Unsupervised Segmentation of Hyperspectral Images Using 3D Convolutional Autoencoders

Abstract

Hyperspectral image analysis has become an important topic widely researched by the remote sensing community. Classification and segmentation of such imagery help understand the underlying materials within a scanned scene, since hyperspectral images convey a detailed information captured in a number of spectral bands. Although deep learning has established the state of the art in the field, it still remains challenging to train well-generalizing models due to the lack of ground-truth data. In this letter, we tackle this problem and propose an end-to-end approach to segment hyperspectral images in a fully unsupervised way. We introduce a new deep architecture which couples 3D convolutional autoencoders with clustering. Our multi-faceted experimental study---performed over benchmark and real-life data---revealed that our approach delivers high-quality segmentation without any prior class labels.

Submitted to IEEE Geoscience and Remote Sensing Letters

References in corpus (4)

• Going Deeper with Contextual CNN for Hyperspectral Image Classification
• BASS Net: Band-Adaptive Spectral-Spatial Feature Learning Neural Network for Hyperspectral Image Classification
• Validating Hyperspectral Image Segmentation
• Transfer Learning for Segmenting Dimensionally-Reduced Hyperspectral Images

Cited by in corpus (10)

• Remote sensing framework for geological mapping via stacked autoencoders and clustering
• A fast and Accurate Similarity-constrained Subspace Clustering Framework for Unsupervised Hyperspectral Image Classification
• Unsupervised Segmentation of Hyperspectral Remote Sensing Images with Superpixels
• Unsupervised Diffusion and Volume Maximization-Based Clustering of Hyperspectral Images
• Sparse Subspace Clustering Friendly Deep Dictionary Learning for Hyperspectral Image Classification
• A distribution-dependent Mumford-Shah model for unsupervised hyperspectral image segmentation
• Capturing implicit hierarchical structure in 3D biomedical images with self-supervised hyperbolic representations
• Hyperspectral Image Segmentation based on Graph Processing over Multilayer Networks
• Segmenting Hyperspectral Images Using Spectral-Spatial Convolutional Neural Networks With Training-Time Data Augmentation
• Hyperspectral Remote Sensing Image Classification Based on Multi-scale Cross Graphic Convolution