Convolutional Neural Networks and Data Augmentation for Spectral-Spatial Classification of Hyperspectral Images

by   Jacopo Acquarelli, et al.

Spectral-spatial classification of remotely sensed hyperspectral images has been the subject of many studies in recent years. Current methods achieve excellent performance on benchmark hyperspectral image labeling tasks when a sufficient number of labeled pixels is available. However, in the presence of only very few labeled pixels, such classification becomes a challenging problem. In this paper we propose to tackle this problem using convolutional neural networks (CNNs) and data augmentation. Our newly developed method relies on the assumption of spectral-spatial locality: nearby pixels in a hyperspectral image are related, in the sense that their spectra and their labels are likely to be similar. We exploit this assumption to develop 1) a new data augmentation procedure which adds new samples to the train set and 2) a tailored loss function which penalize differences among weights of the network corresponding to nearby wavelengths of the spectra. We train a simple single layer convolutional neural network with this loss function and augmented train set and use it to classify all unlabeled pixels of the given image. To assess the efficacy of our method, we used five publicly available hyperspectral images: Pavia Center, Pavia University, KSC, Indian Pines and Salina. On these images our method significantly outperforms other baselines. Notably, with just 1 achieves an accuracy of 99.5 improves over other baselines also in a supervised setting, when no overlap between train and test pixels is allowed. Overall our investigation demonstrates that spectral-spatial locality can be easily embedded in a simple convolutional neural network through data augmentation and a tailored loss function.


page 8

page 15

page 16


Segmenting Hyperspectral Images Using Spectral-Spatial Convolutional Neural Networks With Training-Time Data Augmentation

Hyperspectral imaging provides detailed information about the scanned ob...

Unmixing of Hyperspectral Data Using Robust Statistics-based NMF

Mixed pixels are presented in hyperspectral images due to low spatial re...

HitNet: a neural network with capsules embedded in a Hit-or-Miss layer, extended with hybrid data augmentation and ghost capsules

Neural networks designed for the task of classification have become a co...

Locality-Aware Hyperspectral Classification

Hyperspectral image classification is gaining popularity for high-precis...

SLCRF: Subspace Learning with Conditional Random Field for Hyperspectral Image Classification

Subspace learning (SL) plays an important role in hyperspectral image (H...

ColorUNet: A convolutional classification approach to colorization

This paper tackles the challenge of colorizing grayscale images. We take...

Hyperspectral Image Recovery via Hybrid Regularization

Natural images tend to mostly consist of smooth regions with individual ...

Please sign up or login with your details

Forgot password? Click here to reset