Counting and tracking people with a low-resolution heat sensor
by Max Pettersson
This page contains information about a final project from the Embedded and Distributed AI course of the Spring 2020 semester External link, opens in new window.. The focus has been developing real-time intelligent algorithms which can run on embedded systems.
Studies show that indoor environmental quality impacts the well-being of the occupants, and in turn their productivity. Counting and tracking people in an office could enable companies to improve the workplace through automation of, for example, ventilation or control of lights. However, tracking people raises questions about how the privacy of employees are handled. For this reason, counting and tracking people using cameras may not be the most viable option. An alternate option is to track people using their heat signatures, which ensures the anonymity of the tracked person. One way of detecting heat signatures from a distance is using a thermopile sensor. This sensor represents heat in an matrix where each element contains a heat value.
This project explores a method of tracking and counting people in a room using heat signatures from a thermopile sensor. Image processing techniques are used to try and isolate the heat signatures of a human while filtering out non-human heat sources. The data used for this project is collected from a thermopile sensor sold by ROL Ergo AB. This sensor was placed in the ceiling above an office table and recorded the heat signatures of up to four people sitting around this table.
For more information you can contact Max Pettersson at;