Dropout Concrete Autoencoder for Band Selection on HSI Scenes

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Xu, Lei, Ahishali, Mete, Gabbouj, Moncef
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913087570837504
author Xu, Lei
Ahishali, Mete
Gabbouj, Moncef
author_facet Xu, Lei
Ahishali, Mete
Gabbouj, Moncef
contents Deep learning-based informative band selection methods on hyperspectral images (HSI) recently have gained intense attention to eliminate spectral correlation and redundancies. However, the existing deep learning-based methods either need additional post-processing strategies to select the descriptive bands or optimize the model indirectly, due to the parameterization inability of discrete variables for the selection procedure. To overcome these limitations, this work proposes a novel end-to-end network for informative band selection. The proposed network is inspired by the advances in concrete autoencoder (CAE) and dropout feature ranking strategy. Different from the traditional deep learning-based methods, the proposed network is trained directly given the required band subset eliminating the need for further post-processing. Experimental results on four HSI scenes show that the proposed dropout CAE achieves substantial and effective performance levels outperforming the competing methods.
format Preprint
id arxiv_https___arxiv_org_abs_2401_16522
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Dropout Concrete Autoencoder for Band Selection on HSI Scenes
Xu, Lei
Ahishali, Mete
Gabbouj, Moncef
Computer Vision and Pattern Recognition
Deep learning-based informative band selection methods on hyperspectral images (HSI) recently have gained intense attention to eliminate spectral correlation and redundancies. However, the existing deep learning-based methods either need additional post-processing strategies to select the descriptive bands or optimize the model indirectly, due to the parameterization inability of discrete variables for the selection procedure. To overcome these limitations, this work proposes a novel end-to-end network for informative band selection. The proposed network is inspired by the advances in concrete autoencoder (CAE) and dropout feature ranking strategy. Different from the traditional deep learning-based methods, the proposed network is trained directly given the required band subset eliminating the need for further post-processing. Experimental results on four HSI scenes show that the proposed dropout CAE achieves substantial and effective performance levels outperforming the competing methods.
title Dropout Concrete Autoencoder for Band Selection on HSI Scenes
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2401.16522