Anomaly Detection


This week (03/02 – 03/04) I am presenting a poster at the 2016 Conference on Data Analysis (CoDA) sponsored by the Department of Energy, titled “Anomaly Detection in Hyperspectral Imagery: Statistics- vs. Graph-based Algorithms.” The work captures three different aspects:

  1. algorithm performance under varied scene content,
  2. their empirical vs. theoretical outlier distributions, and
  3. comparisons between the three algorithms.

of three different anomaly detection algorithms:

  1. Reed-Xiaoli (RX),
  2. subspace-RX (SSRX), and the
  3. Topological Anomaly Detector (TAD)

The poster can be seen by clicking the link below.