The file is the introduction for the fine-grained remote sensing classificaiton datasets.
Fine-grained ship classification dataset\cite{fgsc23}, FGSC-23 for short, is a high-resolution optical fine-grained remote sensing image classification dataset, including 23 types of ships and 4052 samples. Each target is given a class label, an aspect ratio label, and a distribution direction label. Compared with the existing optical remote sensing image ship target recognition dataset, the FGSC-23 dataset has the characteristics of diverse image scenes, fine classification and complete labels.
- BaiduDisk Link, extraction code:
n8ra - Google Disk
@article{zhang2020new,
title={A new benchmark and an attribute-guided multilevel feature representation network for fine-grained ship classification in optical remote sensing images},
author={Zhang, Xiaohan and Lv, Yafei and Yao, Libo and Xiong, Wei and Fu, Chunlong},
journal={IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing},
volume={13},
pages={1271--1285},
year={2020},
publisher={IEEE}
}
Fine-Grained Ship Classification in Remote-sensing with 42 classes, FGSCR-42 for short, contains 42 subclasses of 10 broad classes, specifically, subclasses of ships such as Kitty-Hawk class aircraft carrier, Arleigh-Burke class destroyer, and mega-yacht. The images of this dataset are composed of sliced images of the object detection datasets DOTA, HRSC2016, NWPUVHR-10, etc. The slices are obtained by extending the data in the target frame label attached to these datasets to the surrounding by a certain pixel and then cutting them out from the original image. Compared with FGSC-23, the dataset category is further subdivided, specific to the type or class of the ship.
- BaiduDisk Link, extraction code:
w3r3 - Google Disk
@article{di2021public,
title={A public dataset for fine-grained ship classification in optical remote sensing images},
author={Di, Yanghua and Jiang, Zhiguo and Zhang, Haopeng},
journal={Remote Sensing},
volume={13},
number={4},
pages={747},
year={2021},
publisher={MDPI}
}
- BaiduDisk Link, extraction code:
1q3d - Google Disk
@inproceedings{yi2022mha,
title={MHA-CNN: Aircraft Fine-Grained Recognition of Remote Sensing Image Based on Multiple Hierarchies Attention},
author={Yi, Yonghao and You, Yanan and Zhou, Wenli and Meng, Gang},
booktitle={IGARSS 2022-2022 IEEE International Geoscience and Remote Sensing Symposium},
pages={3051--3054},
year={2022},
organization={IEEE}
}