Department of Information Systems
University of Maryland, Baltimore County
Baltimore, Maryland, U.S.A.
Email: serajmost _at_ umbc_dot_edu
Lab: ITE 471
Profiles: Google Scholar, LinkedIn
Short Biography
Seraj is a Ph.D. student at the Department of Information Systems working with Dr. Jianwu Wang. His main focus is using the AI approches in Earth Informatics research in the remote sensing domain. He began as a Ph.D. student in Fall 2021 to work on Ocean Eddy detection and localization project funded by NASA’s Jet Propulsion Lab (JPL). Later, he worked on identifying Gravity Wave in the upper atmosphere, another project funded by NASA goddard space and flight center. His current focus is on Distributed Analytics on HPC platform for large scale remote sensing data and complex deep neural network training to aid in Artificial Intelligence techniques for Earth Informatics. He is also working on designing an RL algorithm to optimize Cloud environments for such complex models with larger datasets. Most of his research work is supported by NASA, NSF and Research Computing of Division of IT (DoIT) at UMBC’s High Performance Computing.
Research Focus
-
- Artificial Intelligence | Computer Vision | Object Detection | Pattern Recognition
- Cloud Computing | Optimization | Distributed Analytics | HPC
- Climate Change | Earth Informatics
Invited Talk
-
- Seraj received an invitation to the U.S. Naval Research Lab (NRL) in Mississippi on September 12, 2023, to discuss his AI-driven Remote Sensing work. During the visit, he presented how AI techniques can be applied to Ocean Research, offering potential benefits to the U.S. Navy.
Education
-
- 2021 ~ ongoing: Ph.D. in Information Systems, University of Maryland, Baltimore County, Maryland, USA
- 2017 ~ 2020: M.Sc. in Computer Science, Montana State University, Bozeman, Montana, USA
- 2010 ~ 2012: M.Sc. in Computer Science and Engineering, Luleå University of Technology, Luleå, Sweden
- 2001 ~ 2004: B.Sc. in Computing Information Systems, London Metropolitan University, UK
Publication
-
- Enhancing Satellite Object Localization with Dilated Convolutions and Attention-aided Spatial Pooling, Seraj Mostafa, C Wang, J Yue, Yuta Hozumi, J Wang, Accepted, IEEE/AMLDS 2025
- gWaveNet: Classification of Gravity Waves from Noisy Satellite Data using Custom Kernel Integrated Deep Learning Method, SAM Mostafa, O Faruque, C Wang, J Yue, S Purushotham, J Wang, ICPR 2025 [springer, preprint]
- YOLO based Ocean Eddy Localization with AWS SageMaker, Seraj Mostafa, Jinbo Wang, Benzamin Holt, Jianwu Wang, IEEE/Big Data 2024 [publication link, arXiv]
- Machine learning based model for mesospheric gravity wave detection from noisy satellite images, J Yue, S Mostafa, C Wang, Y Hozumi, J Wang, S Purushotham, M Cham, AGU23 [link]
- CNN based Ocean Eddy Detection using Cloud Services, Seraj Mostafa, Jinbo Wang, Benzamin Holt, Jianwu Wang, IEEE/Big Data, IEEE Geoscience and Remote Sensing Society (IGARSS) 2023. [IEEE, Publication Link]
- Atmospheric Gravity Wave Detection Using Transfer Learning Techniques. Published by the 2022 IEEE/ACM 9th International Conference on Big Data Computing, Applications and Technologies (BDCAT 2022). [IEEE, Publication Link]
- Benchmarking Probabilistic Machine Learning Models for Arctic Sea Ice. Published by the 2022 IEEE International Geoscience and Remote Sensing Symposium (IGARSS 2022), IEEE. [IGARSS, Publication Link]
- More at Google Scholar
Awards
-
- Student Poster Award, College of Engineering and Information Technology (CoEIT) Research Day, UMBC, April 2024
- PhD Student Research Award, IS Student Research Symposium, Department of Information Systems, UMBC, May 2023
- NSF Student Travel Award, 2021 IEEE International Conference on Big Data.
Advising
-
- Summer 2022 REU program: A NSF funded program to support Research Experience for Undergraduates (REU) from various US universities. Seraj advised 5 students in a Transfer Learning Project to detect Gravity Wave from Remote Sensing Satellite Imagery.
- Advised Four Masters students for AWS cloud environments namely, SageMaker and EC2 to deploy RESTFull API services to automate Machine Learning model training and inferencing.