Rotation Equivariant Deforestation Segmentation and Driver Classification

TL;DR

This paper introduces a rotation equivariant CNN model for deforestation segmentation and driver classification from satellite images, achieving significant accuracy improvements and rotational stability over previous methods.

Contribution

The work presents a novel rotation equivariant neural network that enhances deforestation analysis by improving accuracy and stability under image rotation.

Findings

9% improvement in driver classification accuracy

7% improvement in segmentation map accuracy

Predicted segmentation maps are stable under input image rotation

Abstract

Deforestation has become a significant contributing factor to climate change and, due to this, both classifying the drivers and predicting segmentation maps of deforestation has attracted significant interest. In this work, we develop a rotation equivariant convolutional neural network model to predict the drivers and generate segmentation maps of deforestation events from Landsat 8 satellite images. This outperforms previous methods in classifying the drivers and predicting the segmentation map of deforestation, offering a 9% improvement in classification accuracy and a 7% improvement in segmentation map accuracy. In addition, this method predicts stable segmentation maps under rotation of the input image, which ensures that predicted regions of deforestation are not dependent upon the rotational orientation of the satellite.