Real-time and effective human thermal discomfort detection plays a critical role in achieving energy efficient control of human centered intelligent buildings because estimation results can provide effective feedback signals to heating, ventilation and air conditioning (HVAC) systems. How to detect occupant thermal discomfort is a challenge. Unfortunately, contact or semi-contact perception methods are inconvenient in practical application. From the contactless perspective, a kind of vision-based contactless human discomfort pose estimation method was proposed in this paper. Firstly, human pose data were captured from a vision-based sensor, and corresponding human skeleton information was extracted. Five thermal discomfort-related human poses were analyzed, and corresponding algorithms were constructed. To verify the effectiveness of the algorithms, 16 subjects were invited for physiological experiments. The validation results show that the proposed algorithms can recognize the five human poses of thermal discomfort.

In human-centered intelligent building, real-time measurements of human thermal comfort play critical roles and supply feedback control signals for building heating, ventilation, and air conditioning (HVAC) systems. Due to the challenges of intra- and inter-individual differences and skin subtleness variations, there has not been any satisfactory solution for thermal comfort measurements until now. In this paper, a contactless measuring method based on a skin sensitivity index and deep learning (NISDL) was proposed to measure real-time skin temperature. A new evaluating index, named the skin sensitivity index (SSI), was defined to overcome individual differences and skin subtleness variations. To illustrate the effectiveness of SSI proposed, a two multi-layers deep learning framework (NISDL method I and II) was designed and the DenseNet201 was used for extracting features from skin images. The partly personal saturation temperature (NIPST) algorithm was use for algorithm comparisons. Another deep learning algorithm without SSI (DL) was also generated for algorithm comparisons. Finally, a total of 1.44 million image data was used for algorithm validation. The results show that 55.62% and 52.25% error values (NISDL method I, II) are scattered at (0 °C, 0.25 °C), and the same error intervals distribution of NIPST is 35.39%.

Low visibility on expressways caused by heavy fog and haze is a main reason for traffic accidents. Real-time estimation of atmospheric visibility is an effective way to reduce traffic accident rates. With the development of computer technology, estimating atmospheric visibility via computer vision becomes a research focus. However, the estimation accuracy should be enhanced since fog and haze are complex and time-varying. In this paper, a total bounded variation (TBV) approach to estimate low visibility (less than 300 m) is introduced. Surveillance images of fog and haze are processed as blurred images (pseudo-blurred images), while the surveillance images at selected road points on sunny days are handled as clear images, when considering fog and haze as noise superimposed on the clear images. By combining image spectrum and TBV, the features of foggy and hazy images can be extracted. The extraction results are compared with features of images on sunny days. Firstly, the low visibility surveillance images can be filtered out according to spectrum features of foggy and hazy images. For foggy and hazy images with visibility less than 300 m, the high-frequency coefficient ratio of Fourier (discrete cosine) transform is less than 20%, while the low-frequency coefficient ratio is between 100% and 120%. Secondly, the relationship between TBV and real visibility is established based on machine learning and piecewise stationary time series analysis. The established piecewise function can be used for visibility estimation. Finally, the visibility estimation approach proposed is validated based on real surveillance video data. The validation results are compared with the results of image contrast model. Besides, the big video data are collected from the Tongqi expressway, Jiangsu, China. A total of 1,782,000 frames were used and the relative errors of the approach proposed are less than 10%.

Predicting attributes from face images in the wild is a challenging computer vision problem. To automatically describe face attributes from face containing images, traditionally one needs to cascade three technical blocks --- face localization, facial descriptor construction, and attribute classification --- in a pipeline. As a typical classification problem, face attribute prediction has been addressed using deep learning. Current state-of-the-art performance was achieved by using two cascaded Convolutional Neural Networks (CNNs), which were specifically trained to learn face localization and attribute description. In this paper, we experiment with an alternative way of employing the power of deep representations from CNNs. Combining with conventional face localization techniques, we use off-the-shelf architectures trained for face recognition to build facial descriptors. Recognizing that the describable face attributes are diverse, our face descriptors are constructed from different levels of the CNNs for different attributes to best facilitate face attribute prediction. Experiments on two large datasets, LFWA and CelebA, show that our approach is entirely comparable to the state-of-the-art. Our findings not only demonstrate an efficient face attribute prediction approach, but also raise an important question: how to leverage the power of off-the-shelf CNN representations for novel tasks.