Infrared color-coded aperture optimization for object tracking and spectral classification
Abstract
Compressive spectral imaging (CSI) acquires spectral and spatial information using less amount of data than the traditional scanning-based methods. CSI state-of-the-art has demonstrated that the color-coded aperture (CCA) design allows achieving improvements in the object tracking and spectral classification tasks in the compressive domain. However, it is essential to remark that most of the research is focused on the visible range of the electromagnetic spectrum, which is useless to the infrared spectral range because it does not consider some specific restrictions, for example, the sensor pixel size. Furthermore, conventional object tracking, and spectral classification approaches rely on universal spatial and spectral distributions to design the coded aperture masks, i.e., these approaches do not consider the available prior information about the materials' spectral data in the infrared spectrum. Therefore, designing and implementing a CCA considering statistical information in the infrared spectrum would greatly help to perform object tracking and spectral classification. This project aims to mathematically design the spectral response and spatial structure of a color-coded aperture (CCA) in a compressive infrared imager based on a coded aperture optimization model, considering the predetermined prior target spectral information. Specifically, The CCA structure model exploits the target statistics, spectral behavior, and physical considerations of the compressive infrared systems, which result in improvements in object tracking and spectral classification. At the same time, this approach increases the framerate and reduces the computational complexity and data storage costs of the system. The main novelty of this proposal is introducing the CCA optimization concept for object tracking and spectral classification in the infrared domain, where additional considerations about the propagation phenomena and prior target spectral information need to be included.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Jan 21, 2022
- Source ID
- FA95502110326XX0
Entities
People
- Henry Argüello
Organizations
- Air Force Office of Scientific Research
- Industrial University of Santander
- United States Air Force