UGER: Enhancing the Quality of ROUV Image Segmentation with CLAHE and Color Space-Morphology Masking Techniques

This research aims to enhance the quality of image segmentation captured by Remotely Operated Underwater Vehicles (ROUVs) using contrast-limited adaptive Histogram Equalization (CLAHE) and color space masking morphology techniques. We employ three primary metrics, namely Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index (SSIM), and Video Multi-method Assessment Fusion (VMAF), to evaluate the effectiveness of the proposed methods. The analysis results demonstrate that applying CLAHE and color space-morphology masking significantly improves image segmentation quality. Data obtained from various environments, such as swimming pools, rivers, and fish ponds, show consistent improvements in image reconstruction, structural similarity, and overall video quality. This study makes a significant contribution to underwater image processing, enhancing the applications of underwater monitoring and exploration with more accurate and high-quality image segmentation.