Multiple Detections Application for Indoor Tracking Using PIR Sensor and Kalman Filter

Abstract

Recently, human tracking in multiple indoor environments is getting broadly in demand to enhance security surveillance. Traditional passive video surveillance shown that it has working ineffectively nowadays because the number of cameras has exceeded the ability of operators to monitor them. In this paper, we proposed methods of detecting human presence using Pyroelectric Infrared (PIR) Motion Sensor and tracking people in multiple indoor locations using Kalman filter-based estimation. The proposed method is implemented to analyze the movement of people within the prescribed area and the result will be presented in footprint mapping of the said area. This will further enhanced building security surveillance especially at the sensitive or restricted areas. Experiments for single target tracking in several areas are carried out to verify the application of the developed system. As the results, the maximum error for tracking trajectory reduced from 0.28m to 0.19m and average error for tracking trajectory also reduced from 0.10m to 0.07m after using Kalman filter estimation algorithm.