Using UMAP for Dimensionality Reduction of Hyperspectral Data
Author:
- 6 October 2020
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
Abstract
Uniform Manifold Approximation and Projection (UMAP) is a recent technique for dimensionality reduction. In this paper, we study UMAP for hyperspectral data processing. For well-known hyperspectral images, we search for optimal parameters and compare UMAP to other dimensionality reduction techniques from the viewpoint of the classification quality. The list of alternative techniques includes the linear Principal Component Analysis technique as well as nonlinear methods, namely, Isomap, Locally Linear Embedding, Laplacian Eigenmaps, and Nonlinear Mapping.Keywords
Funding Information
- RFBR (18-07-01312-a.)
This publication has 6 references indexed in Scilit:
- Improving the Efficiency of Automated Precision Robotics-Enabled Positioning and Welding2019
- Nonlinear Dimensionality Reduction by Locally Linear EmbeddingScience, 2000
- A Global Geometric Framework for Nonlinear Dimensionality ReductionScience, 2000
- Principal Component Analysis1986
- 220 Band AVIRIS Hyperspectral Image Data Set: June 12, 1992 Indian Pine Test Site 3
- A Nonlinear Mapping for Data Structure AnalysisIEEE Transactions on Computers, 1969