3. Using an RGB camera.
Previous page | Next page | Table of
Contents
We are studying the ability of polarized light to image the early
onset of esophageal cancer to assist basic research on the initiation
and progression of cancer in a mouse model. A freshly excised tissue
sample (esophagus) from a mouse was placed in saline between a glass
slide and coverslip.
A standard RGB camera acquired images using three channels: RED (1),
GREEN (2), and BLUE (3). The protocol was as follows:
- The images were analyzed using MATLAB software. Images were acquired
as an 8-bit unsigned integer RGB image array,
image(x=1:512,y=1:384,color=1:3), and converted to floating point for
mathematical operations on each channel.
.
- Reference images were acquired of a white 100% reflectance standard
for the purpose of color balance, which means correcting for the
spectral content of the light source and the spectral response of the
camera. The mean pixel values of the sum image, PAR + PER, for the RED,
GREEN and BLUE channels were determined, and called Rstd, Gstd, and
Bstd, respectively.
.
- Tissue images of the esophagus sample were acquired, and normalized
by the Rstd, Gstd and Bstd values to yield reflectance units, R
(dimensionless).
.
- After normalization, the difference image, PAR - PER, was
calculated for each red, green and blue channel yielding pixel values
expressed in reflectance units, 0 < R < 1.
.
- The red, green and blue channels were multiplied by a scalar factor
to bring the R values up toward a maximum of value of 255 pixel units.
This multiplication constituted an increase in image brightness but did
not affect the color balance. Then the three channels were converted
back to an unsigned 8-bit integer image for dispaly as an RGB image
array, RGBimage(x,y), for display.
The result was an RGB image where each color channel (red, green,
blue) was based on the difference image, PAR - PER. The image was
sensitive to the superficially scattered light, and the colors represent
the colors viewed only with superficially scattered photons.
Previous page | Next page | Table of
Contents