Multi-Spectral Imaging of Vegetation with a Diffractive Plenoptic Camera
Abstract
Snapshot multi-spectral sensors allow for object detection based on its spectrum for remote sensing applications. By making these sensors more light-weight, it allows for increased performance for drones and lower costs for satellites. Toaddress this need, I designed and built a diractive plenoptic camera (DPC) which utilized a Fresnel zone plate and a light field camera to detect vegetation via a normalized dierence vegetation index (NDVI). This thesis aims to derivedesign equations of the DPC to get an estimate on its performance and evaluate the DPC's multi-spectral performance. The experimental results yielded that the spectral resolution near the design wavelength was 25 nm which exceed theestimated value of 6:06 nm. In capturing multi-spectral scenes, the DPC was able to identify vegetation in the scene at reduced values in comparison to a grating spectrometer. Additionally, artifacts unique to the DPC were found anddiscussed. The future of this work includes using a dierent diractive optic design, using an intermediate image diffractive plenoptic camera, and using 3D deconvolution techniques.
Document Details
- Document Type
- Technical Report
- Publication Date
- Mar 01, 2020
- Accession Number
- AD1108424
Entities
People
- Tristan R. Naranjo
Organizations
- Air Force Institute of Technology