Adaptive Color Correction and Detail-Preserving Fusion for Enhanced Underwater Image Quality and Segmentation

Underwater images frequently experience significant color distortion, diminished contrast, and decreased visibility. These issues arise due to the selective absorption and scattering of light in aquatic environments. This study introduces an innovative method for enhancing the quality of underwater images through the implementation of adaptive color correction and detail-preserving fusion techniques. The approach addresses the attenuation of red and blue color channels, followed by an adaptive fusion process that enhances contrast and texture while maintaining intricate image details. The findings reveal notable advancements compared to current state-of-the-art techniques, as assessed by widely recognized metrics including the Perceptual Color Quality Index (PCQI), the Underwater Image Quality Measure (UIQM), and the Underwater Color Image Quality Evaluation (UCIQE). Specifically, the method achieves a PCQI score of 1.00 on test shipwreck images, demonstrating a significant improvement in color quality. The enhanced images exhibit more accurate color representation, improved sharpness, and higher contrast, making them suitable for both human observation and further computer vision applications. Additionally, the segmentation performance, as measured by the Geodesic Active Contours (GAC++) algorithm, is markedly improved on the enhanced images, demonstrating the practical utility of the proposed approach for underwater image analysis and object detection. This work sets a new standard for underwater image enhancement, facilitating its application in fields such as marine biology, underwater archaeology, and robotic vision systems.