Deep Clustering for Mars Rover image datasets
This work addresses the time-consuming task of generating ground truth labels for Mars Rover and other planetary images, though it is incremental as it applies existing deep clustering methods to a new domain.
The paper tackles the problem of clustering unlabeled Mars Rover images using deep learning, achieving good accuracy and concordance with ground truth labels on a labeled test dataset.
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.