Underwater Trash Detection Dataset

Overview

The Underwater Trash Detection Dataset is a custom-annotated dataset designed to address the challenges of underwater trash detection caused by varying environmental features. Publicly available datasets alone are insufficient for training deep learning models due to domain-specific variations in underwater conditions. This dataset offers a cumulative, self-annotated collection of underwater images for detecting and classifying trash, providing a strong foundation for deep learning research and benchmark testing.

Dataset Summary

• Total Images: 9,576
• Annotation Types: Trash classification (plastic, trash, underwater debris) vs. environmental factors (fish, flora, fauna).

Dataset Split

| Split | Percentage | Number of Images | |-------------|----------------|-----------------------| | Train Set | 76% | 7,308 | | Validation | 19% | 1,795 | | Test Set | 5% | 473 |

Preprocessing

• Image Resize: All images are resized to 256x256 pixels using stretching for uniform input dimensions.

Purpose

This dataset supports research in underwater trash detection while addressing storage and computational constraints in underwater mobile devices. It enables the development of optimized algorithms for efficient trash detection and classification using minimal resources.