Wearable System for Human Underwater Visual Enhancement

This article presents a wearable visual enhancement system for divers in complex underwater scenarios. An adaptive underwater image enhancement algorithm based on multi-scale fusion is also proposed and implemented in the underwater wearable system to improve the wearer's visual perception of their surroundings. An illumination compensation processing step is introduced to exhibit significantly clearer image features and effectively preserve the true and essential information of the image throughout the fusion process. We conducted extensive experiments across diverse datasets and various underwater scenarios to validate the efficacy of our algorithm. The results show that our approach surpasses state-of-the-art techniques, including traditional, deep learning-based, and multi-scale fusion-based image enhancement methods, as evidenced by both objective quantitative metrics and qualitative evaluations. Also, an underwater experiment is also conducted in the swimming pool and the result shows that the enhanced image's underwater color image quality evaluation value registers a 12% increment relative to the original image, and its underwater image quality measures value experiences a substantial surge of 173%.