Non-intrusive interpretation of human thermal comfort through analysis of facial infrared thermography
Non-intrusive interpretation of human thermal comfort through analysis of facial infrared thermography
复制标题
DOI：
10.1016/j.enbuild.2018.07.025
复制
发表时间：
2018-10-01
影响因子：
6.7
通讯作者：
Kamat, Vineet R.
中科院分区：
文献类型：
--
作者：
Li, Da;Menassa, Carol C.;Kamat, Vineet R.
了解居住者的热感觉和舒适度对于确定建筑物中供暖、通风和空调（HVAC）系统的操作设置至关重要。由于人为因素和环境因素的持续影响，居住者的热感觉和舒适度会随着时间的推移而变化。因此，要对环境进行动态控制，就需要对热舒适性进行实时监测。本文提出了一种新型的非侵入式红外热成像框架，通过使用低成本热像仪测量从不同面部区域收集的皮肤温度来估计乘员的热舒适度。与现有的依赖于将传感器直接放置在人体上进行皮肤温度测量的方法不同，所提出的框架能够检测居住者的存在，提取面部区域，测量皮肤温度特征，并在最小程度上干扰建筑物居住者的情况下解释热舒适条件。通过对12名被试进行制冷、加热和稳态实验，验证了该方法的有效性。结果表明，耳朵、鼻子和脸颊是热舒适的最指示性指标，所提出的框架可用于评估乘员的热舒适，平均准确率为85%。(C) 2018 Elsevier B.V.版权所有
Understanding occupants' thermal sensation and comfort is essential to defining the operational settings for Heating, Ventilation and Air Conditioning (HVAC) systems in buildings. Due to the continuous impact of human and environmental factors, occupants' thermal sensation and comfort level can change over time. Thus, to dynamically control the environment, thermal comfort should be monitored in real time. This paper presents a novel non-intrusive infrared thermography framework to estimate an occupant's thermal comfort level by measuring skin temperature collected from different facial regions using low-cost thermal cameras. Unlike existing methods that rely on placing sensors directly on humans for skin temperature measurement, the proposed framework is able to detect the presence of occupants, extract facial regions, measure skin temperature features, and interpret thermal comfort conditions with minimal interruption of the building occupants. The method is validated by collecting thermal comfort data from a total of twelve subjects under cooling, heating and steady-state experiments. The results demonstrate that ears, nose and cheeks are most indicative of thermal comfort and the proposed framework can be used to assess occupants' thermal comfort with an average accuracy of 85%. (C) 2018 Elsevier B.V. All rights reserved.