Jennifer Newman has extensive experience in image processing, with her most recent work focusing in forensic applications of micro signals in images. Micro signals include small-magnitude signals such as those introduced when performing steganography, as well as photo-response non uniformity (PRNU) from camera sensors used to link a digital photograph to a device. Classification of images containing such low-magnitude signals require effective machine learning algorithms that can address issues including noise modeling, open-set classification, and data-dependency. Her recent project with the Center for Statistics and Forensic Evidence (CSAFE) produced an extensive image database constructed with mobile phone cameras and steganography apps (https://forensicstats.org/stegoappdb/). Another team project Jennifer is working on uses sophisticated clustering, spatial statistics, and machine learning methodologies to develop a tool to diagnose severe wind events (greater than 50 knots). The data contain a broad variety of sources, from highly-uncertain and low volumes of verified severe wind events, to reliable high-resolution data from model analyses. This NOAA proposal is under review but has been recommended for funding.