Performance Evaluation of Underwater Image Enhancement Techniques

Absorption, scattering, and colour distortion make underwater photography difficult. Marine biology, underwater archaeology, and surveillance need better underwater photos. This research compares cutting-edge underwater image-enhancing methods to assess their efficacy and usefulness. This paper provides a comparative analysis of the various filtering approaches for underwater image enhancement, such as median filtering, average filtering, Gaussian filtering, Wiener filtering, and wavelet-based soft thresholding. The performance of these techniques is evaluated based on the mean square error, peak signal-to-noise ratio, average gradient, PCQI, and UICQI. The effectiveness of the suggested filtering scheme is validated on some sample images of the public UIEB dataset. It provides the analytical finding, advantages, disadvantages and performance of the different techniques for wide variety of underwater images enhancement. The paper also explores underwater image enhancement issues and future approaches. The promising research directions are integrating domain-specific information, real-time processing, and adaptive algorithms for dynamic underwater scenes. The discoveries help increase underwater photography visibility and interpretability, increasing underwater exploration and monitoring systems.