Data Discovery in Images Using a Median-Absolute-Deviation Based Filter

The utility of a recently developed, novel, computational digital image curation tool called SOCKS (Statistical Outlier Curation Kernel Software) is demonstrated in this paper. Raw image-data the James Webb Space Telescope (JWST) used for tests. Data decontaminated through an iterative, parametrically tuned sequence operations involving conditional flagging outliers, followed by flagged pixels. Every iteration produces and outlier image, till converges asymptotically producing final that represents ground-truth as accurately possible. intermediate images are shown to contain wealth information about original image. An iteration-by-iteration approach deriving insight out described detail. Traditional filtering methods, contrast, discard outliers causing irreversible loss contained therein.