Self-Sustained Autonomous Wireless Sensor Network with Integrated Solar Photovoltaic System for Internet of Smart Home-Building (IoSHB) Applications.

Conventional wireless sensor networks (WSNs) in smart home-building (SHB) are typically driven by batteries, limiting their lifespan and the maximum number of deployable units. To satisfy the energy demand for the next generation of SHB which can interconnect WSNs to make the internet of smart home-building (IoSHB), this study introduces the design and implementation of a 250 mW to 2.3 W energy harvesting device. The proposed device is dynamically autonomous owing to the integration of embedded solar photovoltaic (PV) modules and power storage through a supercapacitor (SC; 5 V, 0.47 F) capable of powering WSNs for 95 s (up to 4.11 V). The deployed device can harvest indoor and outdoor ambient light at a minimum illumination of 50 lux and a maximum illumination of 200 lux. Moreover, the proposed system supports wireless fidelity (Wi-Fi) and Bluetooth Low Energy (BLE) to do data transfer to a webserver as a complete internet of things (IoT) device. A customized android dashboard is further developed for data monitoring on a smartphone. All in all, this self-powered WSN node can interface with the users of the SHBs for displaying ambient data, which demonstrates its promising applicability and stability.

FACS-Based Graph Features for Real-Time Micro-Expression Recognition.

Several studies on micro-expression recognition have contributed mainly to accuracy improvement. However, the computational complexity receives lesser attention comparatively and therefore increases the cost of micro-expression recognition for real-time application. In addition, majority of the existing approaches required at least two frames (i.e., onset and apex frames) to compute features of every sample. This paper puts forward new facial graph features based on 68-point landmarks using Facial Action Coding System (FACS). The proposed feature extraction technique (FACS-based graph features) utilizes facial landmark points to compute graph for different Action Units (AUs), where the measured distance and gradient of every segment within an AU graph is presented as feature. Moreover, the proposed technique processes ME recognition based on single input frame sample. Results indicate that the proposed FACS-baed graph features achieve up to 87.33% of recognition accuracy with F1-score of 0.87 using leave one subject out cross-validation on SAMM datasets. Besides, the proposed technique computes features at the speed of 2 ms per sample on Xeon Processor E5-2650 machine.