Multi Sensor Underwater Image Fusion Using Modified Filter Bank Reconstruction Model

Abstract

AbstractClear and clean underwater images can provide valuable information, which are crucial for developing, exploring, and protecting the underwater resources. However, the raw underwater image seldom fulfills the requirements concerning the underwater research due to the serious image degradation. To enhance underwater image, we propose an effective and new method combined on the color correction with filter bank technique. Initially the white balancing is applied to remove the color cast while DCP: is used to remove the haze and followed by bilateral filtering is applied to enhance underwater image. Finally introduce a new haar wavelet transform, called Tetrolet Transform, to get an efficient image representation. The appropriate filter bank algorithm is simple but highly effective. Numerical results show the tetrolet transform’s strong efficiency in image fusion. This strategy gives the dark regions a better exposure, improves contrast, and the edges are significantly preserved and enhanced. This algorithm effectively enhances the underwater image that our image experimental results clearly demonstrate. The fusion of information provides the basis for planning, decision-making, and control of autonomous and intelligent machines. The composite image makes it easier for the user to detect, recognize, and identify targets and increase situational awareness. Finally in this work, for various admitible covering, the performance values are calculated. Likewise PSNR and standard deviation values are compared with the conventional model.KeywordsUnder water imagesDehazing algorithmDark Channel Prior Model (DCP)Tetrolet transform