火星图像内容分类：NASA三年部署与近期进展
机器学习
2021-02-10 v1 计算机视觉与模式识别
图像与视频处理
摘要
NASA行星数据系统托管了从火星获取的数百万张图像。为帮助用户快速找到感兴趣的图像，我们开发并部署了针对火星轨道和表面图像的内容分类与检索能力。所部署的系统可通过PDS图像图谱公开访问。我们描述了针对火星任务采集的图像训练、评估、校准和部署两个CNN分类器更新的过程。我们还报告了三年的部署情况，包括使用统计、经验教训和未来计划。
引用
@article{arxiv.2102.05011,
title = {Mars Image Content Classification: Three Years of NASA Deployment and Recent Advances},
author = {Kiri Wagstaff and Steven Lu and Emily Dunkel and Kevin Grimes and Brandon Zhao and Jesse Cai and Shoshanna B. Cole and Gary Doran and Raymond Francis and Jake Lee and Lukas Mandrake},
journal= {arXiv preprint arXiv:2102.05011},
year = {2021}
}
备注
Published at the Thirty-Third Annual Conference on Innovative Applications of Artificial Intelligence (IAAI-21). IAAI Innovative Application Award. 10 pages, 11 figures, 6 tables