Show HN: AstroLens – AI that watches the sky and finds what nobody catalogued (github.com)

AstroLens, an innovative open-source AI tool, has been unveiled to autonomously scan the sky for unusual astronomical objects, marking a significant advancement for both amateur astronomers and researchers in the AI/ML community. With the ability to download and analyze images from major astronomical surveys 24/7, AstroLens identifies anomalies and re-evaluates its findings without human supervision. In a recent three-day test, it successfully analyzed over 20,000 images, flagged nearly 3,500 anomaly candidates, and independently recognized known astronomical objects such as supernovae and gravitational lenses, achieving an impressive 99.5% accuracy in its transient detection model. The significance of AstroLens lies in its robust machine learning architecture, which integrates a Vision Transformer for classification and employs an ensemble of out-of-distribution (OOD) detection methods to assess anomalies. Its self-correcting capabilities not only enhance detection accuracy over time but also provide seamless integration with existing astronomical databases for cross-referencing discoveries. The tool serves as a powerful resource for ML engineers seeking to apply advanced techniques in real-time data analysis, offering a unique glimpse into the potential for AI-driven astronomy to uncover new cosmic phenomena.