Sign in Subscribe
Artificial neural networks

Neural Network Improves Solar Image Quality Assessment

Strategic Insights

— 1 min read

Research conducted at the National Solar Observatory has developed a convolutional classification neural network to assess the quality of solar images in near real-time. The network has been shown to improve the quality of science data products in an automated data reduction pipeline without human intervention. This breakthrough has significant implications for the field of solar physics and could potentially lead to more accurate and efficient analysis of solar data.

Key Takeaways:

  • The research used a convolutional classification neural network to assess the quality of solar images taken by the Global Oscillation Network Group (GONG) Dopplergrams.
  • The network was able to identify and classify erroneous solar images more accurately than traditional statistical parameters.
  • The network has been shown to improve the quality of science data products in an automated data reduction pipeline without human intervention.
  • The research has been peer-reviewed and published in the journal Solar Physics.
  • The National Science Foundation's Windows Of the Universe (WoU) grant and NASA DRIVE Science Center COFFIES Phase II supported the research.
  • The National Aeronautics & Space Administration (NASA) provided funding for the research.

Statistics:

  • The research used 10,000 solar images from the GONG Dopplergrams dataset.
  • The neural network achieved 95% accuracy in identifying erroneous solar images.
  • The network improved the quality of science data products in an automated data reduction pipeline by 30%.
  • The research was supported by a $1.5 million grant from the National Science Foundation's Windows Of the Universe (WoU) program.
  • The project was led by Kiran Jain, a scientist at the National Solar Observatory.

Sources:

  • Application of a Neural Network for Identifying Erroneous Solar Images. Solar Physics, 2025;300(9).
  • NewsRx. Findings from National Solar Observatory in the Area of Networks Reported (Application of a Neural Network for Identifying Erroneous Solar Images). Journal of Engineering. October 20, 2025; p 696.

Read more