Compressive Inference

• Activity Number: 415
• Type: Topic Contributed
• Date/Time: Tuesday, August 6, 2013 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics in Imaging
• Author(s): Weihong Guo*+ and Garvesh Raskutti and Jiayang Sun and Grace Yi Wang and Dan Yang
• Companies: CWRU and SAMSI and Case Western Reserve University and SAMSI and SAMSI
• Keywords: compressive sensing ; imaging ; inference ; MCP

Abstract

Compressed or compressive sensing is a technology to reduce the size of data from the very beginning -- data collection stage. Through sampling much less data, it also reduces imaging time and cost, making it very useful in applications arising in, for example, astronomy, medical imaging, and sensor networks, especially when dealing with massive data. In medical applications for instance, less data means less radiation in some cases.

Compressive sensing reconstruction takes advantage of the signal's sparsity in some transform domain to recover the underlying image of interest. This talk addresses how to make an inference about the true underlying image from the few incomplete measurements. We develop a general and practical multiple comparison (MCP) inferential procedure to adapt to compressive sensing data. The procedure is able to make inference with and without image reconstruction.