Introduction #
The Semantic Segmentation Satellite Imagery dataset was taken from the project for the Kaggle Competition organised by CentraleSupelec Deep Learning course. The training dataset consisted of 261 images taken by a small UAV in the area of Houston, Texas to assess the damages after Hurricane Harvey. Each pixel was segmented into one of 25 classes such as property roof, trees / shrubs, road / highway, swimming pool, vehicle, flooded, etc.
Summary #
Semantic Segmentation with PyTorch Satellite Imagery is a dataset for a semantic segmentation task. It is used in the geospatial domain. Possible applications of the dataset could be in the search and rescue (SAR) and environmental industries.
The dataset consists of 261 images with 2478 labeled objects belonging to 25 different classes including background, grass, trees / shrubs, and other: property roof, vehicle, chimney, secondary structure, swimming pool, power lines & cables, window, road / highway, flooded, dense vegetation / forest, street light, water body, garbage bins, trampoline, satellite antenna, parking area - commercial, solar panels, under construction / in progress status, boat, sports complex / arena, industrial site, and water tank / oil tank.
Images in the Semantic Segmentation Satellite Imagery dataset have pixel-level semantic segmentation annotations. All images are labeled (i.e. with annotations). There is 1 split in the dataset: train (261 images). The dataset was released in 2022.
Here are the visualized examples for the classes:
Explore #
Semantic Segmentation Satellite Imagery dataset has 261 images. Click on one of the examples below or open "Explore" tool anytime you need to view dataset images with annotations. This tool has extended visualization capabilities like zoom, translation, objects table, custom filters and more. Hover the mouse over the images to hide or show annotations.
Class balance #
There are 25 annotation classes in the dataset. Find the general statistics and balances for every class in the table below. Click any row to preview images that have labels of the selected class. Sort by column to find the most rare or prevalent classes.
Class ã…¤ | Images ã…¤ | Objects ã…¤ | Count on image average | Area on image average |
---|---|---|---|---|
backgroundâž” mask | 261 | 261 | 1 | 15.18% |
trees / shrubsâž” mask | 254 | 254 | 1 | 16.24% |
grassâž” mask | 254 | 254 | 1 | 29.21% |
property roofâž” mask | 222 | 222 | 1 | 13.33% |
vehicleâž” mask | 199 | 199 | 1 | 0.63% |
chimneyâž” mask | 150 | 150 | 1 | 0.11% |
secondary structureâž” mask | 144 | 144 | 1 | 0.49% |
swimming poolâž” mask | 142 | 142 | 1 | 0.83% |
power lines & cablesâž” mask | 137 | 137 | 1 | 1.8% |
windowâž” mask | 120 | 120 | 1 | 0.07% |
Co-occurrence matrix #
Co-occurrence matrix is an extremely valuable tool that shows you the images for every pair of classes: how many images have objects of both classes at the same time. If you click any cell, you will see those images. We added the tooltip with an explanation for every cell for your convenience, just hover the mouse over a cell to preview the description.
Images #
Explore every single image in the dataset with respect to the number of annotations of each class it has. Click a row to preview selected image. Sort by any column to find anomalies and edge cases. Use horizontal scroll if the table has many columns for a large number of classes in the dataset.
Object distribution #
Interactive heatmap chart for every class with object distribution shows how many images are in the dataset with a certain number of objects of a specific class. Users can click cell and see the list of all corresponding images.
Class sizes #
The table below gives various size properties of objects for every class. Click a row to see the image with annotations of the selected class. Sort columns to find classes with the smallest or largest objects or understand the size differences between classes.
Class | Object count | Avg area | Max area | Min area | Min height | Min height | Max height | Max height | Avg height | Avg height | Min width | Min width | Max width | Max width |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
background mask | 261 | 15.18% | 64.03% | 0.37% | 3341px | 72.76% | 4592px | 100% | 4159px | 99.79% | 3000px | 100% | 3072px | 100% |
trees / shrubs mask | 254 | 16.24% | 54.68% | 0.14% | 206px | 5.15% | 4592px | 100% | 3207px | 76.53% | 342px | 11.13% | 3072px | 100% |
grass mask | 254 | 29.21% | 95.97% | 0.3% | 279px | 6.97% | 4592px | 100% | 3487px | 83.87% | 201px | 6.7% | 3072px | 100% |
property roof mask | 222 | 13.33% | 35.84% | 0.04% | 54px | 1.18% | 4592px | 100% | 2380px | 55.9% | 104px | 3.47% | 3072px | 100% |
vehicle mask | 199 | 0.63% | 7.84% | 0% | 1px | 0.03% | 4587px | 99.95% | 1724px | 40.14% | 41px | 1.37% | 3072px | 100% |
chimney mask | 150 | 0.11% | 1.37% | 0% | 1px | 0.03% | 4497px | 97.93% | 1392px | 32.16% | 1px | 0.03% | 2997px | 98.1% |
secondary structure mask | 144 | 0.49% | 3.92% | 0.01% | 46px | 1.15% | 4330px | 99.75% | 1185px | 27.6% | 35px | 1.17% | 3065px | 99.87% |
swimming pool mask | 142 | 0.83% | 2.94% | 0.08% | 74px | 1.85% | 4488px | 97.74% | 1325px | 30.54% | 85px | 2.83% | 3069px | 99.97% |
power lines & cables mask | 137 | 1.8% | 25.89% | 0% | 29px | 0.63% | 4592px | 100% | 1303px | 31.06% | 43px | 1.4% | 3072px | 100% |
window mask | 120 | 0.07% | 0.3% | 0.01% | 23px | 0.57% | 4560px | 99.3% | 1120px | 26.37% | 28px | 0.93% | 3066px | 99.8% |
Spatial Heatmap #
The heatmaps below give the spatial distributions of all objects for every class. These visualizations provide insights into the most probable and rare object locations on the image. It helps analyze objects' placements in a dataset.
Objects #
Table contains all 2478 objects. Click a row to preview an image with annotations, and use search or pagination to navigate. Sort columns to find outliers in the dataset.
Object ID ã…¤ | Class ã…¤ | Image name click row to open | Image size height x width | Height ã…¤ | Height ã…¤ | Width ã…¤ | Width ã…¤ | Area ã…¤ |
---|---|---|---|---|---|---|---|---|
1âž” | background mask | 6625.jpg | 4592 x 3072 | 4592px | 100% | 3072px | 100% | 33.07% |
2âž” | grass mask | 6625.jpg | 4592 x 3072 | 3983px | 86.74% | 3068px | 99.87% | 34.7% |
3âž” | trees / shrubs mask | 6625.jpg | 4592 x 3072 | 1730px | 37.67% | 3061px | 99.64% | 12.36% |
4âž” | street light mask | 6625.jpg | 4592 x 3072 | 156px | 3.4% | 50px | 1.63% | 0.03% |
5âž” | garbage bins mask | 6625.jpg | 4592 x 3072 | 45px | 0.98% | 49px | 1.6% | 0.01% |
6âž” | road / highway mask | 6625.jpg | 4592 x 3072 | 788px | 17.16% | 3064px | 99.74% | 7.96% |
7âž” | flooded mask | 6625.jpg | 4592 x 3072 | 1403px | 30.55% | 1430px | 46.55% | 11.87% |
8âž” | background mask | 7239.jpg | 4592 x 3072 | 4592px | 100% | 3072px | 100% | 18.03% |
9âž” | property roof mask | 7239.jpg | 4592 x 3072 | 3959px | 86.22% | 3072px | 100% | 22.44% |
10âž” | swimming pool mask | 7239.jpg | 4592 x 3072 | 1045px | 22.76% | 775px | 25.23% | 0.75% |
License #
Citation #
If you make use of the Semantic-segmentation-Satellite-Imagery data, please cite the following reference:
@misc{alchimowicz_2022,
title={semantic_segmentation_satellite_imagery},
url={https://figshare.com/collections/semantic_segmentation_satellite_imagery/6026765/1}
DOI={10.6084/m9.figshare.c.6026765.v1},
abstractNote={<p>satellite imagery semantic segmentation</p>},
publisher={figshare},
author={Alchimowicz, Jedrzej},
year={2022},
month={Jun}
}
If you are happy with Dataset Ninja and use provided visualizations and tools in your work, please cite us:
@misc{ visualization-tools-for-semantic-segmentation-satellite-imagery-dataset,
title = { Visualization Tools for Semantic Segmentation Satellite Imagery Dataset },
type = { Computer Vision Tools },
author = { Dataset Ninja },
howpublished = { \url{ https://datasetninja.com/semantic-segmentation-satellite-imagery } },
url = { https://datasetninja.com/semantic-segmentation-satellite-imagery },
journal = { Dataset Ninja },
publisher = { Dataset Ninja },
year = { 2024 },
month = { dec },
note = { visited on 2024-12-07 },
}
Download #
Dataset Semantic Segmentation Satellite Imagery can be downloaded in Supervisely format:
As an alternative, it can be downloaded with dataset-tools package:
pip install --upgrade dataset-tools
… using following python code:
import dataset_tools as dtools
dtools.download(dataset='Semantic Segmentation Satellite Imagery', dst_dir='~/dataset-ninja/')
Make sure not to overlook the python code example available on the Supervisely Developer Portal. It will give you a clear idea of how to effortlessly work with the downloaded dataset.
The data in original format can be downloaded here:
Disclaimer #
Our gal from the legal dep told us we need to post this:
Dataset Ninja provides visualizations and statistics for some datasets that can be found online and can be downloaded by general audience. Dataset Ninja is not a dataset hosting platform and can only be used for informational purposes. The platform does not claim any rights for the original content, including images, videos, annotations and descriptions. Joint publishing is prohibited.
You take full responsibility when you use datasets presented at Dataset Ninja, as well as other information, including visualizations and statistics we provide. You are in charge of compliance with any dataset license and all other permissions. You are required to navigate datasets homepage and make sure that you can use it. In case of any questions, get in touch with us at hello@datasetninja.com.