Digital Image Processing using Soft Computing Techniques and Spline Representations

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.