Detection of body posture with low-cost CMOS camera systems for applications in medical care

This paper describes an ongoing research program aimed to develop a network of low-cost, low-power consumption CMOS cameras whose main function is to determine, within a typical house or office room, the posture of a human body, i.e. standing, sitting, or laying down. Fast, low-complexity algorithms have been developed which can be implemented on the limited CPU and RAM resources that are available in the microprocessor chip that comes integrated with this type of cameras. When equipped with wireless communication capabilities, e.g. IEEE 802.15.4, the cameras can perform data fusion and collaborative decision functions that can further increase the probability of correct detection and classification. It is envisioned that systems such as the one described in this paper will work in conjunction with a wireless body area network (WBAN) that measures vital signs. The final goal is to have an accurate and cost effective system capable of providing relevant and timely information to the medical personnel that is responsible for the care of the patients.