Paper Title
Improved Gradient Profile Sharpness Transformation Based Super-Resolution Using De-Hazing

Abstract
This paper presents a Super Resolution (SR) technique considering atmospheric scattering models of hazy images. In which a SR image is considered as combination of radiance of the scene and atmospheric light. This assumption helps us to remove the hazy part in any SR image using dark channel prior. An edge sharpness metric called Gradient Profile Sharpness (GPS) is used in transforming the Low Resolution (LR) image to High Resolution (HR). This results in improved image quality. Extensive simulation results show that proposed method improves the performance of existing SR techniques, in terms of Peak Signal to Noise Ratio (PSNR) and Structural Similarity Index Measure (SSIM). Index Terms - Super Resolution from single image, Gradient Profile Sharpness transformation, Haze removal, Dark channel prior.