Underwater vision is critical for applications such as marine engineering, aquatic robotics, and environmental monitoring. However, severe image degradation–caused by light absorption, scattering, and backscattering–often hampers visual recognition tasks. While underwater image enhancement (UIE) is intuitively expected to improve recognition by restoring visual quality for human perception, the extent of its actual impact on automated computer vision, particularly segmentation, remains underexplored. This study systematically evaluates a range of UIE algorithms, spanning both traditional methods and state-of-the-art (SOTA) deep learning approaches, to assess their effects on underwater instance segmentation. We applied these enhancement techniques to the UIEB benchmark dataset and conducted comprehensive qualitative and quantitative analyses of the enhanced images. Subsequently, a widely used segmentation model, Fishial.AI, was employed to evaluate segmentation performance on both raw and enhanced image sets. Experimental results reveal that existing UIE methods yield promising improvements in segmentation accuracy. These findings provide valuable insights into the role of image enhancement in underwater vision and highlight important considerations for the design of future enhancement algorithms.