Paper
14 May 2018 Synthetic neural network calibration of a hyperspectral imaging camera
Michael W. Kudenov, Clifton G. Scarboro
Author Affiliations +
Abstract
The use of machine learning algorithms is used extensively when analyzing and labeling image data. However, these techniques also offer additional post-processing advantages when they are used to calibrate sensor data. In this paper, a new calibration strategy for a Snapshot Hyperspectral Imaging Fourier Transform (SHIFT) spectrometer is discussed. The method, which is based on use of artificial neural networks, offers greater accuracy in both space and frequency when compared to conventional post-processing techniques. A theoretical model for the calibration procedure is presented, the laboratory data collection protocol is provided, and validation is conducted by measuring NIST-traceable spectral reflectance standards.
© (2018) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Michael W. Kudenov and Clifton G. Scarboro "Synthetic neural network calibration of a hyperspectral imaging camera", Proc. SPIE 10656, Image Sensing Technologies: Materials, Devices, Systems, and Applications V, 1065607 (14 May 2018); https://doi.org/10.1117/12.2305521
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Sensors

Calibration

Spectroscopy

Neural networks

Hyperspectral imaging

Cameras

Fourier transforms

Back to Top