1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
|
function [theColorImage,reconNewWay,reconOldWay] = BitsPlusPackMonoImage(theMonoImage)
% [theColorImage,reconNewWay,reconOldWay] = BitsPlusPackMonoImage(theMonoImage)
%
% In Mono++ mode, the Bits++ box uses the red and green
% channels to provide 14-bits per pixel true intensity
% resolution. The blue channel is set to 0 to let the
% monochromatic image through. Empirically, we concluded
% that the packing is left adjusted. That is, the 8 MSB
% of the 14-bit input go into the red channel, and the
% 6 LSB get left aligned in the green channel, with
% the remaining two bits set to 0.
%
% This routine packs the bits properly for this function.
%
% 11/17/03 dhb, ip Wrote it.
% 8/13/04 dhb Fix bug, the data were not packed quite right.
% 2/26/07 mk Bugfix for LSB conversion: Added modulo operation.
% 3/01/07 mk Bugfix for MSB conversion: Added floor operation.
% 3/04/07 dhb Modified to return some debugging information and
% compare original with recent version.
[m,n] = size(theMonoImage);
theColorImage = uint8(zeros(m,n,3));
% Red channel gets high 8 bits
theColorImage(:,:,1) = uint8(floor(theMonoImage/256));
% Green channel gets low 6 bits, left justified
theColorImage(:,:,2) = uint8(mod(theMonoImage, 256));
% Optionally create debug images for testing:
if nargout > 1
% This is the code from the original OS 9 version. Buggy?
oldRed = uint8(bitshift(theMonoImage,-6));
oldGreen = uint8(bitshift(theMonoImage,2));
% Reconstruct input vales from the two bytes produced for the packing
reconNewWay = double(theColorImage(:,:,1))*256+double(theColorImage(:,:,2));
reconNewWay = bitshift(reconNewWay,-2);
reconOldWay = double(oldRed)*256+double(oldGreen);
reconOldWay = bitshift(reconOldWay,-2);
end
|
