Data Description:

In this competition, Dstl provides you with 1km x 1km satellite images in both 3-band and 16-band formats. Your goal is to detect and classify the types of objects found in these regions. 

3- and 16-bands images

There are two types of imagery spectral content provided in this competition. The 3-band images are the traditional RGB natural color images. The 16-band images contain spectral information by capturing wider wavelength channels. This multi-band imagery is taken from the multispectral (400 – 1040nm) and short-wave infrared (SWIR) (1195-2365nm) range. All images are in GeoTiff format and might require GeoTiff viewers (such as QGIS) to view. Please refer to our tutorial on how to programmatically view the images.

All imagery credit to: Satellite Imagery © DigitalGlobe, Inc.

Imagery details

• Sensor : WorldView 3

• Wavebands :

• Panchromatic: 450-800 nm

• 8 Multispectral: (red, red edge, coastal, blue, green, yellow, near-IR1 and near-IR2) 400 nm - 1040 nm

• 8 SWIR: 1195 nm - 2365 nm

• Sensor Resolution (GSD) at Nadir :

• Panchromatic: 0.31m 

• Multispectral: 1.24 m

• SWIR: Delivered at 7.5m

• Dynamic Range

• Panchromatic and multispectral : 11-bits per pixel

• SWIR : 14-bits per pixel

Object types

In a satellite image, you will find lots of different objects like roads, buildings, vehicles, farms, trees, water ways, etc. Dstl has labeled 10 different classes:

1. Buildings - large building, residential, non-residential, fuel storage facility, fortified building

2. Misc. Manmade structures 

3. Road 

4. Track - poor/dirt/cart track, footpath/trail

5. Trees - woodland, hedgerows, groups of trees, standalone trees

6. Crops - contour ploughing/cropland, grain (wheat) crops, row (potatoes, turnips) crops

7. Waterway 

8. Standing water

9. Vehicle Large - large vehicle (e.g. lorry, truck,bus), logistics vehicle

10. Vehicle Small - small vehicle (car, van), motorbike

Every object class is described in the form of Polygons and MultiPolygons, which are simply a list of polygons. We provide two different formats for these shapes: GeoJson and WKT. These are both open source formats for geo-spatial shapes. 

Your submission will be in the WKT format. 

Geo Coordinates

In this dataset that we provide, we create a set of geo-coordinates that are in the range of x = [0,1] and y = [-1,0]. These coordinates are transformed such that we obscure the location of where the satellite images are taken from. The images are from the same region on Earth.

To utilize these images, we provide the grid coordinates of each image so you know how to scale them and align them with the images in pixels. You need the Xmax and Ymin for each image to do the scaling (provided in our grid_sizes.csv) Please refer to our tutorial on how to programmatically view the images.

File descriptions

• train_wkt.csv - the WKT format of all the training labels

• ImageId - ID of the image

• ClassType - type of objects (1-10)

• MultipolygonWKT - the labeled area, which is multipolygon geometry represented in WKT format 

• three_band.zip - the complete dataset of 3-band satellite images. The three bands are in the images with file name = {ImageId}.tif. MD5 = 7cf7bf17ba3fa3198a401ef67f4ef9b4 

• sixteen_band.zip - the complete dataset of 16-band satellite images. The 16 bands are distributed in the images with file name = {ImageId}_{A/M/P}.tif. MD5 = e2949f19a0d1102827fce35117c5f08a

• grid_sizes.csv - the sizes of grids for all the images

• ImageId - ID of the image

• Xmax - maximum X coordinate for the image

• Ymin - minimum Y coordinate for the image

• sample_submission.csv - a sample submission file in the correct format

• ImageId - ID of the image

• ClassType - type of objects (1-10)

• MultipolygonWKT - the labeled area, which is multipolygon geometry represented in WKT format

• train_geojson.zip - the geojson format of all the training labels (essentially these are the same information as train_wkt.csv) 

filename_to_classType = {
'001_MM_L2_LARGE_BUILDING':1,
'001_MM_L3_RESIDENTIAL_BUILDING':1,
'001_MM_L3_NON_RESIDENTIAL_BUILDING':1,
'001_MM_L5_MISC_SMALL_STRUCTURE':2,
'002_TR_L3_GOOD_ROADS':3,
'002_TR_L4_POOR_DIRT_CART_TRACK':4,
'002_TR_L6_FOOTPATH_TRAIL':4,
'006_VEG_L2_WOODLAND':5,
'006_VEG_L3_HEDGEROWS':5,
'006_VEG_L5_GROUP_TREES':5,
'006_VEG_L5_STANDALONE_TREES':5,
'007_AGR_L2_CONTOUR_PLOUGHING_CROPLAND':6,
'007_AGR_L6_ROW_CROP':6, 
'008_WTR_L3_WATERWAY':7,
'008_WTR_L2_STANDING_WATER':8,
'003_VH_L4_LARGE_VEHICLE':9,
'003_VH_L5_SMALL_VEHICLE':10,
'003_VH_L6_MOTORBIKE':10}