The problem of establishing a disciplined method for characterizing features of data sets and defining a method for comparative analysis of those features is fundamental to most lines of scientific inquiry. The DDMA team has a strong background in building diverse measures, metrics, and data transformations for the more challenging areas of this field.

One of the most difficult problems is defining measures for shapes and images. Domain experts typically obtain qualitative comparisons by looking for features in data such as apparent differences in value statistics, spatial distributions of intensity, and small-scale features such as vortex and shock structures. What is required is a process that can be automated, repeated and defended. DDMA develops tools that are robust to inherent noise and fuzziness present in data yet sensitive to the subtle features of interest.

• Theory
• Methods
• Applications