A GPU Accelerated Hyperspectral 3D Convolutional Neural Network Classification at the Edge with Principal Component Analysis Preprocessing.
Gianluca De LuciaMarco LapegnaDiego RomanoPublished in: PPAM (2) (2022)
Keyphrases
- hyperspectral
- convolutional neural network
- preprocessing
- hyperspectral data
- hyperspectral images
- gpu accelerated
- principal component analysis
- pixel classification
- hyperspectral image classification
- remote sensing
- feature extraction
- infrared
- hyperspectral imagery
- face detection
- multispectral
- dimension reduction
- random projections
- principal components
- hyperspectral imaging
- image data
- pattern recognition
- spatial resolution
- target detection
- covariance matrix
- machine learning
- training set
- satellite images
- edge detection
- feature space
- edge information
- co occurrence
- high resolution
- image analysis
- spectral data
- spectral bands
- face recognition
- computer vision