Multiple sensors-based kernel machine learning in smart environment

Sensor-based monitoring systems use multiple sensors to identify high-level information based on the events that take place in a monitored environment. Identification and health care are important tasks in the smart environment. This paper presents a framework for multisensory multimedia data analysis using a kernel optimization-based principal analysis for identification and health care in a smart environment. Images of faces, palmprints, and fingerprints are used to identify a person, and a wrist pulse signal is used to analyze the person's health condition. The recognition performance evaluations are implemented on the complex dataset of face, palmprint, fingerprint, and wrist pulse signals. The experimental results show that the proposed algorithms perform well for identification and heath analysis.