| Summary: | Imaging systems work on an interplay of physics, electronics, and algorithms. Designing imaging systems by exploiting all these layers of abstraction can give rise to novel applications and interesting solutions to existing problems.
In this thesis, I will explore this design philosophy in 2 problems; automatic calibration of cameras that can see around corners, high resolution optical measurement of ultrasound. I will utilize advanced electronics hardware like femtosecond lasers, and Single Photon Avalanche Photodiode(SPAD) detectors for the automatic calibration project and software tools meant for machine learning in Python. For optical ultrasound measurement, I will use relatively simple hardware like solid state lasers, bench top optics, Field Programmable Gate Arrays (FPGA), and a synchronised global shutter CMOS camera.
Both projects are aimed at solving practical problems in their respective areas. Automatic calibration aims to improve the fidelity of images by algorithmically fixing calibration errors after taking measurements in cameras which can see around the corners. The optical ultrasound project aims to take medical ultrasound images at a very low cost and high resolution by using ideas from optics, laser vibrometry, and signal processing.
|
|---|