blanchon/PatternNet

PatternNet是一个用于遥感场景分类和图像检索的大规模高分辨率遥感数据集。该数据集包含38个类别，每个类别有800张256×256像素的图像，总共有30400张图像。这些图像来源于Google Earth或Google Map API，覆盖了多种地物类型，如飞机、棒球场、篮球场、海滩、桥梁等。图像分辨率为256x256米，包含3个波段（RGB）。

The PatternNet dataset is a large-scale high-resolution remote sensing dataset for remote sensing scene classification and image retrieval. It consists of 38 classes, each with 800 images of size 256x256 pixels. The images are sourced from Google Earth imagery or via the Google Map API for some US cities. The dataset features include images and labels, with labels covering 38 different land cover classes. The dataset is divided into a training set, containing 30400 samples.

原始信息汇总

PatternNet 数据集概述

基本信息

• 语言: 英文
• 任务类别: 图像分类
• 标签: 遥感, 地球观测, 地理空间, 卫星图像, 土地覆盖分类, Google Earth
• 数据集名称: PatternNet

数据集详情

• 特征:

  • image: 图像数据
  • label: 标签数据，包含38个类别
    • 类别名称: 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

• 数据分割:

  • train: 训练集，包含30400个样本，大小为1422177005.0字节

• 数据集大小:

  • 下载大小: 1422316869字节
  • 数据集大小: 1422177005.0字节

数据集描述

• 总图像数: 30400
• 波段: 3 (RGB)
• 图像分辨率: 256x256像素
• 土地覆盖类别: 38

使用方法

• 加载数据集: datasets.load_dataset("blanchon/PatternNet")

引用

• 引用信息: bibtex @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} }

搜集汇总

背景与挑战

背景概述

PatternNet是一个包含38个类别、总计30400张256x256像素图像的大规模遥感数据集，适用于场景分类和图像检索任务，图像来源于Google Earth和Google Map API。