Subject

Ptychographic microscope uses overlapping diffraction images for high-resolution 3D imaging.
This method has widespread applications, such as in semiconductor and biomedical imaging. The challenge lies in the management of large amounts of data captured by camera, necessary for high-quality image reconstruction, as the field of view depends on the number of scanned images. This thesis addresses the efficient representation and storage of such data.

Kind of work

The core objective of this thesis is to spearhead advancements in data compression techniques
specific to ptychographic microscope imaging. The aspect of data compression, specifically its
impact on image quality in ptychographic image reconstruction, is a relatively unexplored area, brimming with opportunities to conserve data storage. By exploiting the redundancy found in the raw data due to overlap, we aim to achieve high-quality image reconstruction without
compromising data integrity. However, our understanding of how compression artifacts influence the reconstructed image is presently limited.

Candidates possessing strong analytical skills and high motivation are encouraged to apply. This multidisciplinary project offers a rare chance to gain comprehensive knowledge in both microscope and data compression. For those interested in academia, the thesis promises a
great opportunity for publication due to its unexplored nature. The tentative activities proposed for the thesis are as follows:

1. Acquire foundational knowledge of both conventional and ptychographic microscopes.
2. Understand the basic concepts of data compression and decompression. Set up a test bench
to analyze compression artifacts objectively and subjectively, a low-risk task with high potential for publication.
3. Design a mathematical model to study the impact of compression noise on the quality of reconstructed images.
4. Utilize the overlap in raw diffraction images to improve the compression rate, a high-risk effort with potential for further exploration.

Framework of the Thesis

Our project's focus is to assess the ptychographic image recovery process with compressed data. Given that the data handled by ptychography depends on the number of image shots, resolution, and field-of-view (FOV), these elements will be scrutinized alongside image reconstruction speed and optimization strategies.
We plan to assess the influence of the fast calculations enabled by JPEG compression on the
quality of ptychographic images and explore the potential of parallel approaches for in-house
phase retrieval methods.

Number of Students

1