arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1806.02583cs.NEcs.CV
收藏

用于高光谱样本真实感合成的生成对抗网络

Generative Adversarial Networks for Realistic Synthesis of Hyperspectral Samples

ONERA, The French Aerospace Lab(法国航空航天实验室(ONERA));Univ. Bretagne-Sud(南布列塔尼大学)

Nicolas Audebert, Bertrand Le Saux, Sébastien Lefèvre

摘要:

这项工作解决了训练深度神经网络所需的注释高光谱数据的稀缺性。特别地,我们研究了生成对抗网络及其在一致性标记光谱合成中的应用。通过在公共数据集上训练这样的网络,我们证明了这些模型不仅能够捕获潜在的分布,而且还能够生成外观相似且物理上合理的光谱。此外,我们通过实验验证了合成样本可以作为一种有效的数据增强策略。我们使用各种深度分类器在几个公共高光谱数据集上验证了我们的方法。

英文摘要:

This work addresses the scarcity of annotated hyperspectral data required to train deep neural networks. Especially, we investigate generative adversarial networks and their application to the synthesis of consistent labeled spectra. By training such networks on public datasets, we show that these models are not only able to capture the underlying distribution, but also to generate genuine-looking and physically plausible spectra. Moreover, we experimentally validate that the synthetic samples can be used as an effective data augmentation strategy. We validate our approach on several public hyper-spectral datasets using a variety of deep classifiers.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑