Search

Recently, it has become a trend to construct a thermal environment with human thermal comfort. The advantage of human thermal comfort is that we could adjust the environment through the thermal sensation of human. Since human thermal comfort is influenced by several factors such as environmental temperature, environmental humidity, airflow, mean radiant heat, and so on, it is usually difficult to be directly evaluated. This paper proposes a method of estimating subjective thermal comfort and objective thermal comfort using CNN by RGB image data, thermal image data, and sensor data. We built a pipe-type booth in a room with a small air conditioner, two heaters, a humidifier, and a dehumidifier to acquired learning data, and we trained CNN under six different learning patterns to estimate thermal comfort. We found that image data is conducive to estimate subjective thermal comfort, and sensor data is conducive to estimating objective thermal comfort.

Realizing an environment that could automatically modify temperature and humidity according to human thermal comfort using an air conditioning system, it is necessary to measure the quantified changes of human thermal comfort continually. While quantifying thermal comfort, human biometric data and environmental data from various environments are needed. Therefore, we proposed a method to estimate human thermal comfort through regression analysis with human biometric data acquired by wearable sensors. Since the PMV model is generally used to evaluate human thermal comfort, the PMV value as the correct thermal comfort could be calculated from the PMV formula. We constructed an experiment environment for acquiring the subject's biometric data and gained the subjects' METs and clo value for calculating the PMV value. The biometric data were analyzed in three regression models to predict PMV value, and MAE and RMSE evaluated each regression model. The fewest number of wearable sensors could be confirmed by gradually reducing the feature values of input data of the regression model. As a result, we acknowledged that the human thermal comfort in an indoor environment could be estimated by heart rate data and left-arm temperature data while applying the proposed method to a daily environment.

Following are two mainstream approaches of video retrieval from large-scale video data using query sentences: (1) an approach to find pre-trained concepts such as objects, persons, scenes, and activities corresponding to a query sentence, and (2) an approach to map a query sentence and images/videos into the same feature space and directly search for images/videos that match the query sentence. In this study, we analyze the advantages and disadvantages of these two approaches using a large-scale video database of TRECVID benchmark and confirm whether the fusion of these approaches can improve video retrieval performance.