A Thermal Comfort Evaluation Method And System Based On Facial Skin Temperature

Physiological parameters involved in heat regulation were proven to be objective indicators reflecting thermal comfort.Skin temperature can be continuously measured and collected by a non-contact method.The measurement method is simple and has no interference with the human body.So,it can be the most suitable physiological parameter for dynamic evaluating thermal comfort.The human face may be the body part whose skin temperature has the most practical potential for noncontact temperature measurement of thermal infrared imagers.This is because they are usually exposed to air without being covered by clothing,and their skin temperatures can be directly measured through the development of facial recognition technology.Hence,this study explored a simple and easily used objective quantitative thermal comfort evaluation method based on facial skin temperature.This study conducted two climate-controlled experiments.The experimental conditions include: 12-24 ℃(winter)and 24-32 ℃(summer),and the relative humidity was set to 60%.During exposure(140 min),a thermal infrared camera and i Button instruments were used to record the participants’ skin temperatures on the forehead,nose,right ear,right cheek,left cheek,left ear,and chin.Their thermal sensation,comfort,and acceptability of perceived thermal environments were documented many times.The main findings are as follows:(1)The forehead skin temperature was the highest among the measured facial parts.The nose skin temperature is lowest when the air temperature is lower than 24 ℃ during winter and under 24 and 26 ℃ for more than 30 minutes during summer.The uneven degree of the skin temperature distribution increased as air temperature or thermal sensation votes decreased.(2)The results from the correlation coefficients,regression coefficients,and significance levels,reveal a good linear association between facial skin temperature and thermal sensation.Then,it confirmed that facial skin temperature can be used to predict thermal sensation.Under the same thermal sensation ratings,summer facial skin temperature is significantly higher than that during winter.Female subjects’ neutral nose skin temperature is higher than that of males.The correlation between facial skin temperature and the thermal sensation was significantly higher after 15 min of exposure time than that during 0– 15 min.(3)A Fisher discriminant analysis method is used to build the evaluation models based on facial skin temperature.These models directly determine whether an individual is in a neutral state through neutral facial skin temperature thresholds and have physiological significance.The built evaluation model is simple and easier to use.The results show that the highest precision of the evaluation model was 80%,and the corresponding false-negative rate(FNR)was 27%.The neutral facial skin temperature ranges were higher during summer than during winter.After 15 min of exposure,the models exhibited higher reliability.The average skin temperature from multiple facial parts was more suitable for evaluating neutral thermal sensations than that from a single facial part.The average skin temperature of the nose,cheeks,and chin is the most appropriate index for future applications of infrared thermal imaging.To a certain extent,the models were further improved by sex.The personal thermal comfort models can significantly improve the evaluation results.When considering individual difference,the highest precision increased by 11%,and the corresponding FNR decreased by9%.(4)The influencing factors of the thermal infrared imager’s measurement of facial skin temperature are analyzed based on infrared radiation theory and a proposed simplified operation range estimation method.Statistical analysis was used to determine the influencing degrees of each factor to the measurement difference.Considering these factors,we proposed an error correction model.The correction results show that the mean difference is-0.01°C(95% confidence interval is-0.19 to0.17°C)with the 95% limits of agreement(Lo A)within ±0.53°C.Combined with the thermal comfort evaluation model,a monitoring system was developed for dynamic assessing thermal comfort based on facial skin temperature.The results of this study provide a theoretical basis for constructing a healthy,comfortable,low-carbon,and high-quality building thermal environment from the perspective of thermal comfort assessment.