|
Jim
Mansfield Spectroscopic
Arthritis Skin Oxygenation
Imaging Works of Art
Personal
|
|
|
|
Visible and Near-IR Spectroscopic Imaging Spectroscopic imaging
has been around for decades, but the majority of its original usage was in
satellite and aircraft based systems for geological, agricultural and
military applications. However, advances in CCD and liquid crystal tunable
filter (LCTF) technologies have allowed the development of inexpensive
spectroscopic imaging systems that operate in the visible and near-infrared. Using spectroscopy
alone, one obtains information about the optical properties of a sample from
a single region. Using imaging, one obtains spatial heterogeneity information
about a sample across a single spectral range, or, in the case of colour
imaging, in each of three spectral ranges (red, green and blue). Spectral
imaging, as its name implies, combines these two methodologies, such that one
obtains complete spectral information about a sample at every pixel in an
image.
There are many means of
doing spectral imaging. The result of spectral imaging data collection,
however it is done, is a spectral imaging datacube, or hypercube, or even
just imagecube. This three-dimensional array can be thought of as an array of
spatially resolved individual spectra, with every pixel in the first image
consisting of an entire spectrum, or as a series of spectrally resolved
images. In either representation, the 3D data cube can be treated as a single
entity and it can contain enormous amounts of spatial and spectral
information about the sample from which it was acquired. LCTFs are an electronically
tunable form of a Lyot filter, with each stage of the Lyot filter having a
liquid crystal retarder stage that allows the interference characteristics of
that stage to be modified, or tuned. CRI is the only commercial supplier of LCTFs
in the world. There are some excellent
descriptions of how LCTFs work on the web. There are many means of
obtaining spectral images, but if one is using an LCTF, then one takes an
image at a large number of wavelengths across the range of the filter. At
right is shown a representation of a spectral imaging “data cube” (or image
cube or hypercube). Each plane in the dataset represents one image taken at a
single wavelength. By taking images at each wavelength, one builds up the
image cube. Following any given pixel through the image cube, then, gives us
the spectrum of the sample at that point.
Schematic
of System Shown at right is a
schematic of a multispectral imaging system. This is CRI’s Nuance system,
however, the components are common to most spectral imaging systems that use
LCTFs. Created
by Jim Mansfield. Last Update: April
1st, 2002. |
