Physics-Inspired Synthesized Underwater Image Dataset

2024-04-05

arXiv.org

AI Summary

Existing underwater image enhancement methods heavily rely on paired atmospheric–underwater image data, which is scarce and difficult to acquire. To address this, we propose PHISWID—the first optical physics–driven synthetic dataset for underwater image enhancement. PHISWID leverages RGB-D atmospheric images and integrates the Jaffe–McGlamery underwater light propagation model to physically simulate chromatic attenuation and snow scattering, generating large-scale, interpretable, and reproducible paired ground-truth/degraded images. The dataset is publicly available, high-fidelity, and provides pixel-level ground truth, thereby filling a critical gap in physics-based synthetic data. Extensive evaluation on multiple state-of-the-art enhancement models demonstrates average improvements of 2.1 dB in PSNR and 0.032 in SSIM. PHISWID has been open-sourced and adopted by several research groups, significantly advancing objective benchmarking and algorithm development in underwater vision.

Technology Category

• Computer Vision: Computational Photography, Image & Video Synthesis
• Humans and AI: Crowd Sourcing and Human Computation
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Search and Retrieval-Augmented AI: Web crawling and indexing
• Web Mining and Content Analysis: Web data generation and simulation
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

This paper introduces the physics-inspired synthesized underwater image dataset (PHISWID), a dataset tailored for enhancing underwater image processing through physics-inspired image synthesis. For underwater image enhancement, data-driven approaches (e.g., deep neural networks) typically demand extensive datasets, yet acquiring paired clean and degraded underwater images poses significant challenges. Existing datasets have limited contributions to image enhancement due to lack of physics models, publicity, and ground-truth images. PHISWID addresses these issues by offering a set of paired ground-truth (atmospheric) and underwater images synthetically degraded by color degradation and marine snow artifacts. Generating underwater images from atmospheric RGB-D images based on physical models provides pairs of real-world ground-truth and degraded images. Our synthetic approach generates a large quantity of the pairs, enabling effective training of deep neural networks and objective image quality assessment. Through benchmark experiment with some datasets and image enhance methods, we validate that our dataset can improve the image enhancement performance. Our dataset, which is publicly available, contributes to the development in underwater image processing.

Problem

• Addresses lack of paired clean-degraded underwater image datasets
• Synthesizes underwater images using physics-based degradation models
• Enables effective training for underwater image enhancement networks

Innovation

• Physics-based underwater image observation model
• Synthetic degradation with color and artifacts
• Public paired atmospheric-underwater image dataset