Give

Graduate Exam Abstract


Ryan Gooch

Ph.D. Final
December 16, 2019, 10:00 am - 12:00 pm
ENGR B105
Transfer Learning and Weather RAdar

Abstract: This work presents the culmination of the doctoral research by the author in exploring modern methods of Data Discovery in weather radar data, improvements in the cyberinfrastructure concerning multi-dimensional gridded data, with a concentration on real-time data streaming, and experimental use cases involving real world datasets.
Included in this work is a successful method for the classification of weather radar image data using convolutional neural networks, with inspiration drawn from the subfield of Transfer Learning in the Computer Vision community.
Once this model was developed, it was deployed on single radar data from each of the radars in the CASA DFW network to assign labels to support a human-in-the-loop semi-supervised method for data discovery in the weather radar scans.
This model has been further trained on the dataset of increased magnitude to demonstrate the model’s generalizability, and its utility in discovering phenomena of interest in vast datasets.
This work discusses the end-to-end development of the data discovery system, with special focus on initial data labeling, choices and tradeoffs in model architecture, and training concerns in the machine learning model.
This represents the first published research known to the authors on utilizing the power of transfer learning to transfer the learning of high quality convolutional neural networks trained on photographic images to the weather radar image domain.
Finally, we examine the current applications of the deep learning technologies developed in this research when applied to real-time streaming weather radar image data, using CHORDS as a platform.


Adviser: Chandra
Co-Adviser: N/A
Non-ECE Member: Jose Chavez, Civil Engineering
Member 3: Margaret Cheney, ECE/MATH
Addional Members: Sid Suryanaryanan, ECE

Publications:
Gooch, S. R. and Chandrasekar, V. (2019). Improving Historical Data Discovery in Weather Radar Image Datasets using Transfer Learning. IEEE Transactions on Geoscience and Remote Sensing. Manuscript submitted for publication.

Gooch, S. R., & Chandrasekar, V. (2019, July). Classifying Meteorological Echoes in Weather Radar Images with Transfer Learning. In IGARSS 2019-2019 IEEE International Geoscience and Remote Sensing Symposium.. IEEE.

Gooch, S.R., Chandrasekar, V., & Daniels, M. (2019, June). Visualization and Storage of Weather Radar Data with CHORDS. EarthCube Annual Meeting, Denver, CO.

Gooch, S. R., & Chandrasekar, V. (2018, July). Advances in Real-Time Weather Radar and Ground Sensor Data with Chords. In IGARSS 2018-2018 IEEE International Geoscience and Remote Sensing Symposium (pp. 3884-3887). IEEE.

Gooch, S.R., Chandrasekar, V., & Daniels, M. (2018, June). Real-Time Weather Radar Data Integration With CHORDS. EarthCube Annual Meeting, Alexandria, VA.

Gooch, S.R., & Chandrasekar, V. (2018, January). Measuring Connectivity in the Earth Sciences. National Radio Science Meeting, Boulder, CO.

Gooch, S.R., & Chandrasekar, V. (2017, July). Integration of real-time weather radar data and Internet of Things with cloud-hosted real-time data services for the geosciences (CHORDS). In 2017 IEEE International Geoscience and Remote Sensing Symposium (IGARSS) (pp. 4519-4521). IEEE.

Rubin, K., Aronson, E., Cruz, I., Hills, D., Daniels, M., Gooch, S.R., & Stamps, D.S. (2019, September). EarthCube Leadership Council 2019 NSF Solicitation Guidance. Retrieved from URL https://docs.google.com/document/d/1w5Mj-gIxWga-QRHZCej2rsZEl-aKobAp6Z0CTbsKnig

Rubin, K., Aronson, E., Cruz, I., Hills, D., Daniels, M., Gooch, S.R., Kelbert, A., & Stamps, D.S. (2019, September). EarthCube Leadership Council 2019 NSF Priorities. Retrieved from URL https://drive.google.com/open?id=12XBoNeRsLp9HJFIpVxejhgIGrtH1n7r0

Daniels, M., Kerkez, B., Chandrasekar, V., Graves, S., Stamps, D.S., Botnick, A., Martin, C., Gooch, S.R., Jones, J., Bartos, M., Keiser, K. (2019, June). Cloud-Hosted Real-time Data Services for the Geosciences (CHORDS): Bringing new observations to scientists in real-time. USGS From Big Data to Smart Data Community for Data Integration Workshop, Boulder, CO.

Daniels, M. D., Botnick, A., Martin, C. L., Stamps, D. S., Kerkez, B., Chandrasekar, V., Graves, S., Gooch, S.R., Bartos, M., Jones, J. & Keiser, K. (2019, April). CHORDS: Building the Internet of Things for the Geosciences (IoT-G). Research Data Alliance 13th Plenary poster, Philadelphia, PA.

Daniels, M. D., Botnick, A., Martin, C. L., Stamps, D. S., Kerkez, B., Chandrasekar, V., Graves, S., Gooch, S.R., Bartos, M., Jones, J. & Keiser, K. (2018, December). CHORDS: Building the Internet of Things for the Geosciences (IoT-G). In AGU Fall Meeting Abstracts.

Daniels, M., Stamps, D.S., Kucera, P., Kerkez, B., Chandrasekar, V., Graves, S., Dye, M., Martin, C., Keiser, K., Gooch, S.R., & Bartos, M. (2017, June). Recent Impacts and Discoveries Arising from the use of Cloud-Hosted Real-time Data Services for the Geosciences (CHORDS). EarthCube Annual Meeting, Seattle, WA.

Daniels, M. D., Kerkez, B., Chandrasekar, V., Graves, S., Stamps, D. S., Martin, C., Dye, M., Gooch, R., Bartos, M., Jones, J., Keiser, K. (2016). Cloud-Hosted Real-time Data Services for the Geosciences (CHORDS) software (Version 0.9). UCAR/NCAR - Earth Observing Laboratory. https://doi.org/10.5065/d6v1236q.



Program of Study:
MATH 550
ECE 513
STAT 525
ECE 656
CS 545
ECE 742
GRAD 544
N/A