Underwater Color Correction

Norzamri, Siti Nor Iffah (2019) Underwater Color Correction. Project Report. UTeM, Melaka, Malaysia. (Submitted)

Text (24 Pages) Underwater Color Correction.pdf - Submitted Version Download (332kB)

Abstract

This report presents the color correction for underwater images. Recovering correct or at least realistic colors of underwater scenes is a very challenging issue for imaging techniques. The crucial reason for this problem is due to the density of the seawater and the penetrating of light. Since the light will be absorbed and scattered by small particles when travels in water, underwater imaging exists in difficulties of color cast. It gives impact to the limitation of visibility in the sea water. The need to correct colors of underwater images is an important task required in all image-based applications like underwater research such as coral bleaching and navigation as it is vital for the researchers to monitor and tracking the corals’s health and pathway for the underwater transportation. To restore the color of the images, the method that is used in this project is Unsupervised Color Correction method and it was technically simulated in MATLAB by using 36 numbers of input images that categorised to; bluish, greenish and normal images respectively. In UCM method, it mainly has two stages to solve the problems mentioned above. First, equalization of RGB colors to achieve equal color values of the RGB component. Second, contrast to RGB color models to improve the contrast of the images. The result is evaluated visually and quantitatively by using Sobel edge detector to detect the edge and proved that whether the color of the images is improved by using Unsupervised Color Correction method. From the results obtained, it shown that the input images that is bluish and greenish improved significantly rather than the normal images that having a deterioration in terms of visual and the number of edges.

Actions (login required)

View Item