EndNet: Sparse AutoEncoder Network for Endmember Extraction and Hyperspectral Unmixing (2017-08-06T00:00:00.000000Z)

TL;DR

This paper proposes a novel endmember extraction and hyperspectral unmixing scheme, so-called EndNet, that is based on a two-staged autoencoder network that is scalable for large-scale data and it can be accelerated on graphical processing units.

Abstract

Data acquired from multichannel sensors are a highly valuable asset to interpret the environment for a variety of remote sensing applications. However, low spatial resolution is a critical limitation for previous sensors, and the constituent materials of a scene can be mixed in different fractions due to their spatial interactions. Spectral unmixing is a technique that allows us to obtain the material spectral signatures and their fractions from hyperspectral data. In this paper, we propose a novel endmember extraction and hyperspectral unmixing scheme, so-called EndNet, that is based on a two-staged autoencoder network. This well-known structure is completely enhanced and restructured by introducing additional layers and a projection metric [i.e., spectral angle distance (SAD) instead of inner product] to achieve an optimum solution. Moreover, we present a novel loss function that is composed of a Kullback–Leibler divergence term with SAD similarity and additional penalty terms to improve the sparsity of the estimates. These modifications enable us to set the common properties of endmembers, such as nonlinearity and sparsity for autoencoder networks. Finally, due to the stochastic-gradient-based approach, the method is scalable for large-scale data and it can be accelerated on graphical processing units. To demonstrate the superiority of our proposed method, we conduct extensive experiments on several well-known data sets. The results confirm that the proposed method considerably improves the performance compared to the state-of-the-art techniques in the literature.

Authors

Savas Ozkan

1 papers

Berk Kaya

1 papers

G. Akar

3 papers

TL;DR

Abstract

Authors

References90 items

Neural network hyperspectral unmixing with spectral information divergence objective

A Stepwise Analytical Projected Gradient Descent Search for Hyperspectral Unmixing and Its Code Vectorization

Cosine Normalization: Using Cosine Similarity Instead of Dot Product in Neural Networks

R-VCANet: A New Deep-Learning-Based Hyperspectral Image Classification Method

Multilayer Unmixing for Hyperspectral Imagery With Fast Kernel Archetypal Analysis

Instance Normalization: The Missing Ingredient for Fast Stylization

Kernel spectral angle mapper

Hyperspectral image nonlinear unmixing and reconstruction by ELM regression ensemble

Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs)

Hyperspectral anomaly detection method based on auto-encoder

A Spatial Compositional Model for Linear Unmixing and Endmember Uncertainty Estimation

Semi-supervised Learning with Ladder Networks

Hyperspectral image unmixing using autoencoder cascade

Empirical Evaluation of Rectified Activations in Convolutional Network

Hyperspectral Image Classification With Limited Labeled Training Samples Using Enhanced Ensemble Learning and Conditional Random Fields

Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift

Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification

Adam: A Method for Stochastic Optimization

Spectral Unmixing via Data-Guided Sparsity

Zero-bias autoencoders and the benefits of co-adapting features

Structured Sparse Method for Hyperspectral Unmixing

Nonlinear Hyperspectral Unmixing With Robust Nonnegative Matrix Factorization

k-Sparse Autoencoders

A Kernel Spectral Angle Mapper algorithm for remote sensing image classification

Dropout Training as Adaptive Regularization

Piecewise Convex Multiple-Model Endmember Detection and Spectral Unmixing

Manifold Regularized Sparse NMF for Hyperspectral Unmixing

Nonlinear Unmixing of Hyperspectral Images: Models and Algorithms

ImageNet classification with deep convolutional neural networks

Representation Learning: A Review and New Perspectives

Supervised Nonlinear Spectral Unmixing Using a Postnonlinear Mixing Model for Hyperspectral Imagery

Using physics-based macroscopic and microscopic mixture models for hyperspectral pixel unmixing

Hyperspectral Unmixing Overview: Geometrical, Statistical, and Sparse Regression-Based Approaches

A Hybrid Automatic Endmember Extraction Algorithm Based on a Local Window

A New Method to Change Illumination Effect Reduction Based on Spectral Angle Constraint for Hyperspectral Image Unmixing

Contractive Auto-Encoders: Explicit Invariance During Feature Extraction

Fully Constrained Least Squares Spectral Unmixing by Simplex Projection

Hyperspectral Unmixing via $L_{1/2}$ Sparsity-Constrained Nonnegative Matrix Factorization

Bilinear models for nonlinear unmixing of hyperspectral images

Non-Linear Spectral Unmixing by Geodesic Simplex Volume Maximization

Nonlinear unmixing of hyperspectral images using a generalized bilinear model

Sparse Unmixing of Hyperspectral Data

Fast Inference in Sparse Coding Algorithms with Applications to Object Recognition

Rectified Linear Units Improve Restricted Boltzmann Machines

Nonnegative Least-Correlated Component Analysis for Separation of Dependent Sources by Volume Maximization

Stacked Denoising Autoencoders: Learning Useful Representations in a Deep Network with a Local Denoising Criterion

HYDROTHERMAL FORMATION OF CLAY-CARBONATE ALTERATION ASSEMBLAGES IN THE

On the use of small training sets for neural network-based characterization of mixed pixels in remotely sensed hyperspectral images

A comparison of kernel functions for intimate mixture models

Nonlinear Hyperspectral Mixture Analysis for tree cover estimates in orchards

Joint Bayesian Endmember Extraction and Linear Unmixing for Hyperspectral Imagery

Minimum Volume Simplex Analysis: A Fast Algorithm to Unmix Hyperspectral Data

Joint linear/nonlinear spectral unmixing of hyperspectral image data

Sparsity Promoting Iterated Constrained Endmember Detection in Hyperspectral Imagery

Approximating the Kullback Leibler Divergence Between Gaussian Mixture Models

Endmember Extraction From Highly Mixed Data Using Minimum Volume Constrained Nonnegative Matrix Factorization

Improved Manifold Coordinate Representations of Large-Scale Hyperspectral Scenes

Evaluation of Kernels for Multiclass Classification of Hyperspectral Remote Sensing Data

Composite kernels for hyperspectral image classification

Vertex component analysis: a fast algorithm to unmix hyperspectral data

Spectral linear mixing model in low spatial resolution image data

A comparison of error metrics and constraints for multiple endmember spectral mixture analysis and spectral angle mapper

HySens-DAIS/ROSIS Imaging Spectrometers at DLR

A quantitative and comparative analysis of linear and nonlinear spectral mixture models using radial basis function neural networks

Fully constrained least squares linear spectral mixture analysis method for material quantification in hyperspectral imagery

Nonlinear dimensionality reduction by locally linear embedding.

N-FINDR: an algorithm for fast autonomous spectral end-member determination in hyperspectral data

Multispectral and hyperspectral image analysis with convex cones

SPIE's International Symposium on Optical Science, Engineering, and Instrumentation報告

Derivative analysis of hyperspectral data

Geometric mixture analysis of imaging spectrometry data

Hyperspectral image classification and dimensionality reduction: an orthogonal subspace projection approach

Backpropagation Applied to Handwritten Zip Code Recognition

Bidirectional reflectance spectroscopy: 1. Theory

A Multilinear Mixing Model for Nonlinear Spectral Unmixing

Deep Learning