Activity Number:
|
266
- Recent Advances in Spatial-Temporal Modeling and Its Applications
|
Type:
|
Topic Contributed
|
Date/Time:
|
Tuesday, August 4, 2020 : 1:00 PM to 2:50 PM
|
Sponsor:
|
Korean International Statistical Society
|
Abstract #313335
|
|
Title:
|
Physics Driven Dynamic Imputation with Application to Pollution Satellite Images
|
Author(s):
|
Youngdeok Hwang* and Won Chang and Hang Kim
|
Companies:
|
Baruch College, CUNY and University of Cincinnati and University of Cincinnati
|
Keywords:
|
|
Abstract:
|
Satellite images using multiple wavelengths channels provides measurements of a wide area, which can be crucial for understanding the nature of pollution generation and transport. They can be, however, masked by the presence of the cloud in the area and contain many missing cells. In this paper, we propose a method to impute the missing cells in satellite images by taking into account for pollution transport by wind. The proposed method incorporates a fundamental physics equation to construct the covariance structure. To mitigate computational challenges due to a large number of spatial locations we adapt stochastic gradient descent based on randomly sampled locations. We demonstrate the benefit of our methodology through a case study.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2020 program
|