Underwater Image Enhancement Based on Fusion Technique via Color Correction and Illumination Adjustment
Shahjadi Silma Subah, Md. Ashfaqul Islam, MD. Mohyminul Islam · 2019 1st International Conference on Advances in Science, Engineering and Robotics Technology (ICASERT) · 2019
Underwater image is affected by absorption and scattering when light propagates through water; resulting in poor visibility, reduced contrast, color attenuation. So before going into any restoration processing activity of an image, it is mandatory to enhance the image first. In this paper, a fusion based technique is introduced, where the method is to use wavelet decomposition. The proposed method has two inputs enhanced from the original underwater image. One is color corrected image and other is illumination adjusted image. Illumination adjustment is done by adaptive histogram equalization, gamma correction and filtering in the proposed system. Histogram equalization and adaptive histogram equalization techniques are used for color correction on the adjusted image. These modified images are then fused based on their maximum coefficient values and the resulted output shows promising result improving the haze of the underwater image. This is a very simple method and the resulted image is clearer than the hazy image.