- Deskripsi :
Dataset EuroSAT didasarkan pada citra satelit Sentinel-2 yang mencakup 13 pita spektral dan terdiri dari 10 kelas dengan 27.000 sampel berlabel dan geo-referensi.
Dua set data ditawarkan: - rgb: Hanya berisi pita frekuensi optik R, G, B yang dikodekan sebagai gambar JPEG. - all: Berisi semua 13 band dalam kisaran nilai asli (float32).
URL: https://github.com/phelber/eurosat
Situs web : https://github.com/phelber/eurosat
Kode sumber :
tfds.image_classification.Eurosat
Versi :
-
2.0.0
(default): Tidak ada catatan rilis.
-
Ukuran kumpulan data :
Unknown size
Cache otomatis ( dokumentasi ): Tidak diketahui
Split :
Membagi | Contoh |
---|---|
'train' | 27.000 |
- Kutipan :
@misc{helber2017eurosat,
title={EuroSAT: A Novel Dataset and Deep Learning Benchmark for Land Use and Land Cover Classification},
author={Patrick Helber and Benjamin Bischke and Andreas Dengel and Damian Borth},
year={2017},
eprint={1709.00029},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
euroat / rgb (konfigurasi default)
Deskripsi konfigurasi : Saluran RGB Sentinel-2
Ukuran unduhan :
89.91 MiB
Fitur :
FeaturesDict({
'filename': Text(shape=(), dtype=tf.string),
'image': Image(shape=(64, 64, 3), dtype=tf.uint8),
'label': ClassLabel(shape=(), dtype=tf.int64, num_classes=10),
})
Kunci yang diawasi (Lihat
as_supervised
doc ):('image', 'label')
Gambar ( tfds.show_examples ):
- Contoh ( tfds.as_dataframe ):
euroat / semua
Deskripsi konfigurasi : 13 saluran Sentinel-2
Ukuran unduhan :
1.93 GiB
Fitur :
FeaturesDict({
'filename': Text(shape=(), dtype=tf.string),
'label': ClassLabel(shape=(), dtype=tf.int64, num_classes=10),
'sentinel2': Tensor(shape=(64, 64, 13), dtype=tf.float32),
})
Kunci yang diawasi (Lihat
as_supervised
doc ):('sentinel2', 'label')
Gambar ( tfds.show_examples ): Tidak didukung.
Contoh ( tfds.as_dataframe ):