A novel elementary spatial expanding scheme form on SISR method with modifying geman & mcclure function

Darun Kesrarat, Kornkamol Thakulsukanant, Vorapoj Patanavijit
2019 TELKOMNIKA (Telecommunication Computing Electronics and Control)  
Because of the feasible and impressive fallout, the classical Super-Resolution Reconstruction (SRR) is the contemporary algorithm for improving spatial information and reducing noise and SISR (Single-Image Super-Resolution) method, which is form on the classical SRR, is solely developed for improving spatial information. Disastrously, deficiency of the classical SISR method is conceptually computed from three specifications (b, h, k) and the simulating calculation of the optimized
more » ... mized specifications for interpolating the better and higher spatial information images with highest PSNR is so burdersome. For figuring out this issue, the Geman&McClure function is proposed to replace with the ordinary SISR function because this function is conceptually computed from only one specification (T), contrary to three specifications similar to classical SISR method hence this analytic article focuses to offer a novel elementary spatial expanding scheme form on SISR method with modifying Geman&McClure function. Therefore, the fallout of a proposed spatial expanding scheme approximately matches to classical SISR method. From these reason, a novel elementary spatial expanding scheme is easily implemented for real works. Comprementary Investigation Because of great requirement of contemporary computer vision and image processing [1-5] for more than one and half decades, images with better and higher spatial information is dramatically required however almost recorded images is usually lower spatial information, which are recorded by marketing build-in CCTV or camera. Therefore, the regular digital image interpolation is conceptual idea in order to creating an image with higher spatial information from one image with lower spatial information. The dominant aspiration of spatial expanding scheme form on SISR method, which is developed from the worldwide famous Super-Resolution Reconstruction (SRR) [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] , is for solely improving spatial information. Conceptually, both SRR and SISR are an ill-posed inversed issue [10, 16] from the reason that copious higher spatial information images can be constituted from only one recorded image with lower spatial information under the identical recording process. In the SISR inquiring in the last two and half decades, one of the most feasible and impressive spatial expanding scheme is the SISR method using boundary constructing techniques, which is originally invented by Bo-Won Jeon [18] in 2006. This method can constitute better and higher spatial information images and, moreover, has low complexity. This SISR method is incorporated by two dominant sub-systems: the boundary constructing sub-system and the regularized function. In the first sub-system, the boundary constructing sub-system is conceptually computed from Laplacian pyramid method [19] [20] [21] [22] [23] for constructing boundary and line of the higher spatial information image. In the second sub-system, the regularized function C (x, y) [18] is conceptually computed from three specifications (b, h, k) and the simulating calculation of the optimized specifications [24, 25] for interpolating the better and higher spatial information images with highest PSNR. For figuring out this issue, the Geman&McClure function [26] , which has been mathematically guarantee and investigated for more than thirty years thereby the Geman&McClure function has an exhaustive inquiring, is conceptually computed from only one specification (T), contrary to three specifications similar to classical SISR method. Therefore, this analytic article focuses to
doi:10.12928/telkomnika.v17i5.12799 fatcat:dmntkqlv7zgjrmchfgy7mqpiyy