Improving Visibility of images - Underwater by Modeling of Back Light and Optimization of the Transmission Map
M Abinithi, K. Priyadharsini, Dinesh Kumar J R, K T Deepadarsini, V T Haripriya, S R Indhuja · 2022 3rd International Conference on Electronics and Sustainable Communication Systems (ICESC) · 2022
Light absorption in the water medium causes degradation and distortion of under-water images. This study uses the technique is image restoration. It restores the images captured and adjusts the color gradient level of the images of underwater. Underwater photographs are not decorated and do not hide naturally. Therefore, to avoid this, various image restoration methods are used to retrieve the original image. This results suggest that the proposed approach has significantly enhanced the visibility and reliability of images in underwater. Two parameters are considered for the image enhancement that is BckL (Background Light) and TrMap (Transmission Map). Color and Contrast adjustment can be developed for the enhancement of the underwater image. The proposed method is developed by enhancing the color and contrast of the image by using the annotation of image manually. A database with the annotated image (ABckL) compared with the histogram model of underwater image statistical analysis is performed. TrMap is estimated roughly for the R channel with the new dark underwater (NDUI) images via a statistical model. Light attention (LA) and underwater reverse saturation (URS) are applied to NDUI to modify the course of the R channel to improve the image enhancement. The difference between red Channel (R) and channel- G-B estimates the TrMap of G-B channel. The white balance method at the end stages improves the color and contrast of the image. Compared with other methods, the proposed method produced has achieved high accuracy, short computation time, color improvement.