Adaptive Color Correction and Multi-Scale Fusion for Underwater Image Enhancement
Ziyu Jia, Yifeng Gu, Xinrui He, Yuhang Zhou · 2025
Underwater images often suffer from severe quality degradation, such as color distortion, low contrast, and blurred details, primarily due to light absorption and scattering in the aquatic medium. These issues significantly impede underwater exploration and analysis. This paper proposes an effective underwater image enhancement method that combines adaptive color correction with multi-scale fusion. Initially, an adaptive color correction technique is employed to address color casts by redefining color channels and applying compensation, followed by a guided fusion strategy with the maximum attenuation map. Subsequently, to tackle low contrast and detail obscurity, the method generates two enhanced versions of the color-corrected image: one focusing on global contrast enhancement using information entropy maximization, and another emphasizing detail enhancement through a combination of histogram equalization and unsharp masking. Finally, these two versions are blended using a multi-scale pyramid fusion algorithm, guided by normalized weight maps derived from luminance and saliency features, to produce a high-quality enhanced image. Experimental results on three public underwater image datasets demonstrate that the proposed method achieves significant improvements in both subjective visual quality and objective evaluation metrics compared to several state-of-the-art techniques.