Page 148 - Computational Colour Science Using MATLAB

IMPLEMENTATIONS AND EXAMPLES 135

% polynomial transforms obtained from getlincam
% caldata is a 3 by 4 matrix produced from getlincam.m
% RGB is an n by 3 matrix of RGB values (in range 0-255)

RGB = RGB/255;
imsize = size(RGB);
RGB = reshape(RGB, prod(imsize)/3, 3);

for i=1:3
x = polyval(caldata(i,:),RGB(:,i));
RGB(:,i) = x*255;
end

RGB(RGB0) = 0;
RGB(RGB255) = 255;

RGBout = reshape(RGB,imsize);

The format for lincam is

[tRGB] = lincam(CALDATA,RGB)

where CALDATA is a 364 matrix obtained from getlincam and RGB is an
m6n63 matrix of camera RGB values where the image size is m6n. The output
of the function is an m6n63 matrix of linearized camera RGB values. Unlike
many of the functions in the earlier part of this book, this function operates on
the whole image with a single function call. In this respect the MATLAB
function reshape is extremely useful.
An image may be read into MATLAB using the simple instruction

image = imread(‘test.tif’);

where test.tif is the name of the image file. For an RGB image of size
1276101, for example, the variable image would now be defined by

image 127610163 38481 uint8 array

It is necessary to use the double command, thus

image = double(image);

in order to convert the variable image from UINT format to DOUBLE format.