Underwater Image Enhancement based on Histogram Manipulation and Multiscale Fusion
Sangeetha Mohan, Philomina Simon
Abstract
The underwater images are a good source of information which explores idea about sea creatures, to study about seafloor hydrother-mal vents. Low contrast, color distortion and poor visual appearance are the major issues that an underwater image has to undergo. Such problems were caused by dispersion and refraction of light as they penetrate from rarer to denser media. The scattering of light reduces color contrast. The influence of water in underwater images is not only due to scattering but also due to the presence of underwater organisms. Here we introduce an improved method for underwater image enhancement based on the fusion method that is capable to restore accurately underwater images. The proposed work takes a single image as the input and a sequence of operations such as white balancing, gamma correction, sharpening, manipulating weight maps are performed on the input image. Finally multiscale image fusion of the inputs is done to obtain the resultant output. In the initial stage, color distorted input image is white balanced to remove the color casts maintaining a realistic subsea image. In the second stage, CLAHE is performed on the gamma corrected image. CLAHE plays a significant role in luminance enhancement of underwater images. At the same time, histogram equalization is performed on the sharpened image. The weight maps analyze image characteristics that properly specify the spatial pixel relationship. Finally in the last stage, multiscale pyramidal fusion of the inputs and weight maps are performed. The fusion performed here is the Pyramidal fusion. Result analysis depict the improvement of the underwater images using proposed method.