HIGHLIGHTS
- who: (RDN). and collaborators from the (UNIVERSITY) have published the research work: Residual Dense Autoencoder Network for Nonlinear Hyperspectral Unmixing, in the Journal: (JOURNAL)
- what: When the first convolution layer of the model is regarded as low-level feature extraction, the consecutive layers aim to mine distinct representations of data. The authors adopted the mean spectral information divergence (mSID) and mean spectral angle distance (mSAD) to quantitatively evaluate the performance of all the methods in endmember extraction. The authors compared the proposed RDAE with other representative hyperspectral unmixing algorithms, which can be categorized into one . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.