Underwater Image Enhancement Based on Color Balance and Edge Sharpening

Underwater optical images often suffer from color cast, edge-blurring and low contrast due to the medium absorption and scattering in water. To solve these problems, we propose an effective technique to improve underwater image quality. First, we introduce an effective color balance strategy based on affine transform to address the color distortion. Then we convert the underwater image from RGB color space to CIE-Lab color space for contrast improvement. In ‘L’ component’s nonsubsampled contourlet transform (NSCT) domain, global contrast adjustment and multi-scale edge sharpening are conducted respectively for lowpass and bandpass direction subbands. Finally, a color-corrected and contrast-enhanced output image can be generated by inverse NSCT and conversion back to RGB color space. The propose method is a single image approach that does not require prior knowledge about the underwater imaging conditions. Experimental results show that our method outperforms state-of-the-art methods both in qualitative and quantitative evaluation. It generally results in good perceptual quality, with significant enhancement of the global contrast, the color, and the image structure details.