollution exposure experiments on 24 healthy adults in twin environmental chambers. Their changes in facial expressions were compared between different exposure conditions and the potential correlation between facial expressions and perceived air quality, environmental satisfaction, and self-reported sick building syndrome symptoms was explored. Preliminary results indicate that a common RGB camera can repeatably detect facial expressions from the same subject. Also, results from a subset of five subjects suggest moderate differences in concentration, calmness, contemplation, and tiredness expressions between the two air quality conditions.