Robust Image Processing Techniques for Construction Environment Monitoring Using Underwater Robots

Abstract

This study proposes a robust image processing framework for underwater robot-based construction environment monitoring, targeting complex degradations observed in actual marine environments. Unlike conventional approaches that mainly consider absorption and backscattering, real underwater imagery is strongly affected by depth-dependent forward scattering blurring and particle-induced degradations such as marine snow. To address this issue, we introduce a staged processing pipeline that sequentially models background degradation via depth-aware forward scattering and foreground degradation using realistic marine snow patterns extracted from actual images. The resulting synthetic data are used to retrain an existing Joint-ID network without modifying its architecture, enabling an isolated evaluation of dataset realism. In addition, a lightweight post-processing scheme is applied to enhance contrast and structural clarity. Experiments on actual underwater datasets collected in Korean coastal environments demonstrate consistent improvements in visual quality and UIQM (Underwater Image Quality Measure) scores. These findings support more reliable underwater robotic monitoring in real marine environments.