Investigating the quality measures of image enhancement by convoluting the coefficients of analytic functions

Abstract

The aim of this research is to enhance image quality by applying convolution methods to a newly generalized subclass of an analytic function, \(k-\Omega S^*(\rho ,\beta )\), which incorporates the concept of the Mittag-Leffer type Poisson distribution associated with starlike functions. Image enhancement processes, such as noise reduction, sharpening, and brightening, improve the image’s suitability for display or further analysis. The proposed method demonstrates superior results based on performance metrics including PSNR (Peak Signal-to-Noise Ratio), SSIM (Structural Similarity Index), MSQE (Mean Squared Error), RMSE (Root Mean Squared Error), PCC (Pearson Correlation Coefficient), and CIR (Contrast Improvement Ratio). For the flower dataset, the technique achieves values of 20.425 for PSNR, 0.8866 for SSIM, 765.044 for MSQE, 27.143 for RMSE, 0.1310 for PCC, and 0.9794 for CIR. Similarly, for the brain dataset, the quality metrics are 24.2981 for PSNR, 0.9773 for SSIM, 268.288 for MSQE, 16.0041 for RMSE, 0.9888 for PCC, and 0.2918 for CIR.