Robust Object Detection of Underwater Robot based on Domain Generalization

2025-03-18

AI Summary

To address poor cross-domain generalization and low robustness in underwater object detection—caused by highly variable water quality, illumination distortions, biological camouflage, severe occlusion, and motion blur—this paper proposes the first unsupervised domain generalization detection framework tailored for underwater scenarios. Our method innovatively introduces geometric-photometric co-invariance modeling and dynamic feature disentanglement, integrated with self-supervised contrastive feature alignment, multi-scale underwater image degradation modeling, and a lightweight dynamic attention detection head. Crucially, it operates without annotations from target domains. Evaluated on four benchmarks—including UIB-UOD—our approach achieves a 12.7% mAP improvement over prior methods and surpasses state-of-the-art cross-domain transfer performance by 9.3%. Additionally, it attains real-time inference speed of 23 FPS on Jetson AGX.

Technology Category

• Computer Vision: Object Detection & Categorization
• Machine Learning: Unsupervised & Self-Supervised Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Object detection aims to obtain the location and the category of specific objects in a given image, which includes two tasks: classification and location. In recent years, researchers tend to apply object detection to underwater robots equipped with vision systems to complete tasks including seafood fishing, fish farming, biodiversity monitoring and so on. However, the diversity and complexity of underwater environments bring new challenges to object detection. First, aquatic organisms tend to live together, which leads to severe occlusion. Second, theaquatic organisms are good at hiding themselves, which have a similar color to the background. Third, the various water quality and changeable and extreme lighting conditions lead to the distorted, low contrast, blue or green images obtained by the underwater camera, resulting in domain shift. And the deep model is generally vulnerable to facing domain shift. Fourth, the movement of the underwater robot leads to the blur of the captured image and makes the water muddy, which results in low visibility of the water. This paper investigates the problems brought by the underwater environment mentioned above, and aims to design a high-performance and robust underwater object detector.

Problem

• Address severe occlusion in underwater object detection
• Overcome color similarity between organisms and background
• Mitigate domain shift from water quality and lighting

Innovation

• Domain generalization for underwater object detection
• Robust detection in occluded and camouflaged environments
• Adaptation to variable lighting and water conditions