Feature Selection for Hyperspectral Images using Single-Layer Neural Networks

Abstract

Hyperspectral image classification by means of Deep Learning techniques is now a widespread practice. Its success comes from the abstract features learned by the deep architecture that are ultimately well separated in the feature space. The great amount of parameters to be learned requires the training data set to be very large, otherwise the risk of overfitting appears. Alternatively, one can resort to features selection in order to decrease the architecture's number of parameters to be learnt. For that purpose, this work proposes a simple feature selection method, based on single-layer neural networks, which select the most distinguishing features for each class. Then, the data will be classified by a deep neural network. The accuracy results for the testing data are higher for the lower dimensional data set when compared to the full data set, indicating less overfitting for the reduced data. Besides, a metric based on scatter matrices shows that the classes are better separated in the reduced feature space.

EEG-Based Brain-Computer Interfaces Using Motor-Imagery: Techniques and Challenges

“…For example, the potential of stacked denoising auto encoders, which has been used to locate robust features [109], could be explored. Also, network structure and training could incorporate feature selection elements such as in [110,111,112], where features which are not strongly discriminative according to some criterion are suppressed by the network. Architectures using statistical tests such as the t -test [113] or chi-squared test [114] to identify the most discriminative features could also be investigated.…”

Task-Driven Learned Hyperspectral Data Reduction Using End-to-End Supervised Deep Learning

“…In hyperspectral imaging, data are collected for a large number of spectral bins from a wavelength range in the electromagnetic spectrum. It is used in various fields [ 1 ], including agriculture classification [ 2 , 3 ], medical imaging [ 2 , 4 , 5 ], luggage and cargo inspection [ 2 , 6 , 7 ] and food quality assessment [ 8 ], as well as with energy-dispersive X-ray spectroscopy (EDX) and electron energy loss spectroscopy (EELS) [ 9 ]. In addition to the spatial dimensions, hyperspectral data include the spectral dimension which is typically large [ 10 ], often in the order of to spectral bins [ 11 ].…”