Projects: CeresAstro
====================

Statistical signal detection, Chandrayaan-2 data · Inter-IIT Tech Meet internal
selection (a past Inter-IIT problem) · Sep – Oct 2026

Method
------

I chose the statistical approach for detecting solar X-ray bursts in
Chandrayaan-2 XSM light curves, based on published flare studies:

- a rolling-percentile baseline
- a robust MAD (median absolute deviation) noise scale
- Gaussian–exponential pulse fits for burst shapes

Evaluation
----------

- One-to-one matching of detections to reference events
- Precision, recall, F1, and false alarms per observed hour
- A written analysis of the method's limits

Result
------

At fixed settings, the tool flags 36 candidate burst windows across ~100
observed hours. It runs as a local web app.