Abstract:
The work presented in this thesis is devoted to digital image processing techniques using
soft computing techniques and spline representations. In particular, it emphasizes on the
problems of image interpolation for finding the optimal digital images. In this thesis, four
image interpolation schemes are developed which utilize one of the soft computing
techniques, Genetic Algorithm (GA) and newly constructed rational Ball Cubic B-spline
representation, the cubic trigonometric B-spline representation, the quadratic trigonometric
B-spline representation along with the rational cubic B-spline representation to get the
resulting interpolated images. Genetic algorithms are used to get the optimal values of
shape parameters in the description of the proposed B-spline representations. The proposed
schemes are demonstrated with both gray scale and color digital images. Numerical results
are collected in terms of well-known image quality metrics including Peak Signal-to-Noise
Ratio (PSNR), Structure SIMilarity (SSIM), Multi-Scale Structure SIMilarity (MS-SSIM)
and Feature SIMilarity (FSIM) indices to analyze and compare the proposed interpolation
schemes with each other and existing ones.
