so2sat

  • Descrição :

So2Sat LCZ42 é um conjunto de dados que consiste em radar de abertura sintética co-registrado e manchas de imagens ópticas multiespectrais adquiridas pelos satélites de sensoriamento remoto Sentinel-1 e Sentinel-2 e o rótulo correspondente de zonas climáticas locais (LCZ). O conjunto de dados é distribuído em 42 cidades em diferentes continentes e regiões culturais do mundo.

O conjunto de dados completo ( all ) consiste em 8 canais Sentinel-1 e 10 canais Sentinel-2. Alternativamente, pode-se selecionar o subconjunto rgb , que contém apenas as bandas de frequência óptica do Sentinel-2, redimensionadas e codificadas como JPEG.

URL do conjunto de dados: http://doi.org/10.14459/2018MP1454690
Licença: http://creativecommons.org/licenses/by/4.0

Dividir Exemplos
@misc{mediatum1483140,
    author = {Zhu, Xiaoxiang  and  Hu, Jingliang  and  Qiu, Chunping  and  Shi, Yilei  and  Bagheri, Hossein  and  Kang, Jian  and  Li, Hao  and  Mou, Lichao  and  Zhang, Guicheng  and  Häberle, Matthias  and  Han, Shiyao  and  Hua, Yuansheng  and  Huang, Rong  and  Hughes, Lloyd  and  Sun, Yao  and  Schmitt, Michael and  Wang, Yuanyuan },
    title = {NEW: So2Sat LCZ42},
    publisher = {Technical University of Munich},
    url = {https://mediatum.ub.tum.de/1483140},
    type = {Dataset},
    year = {2019},
    doi = {10.14459/2018mp1483140},
    keywords = {local climate zones ; big data ; classification ; remote sensing ; deep learning ; data fusion ; synthetic aperture radar imagery ; optical imagery},
    abstract = {So2Sat LCZ42 is a dataset consisting of corresponding synthetic aperture radar and multispectral optical image data acquired by the Sentinel-1 and Sentinel-2 remote sensing satellites, and a corresponding local climate zones (LCZ) label. The dataset is distributed over 42 cities across different continents and cultural regions of the world, and comes with a split into fully independent, non-overlapping training, validation, and test sets.},
    language = {en},

}

so2sat/rgb (configuração padrão)

  • Descrição da configuração : Canais Sentinel-2 RGB

  • Estrutura de recursos :

FeaturesDict({
    'image': Image(shape=(32, 32, 3), dtype=uint8),
    'label': ClassLabel(shape=(), dtype=int64, num_classes=17),
    'sample_id': int64,
})
  • Documentação do recurso:
Característica Classe Forma Tipo D Descrição
RecursosDict
imagem Imagem (32, 32, 3) uint8
rótulo ClassLabel int64
sample_id tensor int64

so2sat/todos

  • Descrição da configuração : 8 canais Sentinel-1 e 10 Sentinel-2

  • Estrutura de recursos :

FeaturesDict({
    'label': ClassLabel(shape=(), dtype=int64, num_classes=17),
    'sample_id': int64,
    'sentinel1': Tensor(shape=(32, 32, 8), dtype=float32),
    'sentinel2': Tensor(shape=(32, 32, 10), dtype=float32),
})
  • Documentação do recurso:
Característica Classe Forma Tipo D Descrição
RecursosDict
rótulo ClassLabel int64
sample_id tensor int64
sentinela1 tensor (32, 32, 8) float32
sentinela2 tensor (32, 32, 10) float32