A novel algorithm for people detection in grey scale thermal images

In this paper, we explore a new algorithm to detect people with thermal cameras based on the standard Implicit Shape Model (ISM) technique. Our approach starts with the ISM to define the proposed centers of people locations. Then we utilize a novel method to detect people based on the density of the concentrated proposed centers by using an auto generated threshold mechanism. Our method is easy to implement and does not require complicated computations; thus resulting in a considerable increase in the speed performance and decrease in the cost of the required hardware on mobile platforms. We evaluated our system by testing it on three image sets for indoor and three for outdoor scenarios taken from four databases. Our system showed promising results in detecting people on images taken by different types of thermal cameras under difficult scenarios. This technique will be used as the vision system for a rescue assist mobile robot currently being built at Flinders University.