Underwater Image Enhancement Using MIRNet

In recent years, enhancement of underwater images is a challenging task, which gaining priority since the human eye cannot perceive under water. The significant details are not clearly captured using conventional image acquisition techniques, and also they expensive. Hence, quality processing algorithms can be enhanced in absence costly reliable techniques. Traditional have certain limitations case these with varying degrees fuzziness color deviation. proposed model, authors used deep learning model for enhancement. First, original pre-processed by white balance algorithm colour correction contrast improved technique. Next, given to MIRNet framework that enhance low-light level images. measured peak signal-to-noise ratio (PSNR), root mean square error (RMSE), structural similarity index (SSIM) parameters.