Tuning Shape Parameter of Radial Basis Functions in Zooming Images using Genetic Algorithm

Image zooming is one of the current issues of image processing that maintains the quality and structure of an image. In order to zoom an image, it is necessary to place the extra pixels in the image data. Adding the data to the image must be consistent with the texture in the image in order to prevent artificial blocks. In this work, the required pixels are estimated using the radial basis functions and calculating the shape parameter c with the genetic algorithm. Then all the estimated pixels are revised on the basis of the sub-algorithm of edge correction. The proposed method is a non-linear one that preserves the edges and minimizes the blur and block artifacts on the zoomed image. This method is evaluated on several images for calculating the suitable shape parameter of the radial basis functions. The numerical results obtained are presented using the PSNR and SSIM fidelity measures on different images as compared to some other methods. The average PSNR of the original image and image zooming was found to be 33.16, which shows that image zooming by factor of 2 is similar to the original image, which emphasizes that the proposed method has an efficient performance.