Hyperspectral Image Classification Using Deep Learning Models: A Review

Abstract

Hyperspectral image (HSI) classification is one of the important topic in the field of remote sensing. In general, HSI has to deal with complex characteristics and nonlinearity among the hyperspectral data which makes the classification task very challenging for traditional machine learning (ML) models. Recently, deep learning (DL) models have been very widely used in the classification of HSIs because of their capability to deal with complexity and nonlinearity in data. The utilization of deep learning models has been very successful and demonstrated good performance in the classification of HSIs. This paper presents a comprehensive review of deep learning models utilized in HSI classification literature and a comparison of various deep learning strategies for this topic. Precisely, the authors have categorized the literature review based upon the utilization of five most popular deep learning models and summarized their main methodologies used in feature extraction. This work may provide useful guidelines for the future research work in this area.

Stochastic Neighbor Embedding Feature-Based Hyperspectral Image Classification Using 3D Convolutional Neural Network

“…They also exploited the labeled and unlabeled samples. Bidari et al also employed a deep learning algorithm to classify hyperspectral images [34]. In addition, more recently, many researchers have proposed deep learning, including some dimension reduction to improve the efficiency of the hyperspectral-image-classification model [31,[35][36][37][38][39][40].…”

AI-based segmentation of intraoperative glioblastoma hyperspectral images

“…Therefore, there is a present demand to research new imaging modalities that could overcome such limitations 2 . Hyperspectral imaging (HSI) is a non-invasive, nonionizing and label-free technique, initially designed for remote-sensing and military purposes, becoming more popular in medicine for cancer detection thanks to the recent technological advances [10][11][12] . Hyperspectral (HS) images measure the reflected or transmitted light from the captured scene, collecting light-matter interaction values associated with several spectral bands, or wavelengths, of the electromagnetic spectrum range 11,13 .…”