How to citeΒΆ

If you use the toolbox in your research, please cite: Windrim et al. Unsupervised Feature-Learning for Hyperspectral Data with Autoencoders. Remote Sensing 11.7 (2019): 864. This paper explains the spectral angle (SA), spectral information divergence (SID) and sum-of-squared errors (SSE) loss functions for training autoencoders.

If you use the cosine spectral angle (CSA) loss function in your research, please cite: Windrim et al. Unsupervised feature learning for illumination robustness. 2016 IEEE International Conference on Image Processing (ICIP).

If you use the classification networks in your research, please cite: Windrim et al. Hyperspectral CNN Classification with Limited Training Samples. 2017 Proceedings of the British Machine Vision Conference (BMVC).