←Back to feed
🧠 AI⚪ NeutralImportance 4/10
A Semi-Supervised Learning Method for the Identification of Bad Exposures in Large Imaging Surveys
arXiv – CS AI|Yufeng Luo, Adam D. Myers, Alex Drlica-Wagner, Dario Dematties, Salma Borchani, Francisco Valdes, Arjun Dey, David Schlegel, Rongpu Zhou, DESI Legacy Imaging Surveys Team||7 views
🤖AI Summary
Researchers developed a semi-supervised machine learning pipeline using vision transformers and k-Nearest Neighbor classifiers to automatically detect poor-quality exposures in astronomical imaging surveys. The method was successfully applied to the DECam Legacy Survey, identifying 780 problematic exposures that were verified through visual inspection.
Key Takeaways
- →A new ML pipeline combining vision transformers with k-NN classifiers can automatically detect poor-quality astronomical images.
- →The method addresses the scalability challenge of manual visual inspection in large astronomical surveys.
- →The pipeline successfully identified 780 problematic exposures in DECaLS Data Release 11.
- →The approach offers a scalable solution for quality control in other large imaging surveys.
- →Self-supervised learning techniques prove effective for astronomical image quality assessment.
#machine-learning#computer-vision#vision-transformers#astronomy#image-processing#quality-control#self-supervised-learning#k-nearest-neighbor
Read Original →via arXiv – CS AI
Act on this with AI
Stay ahead of the market.
Connect your wallet to an AI agent. It reads balances, proposes swaps and bridges across 15 chains — you keep full control of your keys.
Related Articles