Data Proc and Acqui SIG: Poisson Binomial-Based Random Undersampling Scheme for ... - Nov 15th
Complete Title: Poisson Binomial-Based Random Undersampling Scheme for Near-Surface Compressive Sensing
This is a virtual event
NOTE: You Must Be Logged In to Register.
Speaker: Sayan Mukherjee, PhD Scholar, Purdue University
Compressive sensing (CS) is a signal processing technique for efficient acquisition and reconstruction of a signal from fewer samples by exploiting the sparsity of the signal. CS has been widely used for speech recognition and different imaging techniques like photography, holography, magnetic resonance imaging, and so on. A key element of CS-based seismic data acquisition is the choice of an efficient sampling scheme. Among the existing methods, the uniform distribution-based methods sample the traces with equal probability over offset, reducing sampling randomness. Though the Bernoulli distribution-based methods use varying probability, without any direction in the choice of probability, this approach may produce inferior results. The jittered uniform and Bernoulli distribution-based sampling methods also limit the randomness that may result in spatial aliasing. In this webinar, the speaker will present a segmented Poisson Binomial undersampling scheme, which is motivated by the fundamentals of surface wave propagation. The proposed approach is advantageous, as it uses fewer segments than the jittered method to preserve the randomness. It also guides the probability over segments to control the gap size between sampled traces and produces better-reconstructed traces.
Speaker Biography: Sayan Mukherjee, Purdue University
Sayan is a bachelor's in Civil Engineering (2016) from the Indian Institute of Engineering Science Technology (IIEST), Shibpur, and a master’s in Geotechnical Engineering (2021) from the Indian Institute of Technology (IIT), Madras. After his B.E., Sayan worked in the government-owned, Coal India Ltd. for three years, administering multiple constructions. Currently, he is pursuing his Ph.D. at the Sustainability Geophysics Project at the department of Earth, Atmospheric, and Planetary Sciences, Purdue University, working with Dr. Yunyue Elita Li. His current work is focused on the compressive-sensing-based design of an optimum sensor array predominantly for seismic surface wave acquisition.
**Access information will be sent to all registrants after registration closes.
Price List:
|
Pre-Registered |
Late/Walk-Up
|
Member Virtual
|
Free |
Free |
| Non-Member Virtual |
$10 |
$10 |
** Please be aware that ALL registrations must go through the 'Checkout', even if your total is zero. Registration is not confirmed until you complete the checkout process.