Dimensionality reduction based on binary encoding for hyperspectral data

What is it about?

Dimensionality reduction based on binary encoding for hyperspectral data is an approach that aims at summarizing the information contained in various spectral bands into a single image that stores the meaningful information of the bands. This is a feature extraction approach that permits reduce the dimensionality of hyperspectral data for classification purposes. Different options to reduce the radiometric information of the pixels are introduced, such as using a single threshold or multiple thresholds. The thematic mapping of the hyperspectral data with reduced dimension confirms the competitiveness of the binary encoding method compared with other dimension reduction methods.

Featured Image

Why is it important?

The binary encoding method aims at introducing a feature extraction alternative that can be applied to any hyperspectral data set. Also, it enables summarizing the spectral signatures of neighbouring bands, reducing computational complexity and improving data analysis performance. There are advantages when using spectral regions instead of the whole data set because it allows analysing the confusion between classes by region, preserving the original spectral information for classification purposes. Through the literature, there are several manuscripts presented that use hyperspectral images to reduce dimensionality and classification. But this manuscript presents a different approach to others and that could be an alternative when choosing a method to reduce dimensionality.

Perspectives