Fusion based Implementation of Underwater Image Enhancement Technique
Sakshi Mehra, Manjeet Singh, Deepti Kakkar · 2023
Abstract- Light gets dispersed and absorbed when it passes through water, which causes underwater images to typically have low intensity and blurriness. Large, suspended particles, alongside foggy or water that is turbid and has an enormous diversity of particles, etc., are what produce scattering. Objects appear more bluish green over red because longer wavelengths are dispersed by water as depth increases. Red is absorbed by the water because it has a longer wavelength. Images are significantly damaged because of these phenomena, and as a result, they have poor contrast, undesired coloring, and poor visibility. Applications for underwater images can be discovered in a variety of sectors, including mineral identification, aquatic ecosystem analysis and evaluation, marine exploration, and much more. So, it is essential to improve underwater images to boost their quality so that they can be used for a wide range of projects while still retaining the important data they carry. In this paper, we have utilized the fusion-based mechanism to enhance the underwater images so that they can be used for different applications as required. The method we have used here is straight forward and simple to use. In our work, 5 sample images from the UIEB dataset have been taken, and the raw images then are subjected to color correction, image sharpening, and contrast enhancement. Fusion is performed in the last step to provide the improved underwater image. Experimental findings have been performed on raw and reference photos to compare their MSE and PSNR values in order to see the impact of the proposed methodology in underwater image enhancement.