Title :
Image Recovery by Recursive Spline Smoothing and Application to Recognition of Compound Structures
Author :
Grebennikov, Alexandre
Author_Institution :
Fac. of Phys.-Math. Sci., Merited Autonomous Univ. of Puebla, Puebla, Mexico
Abstract :
In this paper we propose approach for special data post-processing, based on Recursive Spline Smoothing method, adapted to image reconstruction. The description of the special MATLAB program package is presented. Constructed algorithms and programs are applied to compound structures reconstruction. Necessity of reconstruction of compound structures appears in different applications. As a rule, tomography images are used for this purpose. However, obtained result usually contains significant visible noise. To eliminate such noise specific approach is applied for structures, which have piecewise constant characteristics with known values. It consists in Recursive Spline Smoothing and includes also projection of the pre-reconstructed data to the known set of values with respect to the absolute or relative criterion.
Keywords :
computerised tomography; image recognition; image reconstruction; mathematics computing; piecewise constant techniques; splines (mathematics); MATLAB program package; compound structure recognition; compound structure reconstruction; data post-processing; image reconstruction; image recovery; piecewise constant characteristics; recursive spline smoothing method; tomography images; Approximation methods; Compounds; Image reconstruction; Mathematical model; Smoothing methods; Spline; Tomography; compounds; computed tomography; images; reconstruction algorithms;
Conference_Titel :
Information Technology: New Generations (ITNG), 2012 Ninth International Conference on
Conference_Location :
Las Vegas, NV
Print_ISBN :
978-1-4673-0798-7
DOI :
10.1109/ITNG.2012.12