The Mars Perseverance rover applies computer vision for navigation and hazard avoidance. The challenge to do onboard object recognition highlights the need for low-power, customized training, often including low-contrast backgrounds. We investigate deep learning methods for the classification and detection of Martian rocks. We report greater than 97% accuracy for binary classifications (rock vs. rover). We fine-tune a detector to render geo-located bounding boxes while counting rocks. For these models to run on microcontrollers, we shrink and quantize the neural networks' weights and demonstrate a low-power rock hunter with faster frame rates (1 frame per second) but lower accuracy (37%).
@article{arxiv.2104.04359,
title = {Rock Hunting With Martian Machine Vision},
author = {David Noever and Samantha E. Miller Noever},
journal= {arXiv preprint arXiv:2104.04359},
year = {2021}
}