Joint Enhancement and Restoration of Underwater Images Using Reflectance and Illumination Components

Abstract

Underwater images are usually suffering from visibility issues like low brightness and low contrast. This is mainly resulted from attenuation of light in the imaging medium by absorption and scattering. In this paper, a joint restoration and enhancement algorithm for improving the quality of underwater images is proposed. Here degraded underwater image is decomposed to reflectance and illumination components. Red channel prior based restoration is applied to the illumination component. The reflectance component is enhanced by Contrast Limited Adaptive Histogram Equalization (CLAHE) and color correction. The contrast and color fidelity improved underwater images is computed by multi-scale wavelet based image fusion of restored and enhanced images. Comparative analysis of the proposed technique is done with the existing techniques both quantitatively and qualitatively. A non-reference metric (underwater image quality measure (UIQM)) is used for evaluating the performance of the proposed method quantitatively. The experimental results reveals that the proposed algorithm effectively enhances the quality of underwater images.