Faculty and Student Teams Program

Project Descriptions

National Renewable Energy Laboratory
Electricity Resources and Building Systems Integration Center

Image Processing-Based Occupancy Detection for Building Energy Management

Project Description

IR, Ultrasonic, or RF-based sensors have long been a weakness for occupancy-based daylighting and HVAC control systems.  Low cost video with embedded microprocessor systems and efficient image processing algorithms make vision-based approaches to occupancy detection viable.  An image processing approach may also enable the creation of additional information that would be useful to a control system beyond a boolean occupied/vacant signal.  Potential feedback could also include the number of occupants, a "centroid of occupancy" for larger spaces, activity level metrics, predicted occupancy based on direction of motion, and raw video feeds for security systems.

This project seeks to develop a proof-of-concept demonstration of an image processing-based occupancy sensor using a suitable networked embedded microprocessor, such as a Freescale Netburner.  At a minimum, the proof-of-concept should provide an occupied/vacant signal over a wired or wireless network connection, but a clear development path toward some of the more sophisticated use-cases described above should be architected into the design.

Applicants Responsibilities and Relationship to Projects

Applicants will receive support under the Department of Energy Faculty Student Team Research Program (FaST) to work collaboratively with the project research team at NREL for up to 10 weeks during the summer starting in June.  Summer and academic year visits by NREL will be schedule by mutual agreement between the Research Project Managers at NREL and the successful applicant.  Faculty will be expected to identify students from the campus to participate in the FaST program offered by the Department of Energy at NREL. Faculty will provide mentorship and/or advising support to students during the summer research activities.  It is expected that the Faculty member will become an integral part of the research team working on this project and will support the project through the academic year on his or her campus.

Qualifications of Ideal Candidate

Faculty: Ph.D. with experience in applied embedded software design and image processing.  Works well in a collaborative environment with students and other researchers.  Currently teaches and collaborates with students in his/her field.  Possesses good written and verbal communication skills, and is familiar with best practices for concurrent software development.  Willing to work at NREL for an extended period during the summer.

Students:  Working towards an undergraduate degree in engineering or computer science with an emphasis on embedded software design or image processing.  Works well in collaboration with faculty, other students, and researchers.  Possesses good written and verbal communications skills.  Willing to work at NREL for an extended period during the summer.

Support and Financial Commitments

See Financial Information.

For More Information contact: