Enhancing Laser Ptychography: Development of Accelerated system

In our laboratory, we are developing an innovative lensless microscope that utilizes coherent diffractive imaging (CDI) to generate 3D profiles of samples. Unlike traditional microscopes that rely on an microscope objective lens to transmit images of the specimen to the detector, our approach uses structured-coherent illumination from a laser to capture diffraction patterns of the object. Many diffraction patterns are collected by a CCD camera on a sample scanning system and a complex amplitude image, including phase, can be reconstructed from data without requiring a lens.

One of the key advantages of ptychography over conventional microscopy is its ability to retrieve the object's phase state. This information provides details about the optical path difference to the sample's surface for a given refractive index in the sample material, which yields surface topography. The phase is essential for measuring transparent objects.

The objective of this project is to improve the laser ptychography microscope and its imaging algorithm. Although off-axis illumination in the system causes distortion of the diffraction image, we have addressed this issue through numerical calculations. The current algorithm is written in MATLAB and requires significant CPU-based computation that can be accelerated on GPUs.

To reduce processing time and enable the use of the microscope on a wider range of samples, your task will be to learn the CUDA programming framework of Nvidia for implementing the image reconstruction method on our high-performance GPU-based workstations.

Type of Project: Thesis; Internship

Master's degree: Master of Engineering Technology; Master of Science; Master of Bioengineering

Master program: Bioscience Engineering; Computer Science

For more information or application, please contact Hyun-Su Kim (Hyun-su.Kim@imec.be)

Only for self-supporting students.