Single-Pixel Thermopile Infrared Sensing for People Counting

People-counting data can be used in building-control systems to improve comfort and space management applications optimize building space. In this work, we consider a combi-sensor with single-pixel thermopile passive infrared sensor for people counting. We first develop signal model object temperature measurements under multiple occupancy. then propose counting method based on: cumulative sum (CUSUM) change detection the signal, forming count estimate using likelihoods of differential mean detected changes, decision fusion an vacancy sensor. The proposed is evaluated generated developed as well experimental from cell office/meeting room environment. obtain average error 0.11 0.19 90% instants respectively when considering 15 minute windows simulated datasets.