Program Summary
Recent imaging techniques inspired by the mathematics of quantum information claim to enable imaging beyond the traditional Rayleigh resolution limit.
The approach is an example of “modal imaging” and is just one recent example of a much broader group of techniques developed over many decades that attempt to efficiently extract information from multidimensional systems using non-Fourier measurement bases. In almost all of these cases, prior information enables more efficient or more accurate information recovery from measurement results. More importantly, even without prior information, intelligently and continually adapting the measurement basis in response to earlier measurement results enables superior performance for many information retrieval tasks.
The Investigating Adaptive Modal Bases for Intelligent Classification (IAMBIC) Disruption Opportunity (DO) is an effort to create new frameworks for analyzing when such adaptive measurement bases can provide significant, real-world advantage for important problems.
IAMBIC performers will develop a cohesive and general theoretical framework for adaptively varying measurement bases in response to initial measurement data. They will apply their adaptive measurement framework on two (or more) tasks: The first task will address the DARPA-defined Constellation Challenge problem of imaging groups of unresolved sources beyond the Rayleigh resolution limit, in order to quantitatively bound the conditions under which adaptive modal imaging provides a positive real- world impact for the specific problem of sub-Rayleigh-resolution imaging. The other measurement tasks are performer-defined. The performers will analyze these tasks using their adaptive measurement framework and show that their strategy provides a significant real-world impact.