Underwater Image Enhancement via RGB Color-Cast Correction and Lab Space Luminance-Chrominance Enhancement

Underwater images often suffer from color casts, insufficient contrast, and blurred details due to wavelength-dependent absorption and scattering, which limits their use in marine observation and underwater target recognition. Therefore, it remains challenging to jointly improve color fidelity, contrast, and detail visibility under varying water types and illumination. To address these issues, this paper proposes an underwater image enhancement method that integrates RGB Space color-cast correction with Lab Space luminance-chrominance synergistic enhancement. First, channel compensation is performed in the RGB space according to color-cast types, together with a Gray-World white-balance strategy, to suppress severe color bias and recover more natural tones. Subsequently, synergistic enhancement is conducted in the Lab space as follows: an improved Auto-MSR is applied on the L channel for contrast restoration while suppressing noise amplification, and an edge-aware detail enhancement module is incorporated to reinforce structural textures. Finally, chrominance balancing is applied to A and B channels to further improve color consistency and rendition. Experiments on the UIEB and UIQS datasets demonstrate that the proposed method achieves the best performance among the methods considered on AG, EI, and CCF. Specifically, it attains AG = 10.716, EI = 107.197, and CCF = 46.935 on UIEB, as well as AG = 10.136, EI = 103.780, and CCF = 39.433 on UIQS. These results indicate clear advantages in image clarity, edge detail preservation, and color rendition. Future work will focus on extending the proposed method to real-time underwater video enhancement.