Accelerating material imaging with hyperspectral microscopy and computer vision

Abstract: Hyperspectral microscopy captures samples across a wide range of wavelengths, generating massive 3D data on the optical properties of materials. With computer vision, we can analyze this data more effectively, using algorithms to identify patterns, reduce noise, and extract features.

In this presentation, I will discuss the application of hyperspectral imaging on semiconductor materials and optoelectronic devices. I will explain how we can surpass the equipment limit to capture fast material degradation using deep-learning image restoration algorithms. I will then demonstrate the combination of computer vision and hyperspectral imaging in a high-throughput self-driving lab scenario, where the optical bandgap of new materials has been characterized 5000x faster than conventional UV-Vis methods.

Bio: Kangyu Ji is currently a postdoctoral researcher at the Massachusetts Institute of Technology. Previously Dr Ji obtained a Bachelor’s degree in Materials Science and Engineering from Imperial College London, and Master and Doctor of Philosophy degrees in Physics from the University of Cambridge.