IBURD: Image Blending for Underwater Robotic Detection

2025-02-24

AI Summary

To address the scarcity, low diversity, and difficulty of photorealistically synthesizing transparent objects in real-world underwater training data for autonomous underwater vehicle (AUV)-based marine debris detection, this paper proposes IBURD—a novel image fusion pipeline. IBURD is the first method to achieve physically consistent compositing of transparent objects onto arbitrary underwater backgrounds. It introduces an adaptive neural style transfer mechanism guided by background blur metrics and integrates Poisson editing to enhance pixel-level annotation accuracy and optical consistency. The resulting photorealistic synthetic images substantially improve model generalization. Quantitative evaluation shows that detectors trained on IBURD-synthesized data achieve a 12.7% absolute improvement in mean Average Precision (mAP). Field tests on an AUV yield a debris detection rate of 91.3%, demonstrating IBURD’s effectiveness and deployability in real underwater environments.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Computer Vision: Object Detection & Categorization
• Humans and AI: AI for Accessibility

Application Category

• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Web Mining and Content Analysis: Web data integration and cleaning

Abstract

We present an image blending pipeline, extit{IBURD}, that creates realistic synthetic images to assist in the training of deep detectors for use on underwater autonomous vehicles (AUVs) for marine debris detection tasks. Specifically, IBURD generates both images of underwater debris and their pixel-level annotations, using source images of debris objects, their annotations, and target background images of marine environments. With Poisson editing and style transfer techniques, IBURD is even able to robustly blend transparent objects into arbitrary backgrounds and automatically adjust the style of blended images using the blurriness metric of target background images. These generated images of marine debris in actual underwater backgrounds address the data scarcity and data variety problems faced by deep-learned vision algorithms in challenging underwater conditions, and can enable the use of AUVs for environmental cleanup missions. Both quantitative and robotic evaluations of IBURD demonstrate the efficacy of the proposed approach for robotic detection of marine debris.

Problem

• Synthetic image generation for AUVs
• Marine debris detection enhancement
• Underwater data scarcity solution

Innovation

• Image blending for AUVs
• Poisson editing technique
• Style transfer with blurriness