Dissertation Announcement for Robiulhossain Mdrafi – 03/24/2022 at 4:00 PM

March 11, 2022

Dissertation title: Data-Driven Sparse Computational Imaging with Deep Learning

When: Thursday, March 24, 2022, 4 PM – 6 PM

Where: In person-Simrall-228 (Conference room)
Remote access: https://msstate.webex.com/meet/acg540

Candidate: Robiulhossain Mdrafi

Committee Members:
Dr. Ali Cafer Gurbuz
(Major Professor)

Dr. Mehmet Kurum
(Committee Member)

Dr. John E. Ball
(Committee Member)

Dr. Jenny Du
(Committee Member)

Inverse imaging problems deal with the reconstruction of images from the sensor measurements where sensors can take form of any imaging modality like camera, radar, X-ray, MRI, and so on. In an ideal scenario, we can reconstruct the images via applying an inversion procedure from these sensors’ measurements, but in the real practical applications, the measurement acquisition process from these sensors is heavily corrupted by the noise, the forward model is not exactly known, and non-linearities or unknown physics of the sensors play roles. Hence, perfect inverse function is not exactly known for immaculate image reconstruction. To this end, in this dissertation, we propose an automatic sensing and reconstruction scheme based on deep learning within the compressive sensing (CS) framework to solve the computational imaging problems. Classical CS utilizes pre-determined linear projections in the form of random measurements and convex optimization with a known sparsity basis to reconstruct signals. Here, we develop a data-driven approach to learn both the measurement matrix and the inverse reconstruction scheme for a given class of signals, such as images. We also propose to extend our analysis to introduce data driven models to directly classify from compressed measurements through joint reconstruction and classification. We develop constrained measurement learning framework and demonstrate higher performance of the proposed approach in the field of typical object and hyperspectral image classification tasks. Finally, we also propose a single data driven network that can take and reconstruct images at multiple rates of signal acquisition. The proposed data driven approaches can work as the pivotal factors for future research to accomplish task-specific smart sensors in several real-world applications.