metadata

language: en
license: unknown
task_categories:
  - image-classification
paperswithcode_id: patternnet
pretty_name: PatternNet
tags:
  - remote-sensing
  - earth-observation
  - geospatial
  - satellite-imagery
  - land-cover-classification
  - google-earth
dataset_info:
  features:
    - name: image
      dtype: image
    - name: label
      dtype:
        class_label:
          names:
            '0': airplane
            '1': baseball field
            '2': basketball court
            '3': beach
            '4': bridge
            '5': cemetery
            '6': chaparral
            '7': christmas tree farm
            '8': closed road
            '9': coastal mansion
            '10': crosswalk
            '11': dense residential
            '12': ferry terminal
            '13': football field
            '14': forest
            '15': freeway
            '16': golf course
            '17': harbor
            '18': intersection
            '19': mobile home park
            '20': nursing home
            '21': oil gas field
            '22': oil well
            '23': overpass
            '24': parking lot
            '25': parking space
            '26': railway
            '27': river
            '28': runway
            '29': runway marking
            '30': shipping yard
            '31': solar panel
            '32': sparse residential
            '33': storage tank
            '34': swimming pool
            '35': tennis court
            '36': transformer station
            '37': wastewater treatment plant
  splits:
    - name: train
      num_bytes: 1422177005
      num_examples: 30400
  download_size: 1422316869
  dataset_size: 1422177005
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*

PatternNet

The PatternNet dataset is a dataset for remote sensing scene classification and image retrieval.

Paper: https://arxiv.org/abs/1703.06339
Homepage: https://sites.google.com/view/zhouwx/dataset

Description

PatternNet is a large-scale high-resolution remote sensing dataset collected for remote sensing image retrieval. There are 38 classes and each class has 800 images of size 256×256 pixels. The images in PatternNet are collected from Google Earth imagery or via the Google Map API for some US cities. The following table shows the classes and the corresponding spatial resolutions. The figure shows some example images from each class.

Total Number of Images: 30400
Bands: 3 (RGB)
Image Resolution: 256x256m
Land Cover Classes: 38
Classes: airplane, baseball_field, basketball_court, beach, bridge, cemetery, chaparral, christmas_tree_farm, closed_road, coastal_mansion, crosswalk, dense_residential, ferry_terminal, football_field, forest, freeway, golf_course, harbor, intersection, mobile_home_park, nursing_home, oil_gas_field, oil_well, overpass, parking_lot, parking_space, railway, river, runway, runway_marking, shipping_yard, solar_panel, sparse_residential, storage_tank, swimming_pool, tennis_court, transformer_station, wastewater_treatment_plant

Usage

To use this dataset, simply use datasets.load_dataset("blanchon/PatternNet").

from datasets import load_dataset
PatternNet = load_dataset("blanchon/PatternNet")

Citation

If you use the EuroSAT dataset in your research, please consider citing the following publication:

@article{li2017patternnet,
  title     = {PatternNet: Visual Pattern Mining with Deep Neural Network},
  author    = {Hongzhi Li and Joseph G. Ellis and Lei Zhang and Shih-Fu Chang},
  journal   = {International Conference on Multimedia Retrieval},
  year      = {2017},
  doi       = {10.1145/3206025.3206039},
  bibSource = {Semantic Scholar https://www.semanticscholar.org/paper/e7c75e485651bf3ccf37dd8dd39f6665419d73bd}
}