Mục lục bài viết

GID-15: A large scale semantic segmentation dataset for remote sensing images

Abstract

In recent years, large amount of high spatial-resolution remote sensing (HRRS) images are available for land-cover mapping.
However, due to the complex information brought by the increased spatial resolution and the data disturbances caused by
different conditions of image acquisition, it is often difficult to find an efficient method for achieving accurate land-cover
classification with high-resolution and heterogeneous remote sensing images.
We create a large scale semantic segmentation dataset for remote sensing images containing 150 Gaofen-2 satellite images, 100 images, 10 images and 40 images for training, validating and testing respectively.

Dataset

We construct a new large-scale land-cover dataset with Gaofen-2 (GF-2) satellite images. This new dataset, which is named as
Gaofen Image Dataset with 15 categories (GID-15), has superiorities over the existing land-cover dataset because of its large coverage, wide distribution,
and high spatial resolution.
The large-scale remote sensing semantic segmentation set contains 150 pixel-level annotated GF-2 images, which is labeled in 15 categories.
Some of the images are from the paper: Land-Cover Classification with High-Resolution Remote Sensing Images Using Transferable Deep Models

Data Compose

We will not release the annotation for testing set, so we can get fair comparative results through online benchmark.
In addition, we provide two types of ground truth, ‘.png’ and ‘.tiff’ format respectively.
the ‘.png’ format ground truth is grey label, while ‘.tiff’ format is ‘RGB’ label, the color palette pls refer to the readme.txt in dataset file.

– large-scale remote sensing semantic segmentation set

– Download

GID-15 can be download from google drive:

Citation

@article{GID2020,
title = {Land-Cover Classification with High-Resolution Remote Sensing Images Using Transferable Deep Models},
author = {Xin-Yi Tong, Gui-Song Xia, Qikai Lu, Huangfeng Shen, Shengyang Li, Shucheng You, Liangpei Zhang},
journal = {Remote Sensing of Environment, doi: 10.1016/j.rse.2019.111322},
year = {2020}
}