Two-step domain adaptation for underwater image enhancement. (February 2022)

• Record Type: Journal Article
• Authors: Jiang, Qun; Zhang, Yunfeng; Bao, Fangxun; Zhao, Xiuyang; Zhang, Caiming; Liu, Peide
• Is Part Of: Pattern recognition. Volume 122(2022)
• Publication Date: 2022-02
• Subjects: Underwater image enhancement -- Transfer learning -- Domain adaptation -- Cycle-consistent adversarial network; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2021.108324
• Languages: English
• ISSNs: 0031-3203

Abstract

Highlights: Inspired by transfer learning, we migrate in-air image dehazing to underwater image enhancement. We propose a novel two-step domain adaptation framework for underwater image enhancement, which realizes cross-domain adaptation from the air domain to the underwater domain. Our method is trained on real-world underwater images without utilizing underwater images synthesized with in-air images, which eliminates the dependence on underwater paired data. Abstract: In recent years, underwater image enhancement methods based on deep learning have achieved remarkable results. Since the images obtained in complex underwater scenarios lack a ground truth, these algorithms mainly train models on underwater images synthesized from in-air images. Synthesized underwater images are different from real-world underwater images; this difference leads to the limited generalizability of the training model when enhancing real-world underwater images. In this work, we present an underwater image enhancement method that does not require training on synthetic underwater images and eliminates the dependence on underwater ground-truth images. Specifically, a novel domain adaptation framework for real-world underwater image enhancement inspired by transfer learning is presented; it transfers in-air image dehazing to real-world underwater image enhancement. The experimental results on different real-world underwater scenes indicate that the proposed method produces visually satisfactory results.