![Dissertations and Theses](../../assets/md5images/e2b873e7ba6b45a43f7aee27fdcee4aa.jpg)
Dissertations and Theses
Date of Award
2019
Document Type
Thesis
Department
Computer Science
First Advisor
Jianting Zhang
Second Advisor
Akira Kawaguchi
Keywords
NEXRAD, GOES-16, NOAA, NCEI, AWS, Detectron
Abstract
Big Data has been playing a major role in the domain of Deep Learning applications as many companies and institutions continue to find solutions and extract certain trends in fields of climate change, weather forecasting and meteorology. This project extracts weather events data from multiple data sources that are supported by National Centers for Environmental information (NCEI) [1] and Amazon Web Services (AWS) [2]. Data sources include Next-Generation NEXRAD [3] Doppler radar reflectivity, GOES-16 [4] multi-channel satellite imagery and NCEI [1] storm events. Then, it integrates and refines data in proper formats to be fed to the open-source Detectron [5] Deep learning software package from Facebook. The integration process involves validation on the respective data source as well as generating geospatial and temporal intersections. The project subsequently shifts to generating training datasets along with annotations to be ingested by Mask R-CNN [6] network architecture. Finally, it passes the generated training dataset as an input for Detectron [5] software application and attempts to train network for the given 2017 and 2018 storm events.
Recommended Citation
alanbari, haidar A. Mr, "Integrating Multi-Source Weather Data for Deep Learning" (2019). CUNY Academic Works.
https://academicworks.cuny.edu/cc_etds_theses/887
Included in
Data Storage Systems Commons, Other Computer Engineering Commons, Other Engineering Commons