An Experimental Approach in Colour Correction and Contrast Improvement for Underwater Images
M.Benedict Tephila, B. Maruthi Shankar, S. Abhinandhan, K.K. Arjun, P.M. Aswini, C.Divya Priya · 2022 Second International Conference on Artificial Intelligence and Smart Energy (ICAIS) · 2022
Underwater captured images often face low visibility and low colour cast because of the light absorbed and scattered when traveling in water. In the marine environment, it is difficult to view an image clearly and with proper details as compared to other environments. This is primarily due to the factors such as pollution, light, depth etc. It becomes an arduous task to produce a clear image in such situations. However, with the help of advanced technology, it has become easier to solve such issues using the appropriate algorithm or methodology. In this paper, a Bi-interval contrast enhancement and color correction is recommended to improve the image quality underwater. The technology or the process adopted here is such that it makes sure that the quality of the image is being restored and an enhanced image is produced as a result. Here the first step involves subinterval linear transformation with effective and simple color correction methodology to address color distortion. Color correcting strategy ensures the quality of the image restoration and hence a single image is produced by information from multiple images. This process is followed by the application of a Gaussian low-pass filter in the L-Channel in order to segregate the high and low frequency components. The last step of the process involves use of Bi-interval histogram using S-shaped function and optimal equalization threshold strategy for high and low frequency components. A comparison of the proposed work with other traditional methodologies indicates that this work shows high-quality underwater images with quantitative and qualitative evaluation.