DRC: Chromatic aberration intensity priors for underwater image enhancement

Abstract

Underwater imaging technology is a crucial tool for monitoring marine flora and fauna. However, selective light absorption and scattering properties of water make underwater imagery frequently appear blurred and exhibit color biases, hindering the extraction of vital aquacultural insights. To address this challenge, we propose a method, namely DRC, which is a holistic approach to enhancing underwater image clarity and color fidelity. This method comprises three integral components: D-procedure, R-procedure, and C-procedure. The D-procedure intricately accounts for the trichromatic underwater attenuation dynamics, formulating a chromatic aberration intensity prior. This prior counters the disparities in degradation levels seen in conventional single-channel prior depth estimations, achieving a dynamic depth representation approaching binocular image precision. The R-procedure, utilizing an adaptive dark-pixel prior, pinpoints corresponding points across varied depth zones to counteract backscattering, thereby mitigating the image’s hazy appearance. The C-procedure bolsters image luminance and color fidelity through opponent channel rectification and amalgamates pronounced image contrasts and intricate details via Gaussian pyramid fusion. The method was tested on several publicly available datasets and compared with nine popular underwater image enhancement techniques. Both subjective and objective assessments underscore the superiority of our DRC method over existing underwater image enhancement techniques.