This repository contains the code and files for Spatial and Temporal Data Mining Project titled "Spatial Predictions using Satellite Images".
Defense Science and Technology Laboratory (DSTL) has provided 1km x 1km satellite images in both 3 band and 16 band formats. It can be downloaded from here.
The dataset has been labelled into 10 different classes:
• Buildings
• Small Vehicle
• Road
• Large Vehicle
• Track
• Miscellaneous Structures
• Trees
• Waterway
• Standing Water
• Crops
| Type | Value |
|---|---|
| Sensor | WorldView3 |
| 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 |
• TensorFlow
• Numpy
• OpenCV
• Pandas
• Shapely
• Os
• Keras
• CSV
• Tifffle
• Skimage
• sys
Run train.py for training.