GAMA-1’s Haibing Sun to Present Cutting-Edge Deep Learning Research at AMS 105th Annual Meeting

Greenbelt, MD – January 2025

 

GAMA-1 Technologies is proud to announce that Haibing Sun, a key member of our team, will present his groundbreaking research, “Anomaly Detection and Classification in NOAA Satellite Imagery with Deep Learning,” at the American Meteorological Society’s (AMS) 105th Annual Meeting. The presentation is part of the 21st Annual Symposium on Operational Environmental Satellite Systems, scheduled from January 12 to 16, 2025, in New Orleans, Louisiana, and online.

Haibing Sun’s presentation will take place during the 21OESS Posters III session on Wednesday, January 15, 2025, at 3:00 PM Central.

Abstract Highlights: Transforming NOAA Satellite Data with Deep Learning

  • Satellite imagery from NOAA’s GOES ABI and NPP/J1/J2 VIIRS systems plays a critical role in monitoring hazards and supporting weather forecasting. However, visual anomalies—ranging from predictable issues linked to satellite maintenance to rare, random occurrences—often disrupt operations and impair the interpretation of imagery. Traditional methods of anomaly detection rely heavily on manual visual inspection and ancillary data, making them prone to error, complex, and time-intensive.
    To address these challenges, Haibing Sun has spearheaded the development of a binary classification deep learning model to automatically detect anomalies in NOAA satellite imagery. This work is now expanding to a multilabel model that identifies not just the presence of anomalies but also classifies their types and causes. By leveraging a curated training dataset and sophisticated evaluation methods, this innovative approach enhances anomaly detection accuracy, offering NOAA and its stakeholders a robust, efficient solution.
    Key elements of the research include:
    Deep Learning Advancements: Transitioning from binary to multilabel classification for nuanced anomaly detection and classification.
    • Data-Driven Insights: Analyzing training datasets, model performance, and anomaly classification outcomes.
    • Epoch Optimization: Exploring the balance between underfitting and overfitting to achieve optimal prediction accuracy through techniques like early stopping.

GAMA-1’s Commitment to Innovation

This presentation underscores GAMA-1’s dedication to advancing environmental science through cutting-edge AI and machine learning solutions. By empowering NOAA and its partners with innovative tools, we continue to strengthen the operational reliability and efficiency of critical satellite systems.
Join us in supporting Haibing Sun’s important contribution to the field of operational environmental satellite systems. His work not only demonstrates the transformative power of AI but also reflects GAMA-1’s unwavering commitment to excellence and innovation.
For more information about the AMS Annual Meeting, visit https://www.ametsoc.org/.

About GAMA-1 Technologies, LLC

GAMA-1 Technologies is committed to leveraging science and data to create impactful solutions for humanity. As a certified Small Disadvantaged Business, we specialize in secure, high-performance enterprise technologies that support future scientific endeavors. Our core principles—reputability, reliability, and results—guide our mission to manage public scientific data and environmental intelligence effectively.

Our certifications and partnerships underscore our commitment to excellence. GAMA-1 is an ISO 9001, ISO 20000, and ISO 27001 certified organization, adhering to CMMI and ITIL best practices. As an AWS Advanced Tier Services Partner with the Migration Competency and Public Sector Partner Designation, a Scaled Agile Silver Partner, and a Weather-Ready Nation Ambassador, we combine technical expertise with mission-driven objectives to deliver meaningful solutions. Our goal is to be the trusted partner in digital transformation, earth observations, and applied science technology, leveraging our experience to enable data-driven benefits for humanity.

A Great Place to Work

In the competitive landscape, GAMA-1 stands out as a company that fosters a family-oriented, small-business culture. Our workplace cultivates camaraderie and community, with 97% of employees feeling this strong sense of connection. We value individual contributions and provide opportunities for career growth and interaction with senior executives, attracting candidates who seek meaningful engagement and a rewarding career trajectory.

GAMA-1 Technologies, LLC is more than just a company; we are a partner in driving meaningful change. With certifications including ISO 9001, ISO 20000-1, ISO 27001, CMMI, and ITIL best practices, we align technical expertise with mission objectives to craft solutions that matter. As an Amazon Web Services Advanced Tier Partner, Scaled Agile Silver Partner, and Weather-Ready Nation Ambassador, we combine competence and commitment to create a better future. Learn more about us at www.gama1tech.com and join us in shaping a brighter tomorrow.

Contact:
Stan Coachman, Contracts Manager – 301-982-4262, [email protected]


GSA Multiple Award Schedule (MAS)


SIN 54151S – Information Technology Professional Services

SIN 518210C – Cloud Computing and Cloud Related Services

OLM – Order-Level Materials (OLM’s)


Contract Nº: GS-35F-131CA


Contract End Date: January 3, 2030

* GSA may exercise a total of up to 2 additional          5-year option periods under the current contract.


GSA Multiple Award Schedule (MAS)


SIN 54151S – Information Technology Professional Services

SIN 518210C – Cloud Computing and Cloud Related Services

OLM – Order-Level Materials (OLM’s)


Contract Nº: GS-35F-131CA


Contract End Date: January 3, 2030

* GSA may exercise a total of up to 2 additional          5-year option periods under the current contract.


GSA Multiple Award Schedule (MAS)


SIN 54151S – Information Technology Professional Services

SIN 518210C – Cloud Computing and Cloud Related Services

OLM – Order-Level Materials (OLM’s)


Contract Nº: GS-35F-131CA


Contract End Date: January 3, 2030

* GSA may exercise a total of up to 2 additional          5-year option periods under the current contract.

  • Discovery
  • Analysis
  • Vision Definition and Development
  • Documentation