DeepSat - A Learning framework for Satellite Imagery

TL;DR

This paper introduces two new high-resolution satellite datasets, SAT-4 and SAT-6, and proposes a deep learning framework using Deep Belief Networks that significantly improves classification accuracy over existing methods.

Contribution

The paper presents novel satellite datasets and a deep learning classification framework that outperforms existing algorithms in satellite image recognition tasks.

Findings

Achieved 97.95% accuracy on SAT-4, outperforming other algorithms by ~11%.

Achieved 93.9% accuracy on SAT-6, outperforming others by ~15%.

Demonstrated the effectiveness of unsupervised learning over supervised methods.

Abstract

Satellite image classification is a challenging problem that lies at the crossroads of remote sensing, computer vision, and machine learning. Due to the high variability inherent in satellite data, most of the current object classification approaches are not suitable for handling satellite datasets. The progress of satellite image analytics has also been inhibited by the lack of a single labeled high-resolution dataset with multiple class labels. The contributions of this paper are twofold - (1) first, we present two new satellite datasets called SAT-4 and SAT-6, and (2) then, we propose a classification framework that extracts features from an input image, normalizes them and feeds the normalized feature vectors to a Deep Belief Network for classification. On the SAT-4 dataset, our best network produces a classification accuracy of 97.95% and outperforms three state-of-the-art object recognition algorithms, namely - Deep Belief Networks, Convolutional Neural Networks and Stacked Denoising Autoencoders by ~11%. On SAT-6, it produces a classification accuracy of 93.9% and outperforms the other algorithms by ~15%. Comparative studies with a Random Forest classifier show the advantage of an unsupervised learning approach over traditional supervised learning techniques. A statistical analysis based on Distribution Separability Criterion and Intrinsic Dimensionality Estimation substantiates the effectiveness of our approach in learning better representations for satellite imagery.