Research Interests
Hyperspectral imaging (HSI) is used in many applications where there is a need to identify visually similar materials having fine spectral signatures. While the high spatial and spectral resolutions are an advantage of HSI, the volume of data presents processing challenges, particularly in time-sensitive, real-time or near real-time applications. One such application area requiring real-time processing is robotic vision for recycling. HSI has been researched for vision systems to separate waste streams including plastics, textiles, wood, electronics, and building materials.
Given the increased adoption of circular economy frameworks, my research aims to develop applications for material classification through full material and product life cycles. These might include sensor fusion to improve machine vision systems, tracer chemicals to track product and material flows, or identifying pollutants or toxins in waste streams (in addition to raw materials). My research also aims to improve machine learning algorithms used in existing waste material classification applications.