Satellite Image Dehazing Using Fast Iterative Domain Gaussian Guided Image Filtering

Abstract

In general, satellite images are foggy due to factors such as noise, snow, thin cloud, dust, and so on, resulting in image contrast reduction. Dehazing is described as the elimination of noise or ambient contaminants from a image in order to improve image quality. However, most state-of-the-art technologies failed to completely eliminate atmospheric influences and noise from satellite images. To address this issue, this study focuses on the creation of a grey world optimization method for the correct estimate of satellite atmospheric light. The study also creates a unique technique for estimating and refining dark channel prior-based transmission maps at the pixel and patch levels. As a consequence, every pixel-based patch contains information about how satellite atmospheric effects were addressed. The fast iterative domain gaussian guided image filtering (FID-GGIF) method was created to provide output that is smooth with dehazing properties. According to the simulation results, the suggested study beats cutting-edge approaches in terms of both quantitative and qualitative results