Deep Clustering for Mars Rover image datasets

TL;DR

This paper introduces a deep learning-based unsupervised clustering method using autoencoders and k-means to analyze Mars Rover images, aiming to facilitate faster labeling and analysis of planetary image datasets.

Contribution

It is the first to apply deep learning unsupervised clustering to Mars Rover images, improving data annotation efficiency.

Findings

Achieved good accuracy in clustering Mars Rover images

Demonstrated concordance with ground truth labels

Potential to accelerate dataset annotation processes

Abstract

In this paper, we build autoencoders to learn a latent space from unlabeled image datasets obtained from the Mars rover. Then, once the latent feature space has been learnt, we use k-means to cluster the data. We test the performance of the algorithm on a smaller labeled dataset, and report good accuracy and concordance with the ground truth labels. This is the first attempt to use deep learning based unsupervised algorithms to cluster Mars Rover images. This algorithm can be used to augment human annotations for such datasets (which are time consuming) and speed up the generation of ground truth labels for Mars Rover image data, and potentially other planetary and space images.