Author: Kriti Sen Sharma <>
Revision 1: 29th April 2012


In accordance with the Reproducible Research policy, this folder contains 
Scripts and Data for regenerating figures in paper on Few-View Micro-CT.

	- The scripts are named RR1_.., RR2_.. and so on (hereinafter referred 
	to as “the codes”)
	- Data used in these codes are stored in the ‘data’ folder (See note 
	‘Data’ below for more details on data).


	First, 2 external libraries need to be installed for using these codes:

i > ** OpenCT ** 
(library for image reconstruction)
A copy of this library which has been tested to work with the codes RR1_.. 
etc. is provided. Follow instructions in <<openct/README.rtf>> to install 
this library.
[NOTE: This library is not maintained anymore, and current development has 
been transfered to the OpenRecon fork at . However the results of 
few-view micro-CT paper were generated with the older 
OpenCT project, and hence OpenCT is used here. A copy of OpenCT repo is also 
archived here for sake of completeness.]

ii > ** utilitiesmatlab **
(This is a bunch of useful utility functions in MATLAB that I use across 
projects and is available for free download for others)
A copy of this library which has been tested to work with the codes RR1_.. 
etc. is provided. Follow instructions in <<utilitiesmatlab/README.rtf>> 
to install this library.

[NOTE: Contains a lot of functions for many purposes. Not all of the 
functions are required in reproducing the results of the paper. 
However the simple way to use this library is to copy the entire 
folder to your system, and add the folder path (with sub-directories) 
to the MATLAB path.]



Run the codes RR1_.., RR2_.. etc. to regenerate the results of the 



(1) You need to have a folder called ‘box’ in the same folder as 
the code. This folder is used to store all the config files 
generated while running the codes.

(2) Make sure the paths to external libraries are OK --- check 
in codes like following (here a relative path addressing has 
been used):

These may be changed to point to the direct folder path on 
your system.

(3) The data folders should point to the correct locations e.g. 
here a relative path addressing system is used:
DATAFOLDER = ‘data/mastic/‘;



Data used in this project was acquired on the Xradia MicroXCT-400 
scanner. Projection data is stored in the scanner in a native .TXRM 
format. The projection data was exported as binary (.BIN) images 
and used in the codes here. 

Each projection is a 2D image resulting from the cone-beam geometry 
of the micro-CT scanner. From these 2D projections, the mid-line 
is extracted which corresponds to fan-beam geometry in the central 
slice of the cone-beam. Mid-lines across all projection angles are 
collected to form a sinogram. This procedure of extracting the 
sinogram from 2D projections is quite time-consuming and is carried 
out only once by using the following function:
 The resulting sinogram is stored as a MATLAB .MAT file and stored 
 in same folder as the 2D projections. In subsequent runs of the 
 code, loadSinogram() senses that a .MAT file containing the sinogram 
 already exists, and directly loads the sinogram.

The entire tomography consisting of 2D projections at all projection 
angles requires a lot of disk space. To enable faster downloading, 
only the .MAT file containing the sinograms are supplied in this 
folder. However if you are interested in downloading the complete 
data, please download and extract the 
<<>> file.  In case of any 
problem getting the complete data, write to, or