Development of next generation vision sensor with coded multi-exposure pixel and compressive sensing based readout

Vision sensor is essential components in sensor applications of industry 4.0 such as autonomous vehicle, Internet of Things (IoT), Neural Network. The main goal of this research is to study, design, and develop a new concept of computational vision sensors called “transport-aware”. Unlike conventional vision sensors which record all incident light, transport-aware vision sensor can be programmed to block some of that light, based on the actual 3D paths it followed through a scene. By using coded exposure pixel (CEP) image sensor, we can control the exposure of the vision sensor at the individual pixel level. And the compressive sensing(CS) based readout will enable very high-speed imaging. In this research, I will do research on designing the novel vision sensor which can see high-speed imaging by utilizing CEP and CS-based readout circuit. This novel image sensor can be used to applications such as self-driving cars, biomedical imaging, drones, robots, and machine vision.

Faculty Supervisor:

Roman Genov

Student:

Partner:

Gwangju Institute of Science and Technology

Discipline:

Engineering

Sector:

Technology; Information and Communications Technology; New and Digital Media

University:

University of Toronto

Program: