Dataset Card for OpenSARWake

OpenSARWake is a benchmark dataset built for ship wake detection. This collection provides 3,973 images containing two polarization modes and 4,096 instances. Most importantly, it encompasses SAR images in the L-, C-, and X-bands, which have not been provided by previous datasets. The images in the dataset have spatial resolutions of 1.25 m to 12.5 m. The image size is 1024× 1024 pixels.

This is a FiftyOne dataset with 2383 samples.

Installation

If you haven't already, install FiftyOne:

pip install -U fiftyone

Usage

import fiftyone as fo
from fiftyone.utils.huggingface import load_from_hub

# Load the dataset
# Note: other available arguments include 'max_samples', etc
dataset = load_from_hub("dgural/OpenSARWake")

# Launch the App
session = fo.launch_app(dataset)

Dataset Details

Dataset Sources

• Repository: https://github.com/libzzluo/OpenSARWake?tab=readme-ov-file

Dataset Structure

The dataset includes ground_truth field as well as clip embeddings for visualization

Source Data

Google Drive Link

Who are the source data producers?

Xu, Chengji and Wang, Xiaoqing

Citation [optional]

@ARTICLE{10507047, author={Xu, Chengji and Wang, Xiaoqing}, journal={IEEE Geoscience and Remote Sensing Letters}, title={OpenSARWake: A Large-Scale SAR Dataset for Ship Wake Recognition with a Feature Refinement Oriented Detector}, year={2024}, doi={10.1109/LGRS.2024.3392681}}